Hearted Youtube comments on Brodie Robertson (@BrodieRobertson) channel.
-
3100
-
1800
-
1600
-
1600
-
1400
-
1400
-
1100
-
1000
-
1000
-
1000
-
1000
-
1000
-
1000
-
1000
-
1000
-
1000
-
1000
-
996
-
995
-
948
-
923
-
914
-
898
-
866
-
851
-
828
-
827
-
785
-
783
-
781
-
780
-
777
-
766
-
765
-
746
-
710
-
Hello, I worked on this project quite a bit (you can see me credited at 1:54) and thought I'd let you know that while bilman66 is no longer working on the project, I still am. I built the ROM that is used in the published version of the project. You're right that it's not a "normal" distro, but it is a patched version of buildroot, a distro/linux fork made to be very configurable and to operate on minimal embedded systems.
I'm currently working on a new version of the emulator with 2 major new features; a memory management unit (MMU) and proper hard drive emulation. Right now, the emulated CPU has no MMU, which means the kernel cannot control virtual addressing or privileged levels of RAM. As a result most programs cannot run on it, as basically any code not written with no-MMU environments in mind will not work. When I finish adding the MMU, hopefully way more normal programs will be able to run inside this emulator. I am also adding an emulated hard disk, sense right now the entire filesystem is just the initramfs which the kernel will accept, but also really isn't a huge fan of and it can cause problems.
Happy to answer any other questions you might have 😊
699
-
695
-
694
-
689
-
685
-
681
-
680
-
663
-
659
-
655
-
648
-
647
-
646
-
645
-
643
-
620
-
619
-
617
-
604
-
602
-
597
-
595
-
589
-
582
-
582
-
580
-
574
-
573
-
570
-
555
-
553
-
550
-
549
-
548
-
546
-
519
-
517
-
516
-
513
-
512
-
505
-
505
-
498
-
497
-
497
-
495
-
488
-
484
-
484
-
480
-
479
-
479
-
472
-
471
-
469
-
463
-
463
-
462
-
460
-
458
-
452
-
451
-
450
-
449
-
448
-
447
-
445
-
439
-
434
-
431
-
428
-
428
-
426
-
423
-
423
-
423
-
cheat dev here, kinda want to explain why this is happening on a tiny bit more technical level, basically: windows anticheats love checking signatures. digital signatures are great, because people who have them cant do anything bad. if their signature is spotted in a cheat then you just blacklist the signature, buying a new signature from microsoft can cost around 5 digits so not the most reusable thing. On linux we don't have signatures, since if you want to recompile your entire system from scratch, lfs or gentoo style, you can do that. EAC has a massive skill issue here because it cant actually differentiate cheats from, lets say, anti viruses. thats basically it, there is nothing inherently different about linux and windows (i mean wine is a good example). and no there were still plently of checks that EAC could've used but didnt, linux just isnt that big of a threat considering the player numbers. cheating on linux is still a minority of a minority.
now if you want to hate me because i make cheats, go for it, but just remember that in the end im just giving you the message.
also how do you manage to mistake a windows system for a linux one? do you launch the game in wsl or what?
420
-
420
-
417
-
414
-
409
-
409
-
407
-
407
-
405
-
405
-
402
-
401
-
397
-
397
-
395
-
395
-
394
-
390
-
387
-
386
-
384
-
380
-
378
-
374
-
371
-
368
-
367
-
366
-
366
-
364
-
363
-
362
-
361
-
361
-
360
-
360
-
358
-
357
-
357
-
356
-
353
-
353
-
352
-
350
-
350
-
349
-
349
-
348
-
346
-
345
-
343
-
343
-
342
-
342
-
337
-
336
-
336
-
335
-
335
-
334
-
334
-
333
-
330
-
330
-
329
-
329
-
329
-
325
-
324
-
320
-
317
-
317
-
317
-
317
-
315
-
315
-
314
-
313
-
313
-
312
-
311
-
309
-
307
-
306
-
306
-
304
-
304
-
300
-
298
-
296
-
294
-
293
-
292
-
290
-
289
-
288
-
288
-
288
-
288
-
287
-
287
-
284
-
284
-
284
-
283
-
283
-
281
-
Some thoughts, off the top of my head. Here is a question for you, since I've tried to push the ball forward with the license thing for a year (particularly with getting the horible rugpull part removed from the dumpster fire that was the year-long "temporary" license)
If we referred to this license as "source first" rather than "open source", how would you feel about that? I'd like to give some history as to why this was referred to the way it was, the organization's thoughts, my personal thoughts, and the direction we'd like to move in into the future.
“Open source” has a particular meaning to the community, and suggested we call it “source available” instead. Here’s why we haven’t done that:
“Source available” commonly means you can’t redistribute modified versions, or unmodified versions, of the software. This doesn’t apply to FUTO software.
“Source available” commonly means you can’t create derivative works, or modified versions. This does not apply to our software.
“Source available” commonly means that you must pay to see the source code. This does not apply to our software.
“Source available” commonly means that software can only be used within a specific organization, but not be available outside of that organization. This does not apply to our software.
Thus, FUTO called these projects open source. FUTO didn’t care about OSI’s definition, and thought it arrogant of an organization that has confidential charter members and Microsoft as a sponsor to define it. I also believe they haven't done much to prevent the spread of closed source, subscription based, spyware from becoming the norm. People use google apps over nextcloud, and adobe premiere over kdenlive for a reason. google calendar works, and adobe premiere allows me to use a compressor plugin that actually allows me to see a visual representation of gain reduction. While open source software runs the server world & the internet, just using the linux kernel & apache without a doubt; in the world of niche professional software, or consumer software... I think it's lost very badly.
“Source available” is commonly understood to encompass projects with far more restrictive terms than our software.
“Open source” is commonly understood to have no financial limitations on one’s ability to use the software commercially.
Neither one of the community’s definitions fully fits what we’re doing. Using either of them is bad. The former is a disservice to us, and the latter is a disservice to you. A fair compromise, I think, would be making our own term.
“Source first” would describe our software, and fit our values;
Here’s where source first & our values align with the community’s definition of open source:
1. Our licenses allow users to see source code of all of our software.
2. Our licenses ensure that you can modify the source code for your own use, and redistribute it.
3. Our licenses ensure that our software is not limited to use by a particular organization.
4. Our principles demand that any client we release that requires a server, also releases the server software under principles as free as the client software.
5. Our software avoids integration of crypto shitcoin scams.
6. Our software rejects “the customer is the product” as a business model.
Here’s where source first & our values part ways with the community’s definition of open source:
1. We believe in a programmer’s ability to have the legal right to demand financial compensation for commercial use of their code. It's not enough for a programmer to have the ask for money politely; we want them to have the legal right to demand commercial entities pay. ffmpeg getting a few programmers for google summer of code, while the backbone of youtube is key to them bringing in 31 billion dollars a year isn't a fair exchange of value. Under all current open source licenses, they can ask for money, but they can't demand money.
2. We believe that community ownership of software has not led to consumer-facing-software that beats closed source alternatives, and that this has not, and will not, be a winning model.
I think the community likes what we’re doing. They like that we fund consumer-facing software projects like Immich that seek to beat google photos & iCloud. They like that we’ve donated millions to non-profits like Signal, Tor, Repair Preservation Group, GrapheneOS, MicroG, Mobility Independence Foundaiton, and many other organizations that use traditional, OSI-compliant license, that fit the community's definition of open source.
They don’t like that we’ve used the term open source. Technically, legally, nobody “owns” that term. OSI doesn’t own that term. So we used it; but culturally, the community own that term. Frankly, as I've told my boss personally many times, you can be technically correct, and still be a complete dickhead. You're not going against OSI when you do this; you're going against the community.
I don't have it in me to give you all some adobe crisis management blog post bullshit, so I'll put my thoughts here:
I've had an idea of what open source meant since I first bought & installed Suse Linux professional in 2002. I had no idea what OSI was, or how they defined software, or what they even did until i started working here. While working here, I've made my stance unequivocally clear; I don't think it makes sense to have all of this good will created by having the fellowship programs we have, donating millions to open source that ARE MIT/Apache/AGPL/GPL licensed, funding projects internally that kick ass & use AGPL licenses for some; all to lose it over what you call your licenses. Being technically correct has, and always will, in the public eye, be less important than being culturally correct & taking action that does not cause the community to see us all as dickheads.
I'd like to advocate internally for that to change, without compromising our values, or the community's.
This is all an experiment; at the end, it comes down to this. If we create software that doesn’t abuse the public; will people pay for it? Will programmers quit working for google, apple, facebook, amazon, and microsoft; and decide to work on GOOD software instead? Because they believe they can actually make a living from it?
We don't see people doing that right now for consumer software. For many pieces, there isn't even an option to not be abused by traditional closed souce/cloud bullshit, and that sucks. Someone has to go first and give this a shot, to even have a chance of success, prior to others giving it a go. If we fund that software to the tune of millions of dollars per year, in order to make it great under the terms we've created; will people support it financially? If we can demonstrate that they WILL, can we get more programmers to change their habits? That is FUTO's goal.
It'll include stumbles along the way. I believe this was one of them, that can be addressed, improved, & fixed along the way.
Thoughts, as always, appreciated.
280
-
279
-
278
-
278
-
278
-
276
-
276
-
276
-
276
-
276
-
275
-
275
-
274
-
274
-
273
-
273
-
272
-
272
-
270
-
269
-
269
-
268
-
Rick Yorgason here. A big piece of feedback on the original protocol was "we can't accept this when we haven't even explored the possibility of a better alternative", so Matthias is doing his best at exploring alternatives.
I'm just doing a bit of light tire kicking on his proposals. I legitimately think his latest proposal is a better fit for non-traditional compositors (tiling, scrolling, VR, etc...). It's moderately harder for traditional compositors, toolkits, and Wine, but it's flexible enough that they should be able to do something reasonable. And the loudest voices in the Wayland community have this weird double standard where supporting unusual compositors is of the utmost importance, but unusual apps are "not our problem", so this might have more success at getting accepted.
But if the end result is just saying "okay, we explored the alternatives, and they're more complicated than they're worth, so we should just go along with the industry standard", then that's a win too.
268
-
266
-
I said it in a previous video, but Mozilla needs to focus on doing a small hundful of things and doing it well instead of half-heartidly chasing trends, failing at them, and then scrapping the project, wasting everyone's time, effort, and money. Meanwhile, the products their user base does care about slips a little more in quality, leading to people jumping ship and Firefox loses more shares. At the rate they're going, Firefox is going to end up in the same place as their OG name, NetScape, and being buired and forgotten.
266
-
265
-
265
-
263
-
263
-
261
-
261
-
261
-
260
-
260
-
257
-
256
-
256
-
256
-
254
-
252
-
248
-
247
-
247
-
245
-
244
-
244
-
242
-
241
-
240
-
239
-
239
-
239
-
238
-
238
-
238
-
237
-
236
-
235
-
234
-
234
-
233
-
232
-
232
-
231
-
231
-
230
-
229
-
228
-
228
-
227
-
226
-
225
-
225
-
225
-
225
-
225
-
224
-
224
-
224
-
223
-
222
-
222
-
222
-
221
-
221
-
221
-
220
-
220
-
219
-
219
-
217
-
215
-
215
-
215
-
213
-
212
-
212
-
211
-
210
-
209
-
209
-
209
-
208
-
208
-
208
-
207
-
206
-
206
-
205
-
205
-
205
-
199
-
199
-
198
-
197
-
196
-
196
-
196
-
196
-
195
-
195
-
195
-
194
-
194
-
194
-
194
-
194
-
192
-
191
-
190
-
190
-
190
-
190
-
189
-
188
-
188
-
188
-
188
-
188
-
188
-
188
-
188
-
188
-
187
-
187
-
187
-
186
-
185
-
185
-
185
-
184
-
184
-
183
-
183
-
182
-
182
-
182
-
182
-
182
-
182
-
182
-
182
-
181
-
181
-
181
-
181
-
180
-
Hi! I'm the maintainer of Osiris and MiDE (KDE1 fork)! Wow, what a surprise to see a video from my Reddit post! Sorry I didn't include more links, I didn't expect there to be as much interest as there was! My work on MiDE is indeed based on Helio Chissini de Castro's (heliocastro) initial work, from around 2016. My Qt2 fork, Osiris, was done independently. There are many people that deserve thanks/credit for the project being possible!
The name MiDE was chosen, from the Japanese verb "miru" - to see/to look. As simple and plain as it is, KDE1 is easy to look at, but unimposing enough to not distract the user from seeing what they are doing.
Osiris 2.4.2.1 is releasing here in the next couple of weeks as a security release, and with it, will come packages for a few distributions (planned: Debian, Ubuntu, and Fedora). It is considered generally stable at this point. Hopefully, I'll have a test release of MiDE out soon :)
180
-
179
-
179
-
179
-
179
-
Ex gentoo user now on arch, and wishing that I never changed. I changed because of the time commitment, but I find that anything I want to do something a bit off the beaten path, it is a lot more painful. One example is that I recently wanted to use some academic code to play with some bleeding edge 3D graphics research. The library required a particular implementation of a maths function. Because this implementation was faster in some cases, but not in all, it is not in the default library. I had to manage this myself, or change the code. What was in the AUR didn't work. In gentoo, whenever I had problems like that, it was either change a USE flag and do an update to fix, or just go and install the new library myself, then do an update. Because everything was compiled from source on the system, I generally had all the dependencies, instead of having to run around looking for the right devel packages.
178
-
178
-
177
-
177
-
177
-
176
-
175
-
175
-
174
-
174
-
174
-
173
-
173
-
173
-
173
-
173
-
172
-
172
-
172
-
171
-
170
-
169
-
168
-
168
-
168
-
167
-
167
-
166
-
166
-
166
-
165
-
165
-
165
-
165
-
164
-
164
-
164
-
164
-
164
-
163
-
163
-
162
-
162
-
162
-
161
-
161
-
161
-
160
-
160
-
159
-
159
-
158
-
157
-
157
-
156
-
156
-
156
-
155
-
155
-
154
-
154
-
154
-
154
-
154
-
154
-
153
-
153
-
153
-
152
-
151
-
150
-
149
-
149
-
149
-
149
-
148
-
148
-
148
-
148
-
148
-
148
-
147
-
147
-
147
-
147
-
146
-
146
-
145
-
145
-
144
-
143
-
143
-
143
-
143
-
143
-
143
-
143
-
143
-
142
-
142
-
142
-
141
-
141
-
this is actually an amazing project, now I know I might sounds joking, but I'm not, this is actually the best feature there is on windows, why? because it lets you know that you haven't activated your os, it reminds you, we as a human forget things easily all the time, but this is a free reminder from windows (thanks a lot windows), now who in the world doesn't wanted to activated windows? a scummy pirate? please, windows is the best os of all times, it lets you change your desktop wallpaper, and uh.. you could change your admin password, it even lets you... do all things that you wouldn't be able to do on linux or even macos, for example it lets you activate your own os, what's that? macos and linux is free? well you know what they say, if it's free, you're the product, so lets thanks windows for letting us pay for that amazing os
141
-
141
-
Speaking as a longtime Plasma user:
The application dashboard is an alternative launcher that's meant to go on panels, and not the desktop itself. It's basically a macOS homage, if the title didn't give it away.
I don't know anyone who considers the Super+T anything close to "tiling", including myself, someone with little if any experience with tiling WMs. Like you said, it's far closer to being an expanded version of Windows 10 and 11's FancyZones powertoy than an actual alternative to tiling WMs.
Your enthusiam for widgets has resparked a bit of joy in me, knowing how the legacy of widgets lives on in KDE Plasma. The Plasma team could've changed their minds like Windows and macOS did by around, say Plasma 5, but they didn't, and all the kudos to them for that. That said, I don't get why widgets are used for both the desktop and panels -- that's definitely a problem, and extremely likely why you got confused by the aforementioned dashboard.
Around the 15:00 mark you bring up the pop-up panel menu, and that only shows up when you right-click the panel itself and not the desktop. Failing that, there's a cog in the far right of the panel to bring it up manually while in Edit Mode.
Anyways, dunno why I'm bringing all this up other than to provide fuel for the algorithm. I don't even work at KDE, let alone contribute, I just also think the stuff they do is neat lol
141
-
141
-
140
-
140
-
140
-
140
-
140
-
139
-
139
-
138
-
138
-
138
-
138
-
138
-
138
-
138
-
137
-
137
-
137
-
137
-
137
-
136
-
136
-
136
-
136
-
135
-
135
-
134
-
134
-
134
-
134
-
134
-
134
-
134
-
134
-
134
-
133
-
133
-
133
-
132
-
132
-
132
-
132
-
132
-
132
-
131
-
131
-
131
-
131
-
130
-
130
-
130
-
130
-
130
-
130
-
129
-
129
-
128
-
128
-
128
-
127
-
127
-
127
-
127
-
127
-
126
-
126
-
126
-
125
-
125
-
125
-
125
-
124
-
124
-
123
-
123
-
123
-
123
-
122
-
121
-
121
-
121
-
Not only do I remember the lawsuit, I followed it as it happened. For one, as I recall, their ridiculous claims of copyright violation were largely for standard C header files that exist in every C development system anywhere.
A few minor nits, though. For one, Novell was the primary corporate network operating system before the internet and the migration to TCP/IP networks and the rise of Windows servers. In the late 90's my workgroup was at a disconnected site and I got the collateral duty of computer support. We were running a Novell 3.2 (I think) network at the time and migrated to Netware 4 while I had it. One of the nice things about them buying SuSE linux from its original German company, was that improved networking support (I was already a SuSE user at the time, still am). Also, you're pronouncing the name wrong. It's not pronounced like a book novel. It's "No-veil".
I also remember there was a woman, a lawyer I believe, that setup a website for the sole purpose of tracking and documenting the lawsuit, every single step of the way. And did a really good job of explaining the "legalese" for everyone. I couldn't remember her name, but it turns out I still have an old bookmark buried that links to where the site was. Site still there, but the article I originally linked to isn't. Anyway, it appears the woman was Pamela Jones and the site was/is Groklaw.net. I just checked, you can still go back as far as 2003 and find articles about the SCO-IBM suit. Might be worth a visit for anyone interested in more information.
121
-
121
-
120
-
120
-
120
-
120
-
120
-
119
-
119
-
119
-
119
-
119
-
119
-
118
-
118
-
118
-
117
-
117
-
117
-
116
-
116
-
116
-
116
-
Dev guy here: Regarding Luke's "removal" issue with Steam:
- On Mint at least, APT locks itself down so that only one instance (by default, as there are alternatives and modded versions people use) can run in the system at the same time.
- Upon that lock, no other application can hook onto APT while it's in use.
What likely happened is Software Manager woes where you may have accidentally clicked twice on an option and it tried to bring up APT to both install and remove it. Obviously, installation comes first and locks down APT. Removal has to wait for the lock to lift. Since it couldn't "grab" APT within a reasonable amount of time, Software Manager just hung there. In that case, closing it and running the installed app is fine.
I very much dislike Mint's software manager because it still has a lot of issues that remain unresolved and it's not the safest thing to use. When using any Debian-based distro I tend to prefer terminal, which is NOT good for UX especially for newbies.
I am more partial to the way Pamac does things (high five to Brodie for using the same distro as me :D ), but that application also has issues of its own whereas Octopi provides more surgical control of the system. Basically when I'm lazy, Pamac. When I feel like I need to be more careful with the system during a particular procedure (like excising 495 nvidia drivers that I accidentally installed in Pamac and reverting to 470 with the lib32s), Octopi.
In a way, Mint has this too with Software Manager / Synaptic. Lazy install? Software Manager. Need more attention to detail on what you're doing? Synaptic.
Still, I think Pop! has a better manager for new users even with the problems Linus had. Apps like octopi/synaptic can be very intimidating to use if you're not "invested" in Linux culture/know the proper package names and how to sanitize your app suite & repos.
116
-
116
-
115
-
115
-
115
-
115
-
114
-
114
-
113
-
113
-
113
-
112
-
112
-
112
-
112
-
112
-
Wendell and Steve are awesome people. OpenPleb is an amazing initiative and it's absolutely necessary with how many proprietary one-off systems are out there and how every piece of hardware you buy (not even just counting PC hardware) requires some form of software to control. Of course I'm in it for the RGB, but there are so many other things that this initiative will help. libratbag for mouse control is a big one, along with any project for keyboard remapping such as OpenRazer and ckb-next. Sensors are another area that was talked about, especially with super IO chips where some chips have documentation and others do not. Super IO also encompasses motherboard fan control. Smart power supplies. LCD displays on things like cases and coolers. Heck, the initiative doesn't have to be limited to just PC (though in its current implementation it does seem like consumer PC is the focus) because there are a ton of smart home devices using proprietary protocols and apps that could also benefit.
I'm glad this is getting media attention. I wanted to standardize RGB as soon as two vendors had RGB keyboards on the market and worked with their official SDKs for several years before so much RGB was coming out that SDKs were no longer guaranteed. Then I dove into the reverse engineering side because I wanted my entire PC to sync up and work in Linux. Reverse engineering even the 5 or so things I had in my PC at the time was like a year long effort, but once we figured those out and had some documentation and tools for reverse engineering the underlying interfaces it started to get easier. Still, reverse engineering is never truly easy and it will always be time-consuming work.
It looks like OpenPleb is already making an impact. Last night, HYTE tweeted out:
At HYTE, the software user experience is our key value proposition. We appreciate and support the direction of Open Pleb. We're eager to learn more about how we can contribute and be part of this! @GamersNexus @Level1Techs @CalcProgrammer1
They've offered to send me some sample products along with DOCUMENTATION! This is big. At least some companies are listening and that's incredible after years of working on OpenRGB with pretty much no acknowledgement by any of the vendors (except HyperX early on, good on them, though they still only sent me documentation in private rather than release it publicly).
112
-
Oooooh, memories! I'm the original poster of that bug. Thanks a lot for the walkthrough! I think the lack of this, over the years, pushed me away from using workspaces alltogether... I'm still on KDE, though. Even though I'm not too concerned anymore, I get reminded of the issue every now and then, because I can't unsubscribe.
At some point, my setup was two screens on an Intel NUC and a laptop with synergy and later barrier between them. I think I stopped around that time. Nowadays, I have monitors with very different resolution, one 4K, and I'm more concerned how things would behave on really large monitors. At some point I suppose they will come down in price, and say you have a 7680x2160, like for example Samsung Odyssey Neo G9 🤤 No way I could afford that now, but some day. I would suppose you would section the display for different windows, so "maximize" wouldn't imply "cover the whole screen", but more "fill this section of the screen". And then, it would be interesting to assign workplaces again in interesting ways. I don't know if anything is happening in that regard.
112
-
111
-
111
-
111
-
111
-
111
-
110
-
110
-
110
-
110
-
110
-
110
-
110
-
109
-
109
-
109
-
109
-
109
-
109
-
109
-
108
-
108
-
108
-
108
-
108
-
107
-
107
-
107
-
107
-
107
-
107
-
107
-
107
-
It really shows that the GPU division of AMD has a lot less money and manpower in it, compared to their CPU division. People always praise AMD for great linux support, but the docs are lacking, especially when it comes to ML related stuff, from what I've heard. If not going with NVIDIA, I have to say, Intel Arc really looks like a very compelling option for GPUs on Linux. Sure, their drivers are still buggy, features are missing and so on. But overall, the docs seem very organized, the drivers are improving at massive speed, and Intel has a well tought out roadmap for their GPU architecture in the next years. Meanwhile AMD just seems to try random stuff, kinda half assed. Sometimes, it really looks like they don't know what they are doing. I might get a second gen Intel Arc GPU, when drivers are more mature. Because their hardware stack is actually very good, the docs are good, and basically everything is open src
107
-
106
-
106
-
106
-
106
-
106
-
106
-
106
-
106
-
106
-
106
-
106
-
105
-
105
-
104
-
104
-
104
-
104
-
104
-
104
-
104
-
104
-
103
-
103
-
103
-
The historical reason for chat and email apps etc. to have systray icons is because you want them running but you don't want them to show giant chunky taskbar entries all the time, crowding out everything else. So they close to the systray and you use the systray to keep track of them instead of your taskbar.
...Except if you use an icons-only/dock-style taskbar, in which case those apps' icons takes up very little extra space while open--or even none, for apps that are pinned in place. Taskbar icons have the ability to show various statuses already--unread count, progress indicator, etc. So when using this kind of taskbar, you genuinely don't need tray icons for those apps and they're just redundant with the taskbar icons.
What I think we need is for some kind of API that lets the shell tell apps whether a tray icon makes sense in the current context. So when using an icons-only taskbar, apps implementing this API would ask the shell, the shell would say "no" and then when the app has a visible taskbar icon, they wouldn't show a systray icon or silently stay running when closed. They'd just behave like normal apps, without the user having to manually disable the tray icon.
However, none of this applies to apps that never show an icon on the taskbar, like the Nextcloud client thingy, or Dropbox, or similar background-only apps. For those, it does make sense for them to show a tray icon to make the user aware of their background status. Asking them as an alternative to opt into some DE-specific API is a nonstarter for them since they're cross-platform and simply will not care about us enough to do the large amount of work to write a Plasma applet or adapt to whatever API the GNOME folks want, or whatever. We can't have our heads in the sand so much that we think we can dictate their design. I'd love it if they did write lovely custom UIs for us, but it just won't happen--not until our combined market share becomes 10x bigger or more.
103
-
103
-
103
-
103
-
103
-
103
-
103
-
103
-
102
-
102
-
102
-
102
-
101
-
101
-
101
-
101
-
101
-
100
-
100
-
100
-
100
-
100
-
100
-
100
-
100
-
99
-
99
-
99
-
99
-
99
-
99
-
99
-
99
-
98
-
98
-
98
-
97
-
97
-
97
-
97
-
97
-
97
-
97
-
97
-
97
-
97
-
97
-
96
-
96
-
96
-
96
-
96
-
96
-
96
-
96
-
96
-
95
-
95
-
95
-
95
-
95
-
95
-
94
-
94
-
94
-
94
-
94
-
94
-
94
-
94
-
94
-
94
-
94
-
94
-
93
-
93
-
93
-
93
-
93
-
93
-
93
-
93
-
93
-
92
-
92
-
92
-
92
-
92
-
92
-
92
-
92
-
92
-
92
-
92
-
92
-
92
-
91
-
91
-
91
-
91
-
91
-
91
-
91
-
91
-
91
-
91
-
91
-
91
-
90
-
90
-
90
-
90
-
90
-
89
-
89
-
89
-
89
-
89
-
89
-
89
-
89
-
88
-
88
-
88
-
88
-
88
-
88
-
88
-
88
-
87
-
87
-
87
-
87
-
87
-
87
-
87
-
87
-
87
-
87
-
87
-
87
-
86
-
86
-
86
-
86
-
86
-
85
-
85
-
85
-
85
-
85
-
85
-
85
-
85
-
85
-
85
-
85
-
85
-
84
-
84
-
84
-
84
-
84
-
84
-
84
-
84
-
83
-
83
-
83
-
83
-
83
-
83
-
82
-
82
-
82
-
82
-
82
-
82
-
82
-
81
-
81
-
81
-
81
-
81
-
81
-
81
-
80
-
80
-
80
-
80
-
80
-
80
-
80
-
80
-
80
-
80
-
80
-
80
-
79
-
79
-
79
-
79
-
79
-
79
-
79
-
79
-
79
-
I think that Apple gets less hate because they're a more "opt-in" ecosystem / playground. That is, the default is windows, when you have no choice or don't know what to pick. So you'll use it and, in many cases, find things that irk you and some that you'll absolutely hate. But going to an Apple... you usually reasearch it a bit, before you choose to buy one. That is, you already have some idea if you'd like it or not and there's good chances that you'll simply not switch to it if there's possibility of incompatibility, so to speak. Getting back to Windows being the default option - you rarely are forced to use Apple, for, say, work.
So bottom line of the above, when going Apple you usually know what you're getting into, significantly reducing the number of people frustrating with using it. As some simply choose to not go Apple, they might've realized beforehand that what they're doing is simply incompatible (like most gaming). And the rest might've had done some research and learned how to do the basic things.
Me personally, I do hate Apple more than Microsoft. I do not deny that their engineers and designers are usually doing a very good job. Most people I know using Apple's products are happy, things work. Well, the thing they're using it for. But Apple is so focused on control and walling the garden as much as possible, so anti consumer, that I do not care how good their products are.
Microsoft, to be fair, is not that far off. But, I guess, because their current position, they have a much bigger garden, so closing it is much much harder. But their strides for requiring an online Microsoft account, and what they're doing to secure login and I forgot the next thing after secure login, that's also a no-no. I've used Windows since Windows 95 (used a bit 3.11 too, but it was on old computers in some places) to Windows 10, and I've been a happy Windows 10 user. I know I won't drive Windows 11, not by personal choice. I might have to, for work, but unless I REALLY have to for something specific, I won't install it on any of my personal systems. Even if their bullshit if bypassable.
79
-
78
-
78
-
78
-
78
-
78
-
78
-
78
-
78
-
78
-
78
-
78
-
78
-
77
-
77
-
77
-
77
-
77
-
77
-
77
-
77
-
77
-
77
-
77
-
77
-
77
-
77
-
77
-
76
-
76
-
76
-
76
-
76
-
I just wrote a lengthy reply to someone asking why there's hesitation to adopt absolute positioning, but it looks like the original post got deleted. So if anyone's interested, here's what I wrote:
I like the original proposal; that's how I found it in the first place, but the people who don't like the original proposal have a few different reasons.
The best complaint I heard was from someone who is building a VR compositor, and yeah, XY coordinates really don't make sense in a 3D space, so I get it. I think the latest proposal works better here, because you can easily place a virtual screen in 3D space.
It's a similar situation with tiling compositors. Smart compositors could theoretically do something like this already to keep app windows clumped together, but it would be guess work that sometimes guesses wrong, and wouldn't work for multiple apps coordinating together. I feel like these developers are holding out hope that Qt, SDL, and all the other toolkits drop support for absolute positioning if Wayland doesn't support it, thereby forcing app authors to stop making these "distasteful" apps that work poorly with tiled managers, but I think that's unrealistic, especially now that JavaScript is getting support for absolute positioning.
There's some people who are worried about security risks, where a malicious app does something like placing a borderless window over a credit card input field to capture your input. It's a stretch given you would have already had to give the malicious app permission to run, and to inspect your screen, but I'm glad someone's thinking about it.
Then there's people who are just worried that if an app can place a window wherever it wants, then apps will start doing that all the time, and the desktop experience will be garbage, but I never bought that argument, because apps have been able to do that for about 40 years and generally haven't.
76
-
76
-
76
-
76
-
76
-
76
-
76
-
76
-
76
-
75
-
75
-
75
-
75
-
75
-
75
-
75
-
75
-
74
-
74
-
74
-
74
-
👋Hello there! I'm the creator of tuifi. First of all, I want to thank you for your review, I do really apreciate it, you gave me some interesting insights and ideas for the next version. Second of all, I'm sorry for your deletion-losses you had 😅 (ngl it was pretty dumb of me, not to say anything about this at all)
0:52 That's mainly because It uses the uni-curses library which is (ncurses/pdcurses) pure C in it's backend
1:38 in TERMUX it works with touch like with a normal smartphone file manager which makes it perfect for navigation inside it
1:56 I had it removed because there was an issue with the decoading of non latin letters, but now it works fine
2:55 It is possible I just didn't implemet it yet
3:53 You can Drag&Drop things into folders, there is just no visual animation to it yet
4:04 True, I'll add that to the todo list, hehe.
4:26 Oooh noooo, sorry for that 😅
5:13 That's definately something I'll add, thanks!
7:54 Hahahahaha, I was board ngl
Besides the use in TERMUX, it's also a great tool for when you want to make changes, to a remote machine (like raspberry pi) over ssh, easily and effectively, with the familiarity of a GUI.
PS. This is my main file manager at the momment lol
74
-
74
-
74
-
74
-
74
-
74
-
73
-
73
-
73
-
73
-
73
-
73
-
73
-
73
-
73
-
72
-
72
-
72
-
72
-
72
-
72
-
72
-
72
-
72
-
72
-
72
-
72
-
72
-
72
-
72
-
72
-
72
-
71
-
71
-
71
-
71
-
71
-
71
-
Jobs and Gates don't get their recognition for their programming/technical abilities, rather for being great businessmen with novel ideas and leading groundbreaking innovations (while not necessarily actually creating them). I mean it's enough to compare Jobs and Woźniak to see this and tbh. Linus will get more recognition once desktop Linux becomes more popular but even then, he's a lead engineer, not a UI designer or visionary (well, he might be in the programming world but that's not enough). Like you said, Jobs and Gates got computers in hands of average people, this is something that's already present in history books so it simply can't compare. I imagine Linus will get more recognition once the FOSS movement takes off more among normal people or hopefully even becomes enforced by some governments one day, he'll be remembered for his contributions to consumer freedom in the world of computing rather than his technical contributions, and in fact, if that moment ever comes, it's Stallman who should and I think will get more recognition than anyone else, whether we agree with him and the FSF or like everything they do, he's the man we can thank for starting this entire movement and he'll one day take his place in the history books, possibly as an equal to Jobs and Gates.
71
-
70
-
70
-
70
-
70
-
70
-
69
-
69
-
69
-
69
-
69
-
69
-
69
-
69
-
69
-
69
-
69
-
69
-
69
-
69
-
69
-
68
-
68
-
68
-
68
-
68
-
68
-
68
-
68
-
68
-
68
-
68
-
68
-
67
-
67
-
67
-
67
-
67
-
67
-
67
-
67
-
67
-
67
-
67
-
67
-
67
-
I had an RX6900XT card. It was a terrific pick for doing general tasks and gaming. Worked perfectly out of the box with pretty much any distro. It was an Asrock that overclocked insanely, ran very cool, total silicon-lottery winner. Games were a joy with it.
I really loved it until I wanted to use it to accelerate Blender.
Getting ROCm installed was **horrific**. There is simply no clear set of instructions, even if you're using one of the supported distro/version combinations. After much hair-pulling, I got it working... but installing the various components broke other things. Like some apps not displaying correctly in X. Like apt breaking due to version and naming issues from the AMD repos. I've been using Linux for decades at this point, and this situation gave me PTSD of how Linux often used to be when getting hardware to work.
In the end, it was easier to reverse everything and replace the card with a used EVGA RTX3090. I really hate Nvidia in so many ways. But, while the replacement uses proprietary drivers, everything Just Works after those are installed.
Am I thrilled with the RTX3090? Well, not entirely, as it works worse for the original use case - general tasks and gaming. But not so much worse that I'd even think about switching back. My next card is likely to be from Nvidia, although I'll wait for the 4000 series to hit the used market in quantity.
I used to tell everyone that AMD was best for Linux. I now answer in a much more nuanced way. AMD should be ashamed.
67
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
66
-
65
-
65
-
65
-
65
-
65
-
65
-
65
-
65
-
65
-
65
-
65
-
65
-
65
-
65
-
64
-
64
-
64
-
64
-
64
-
64
-
64
-
64
-
64
-
64
-
64
-
63
-
63
-
63
-
63
-
63
-
63
-
63
-
63
-
63
-
62
-
62
-
62
-
62
-
62
-
62
-
62
-
62
-
62
-
62
-
62
-
62
-
62
-
62
-
62
-
61
-
61
-
61
-
61
-
61
-
61
-
61
-
61
-
61
-
61
-
61
-
61
-
61
-
61
-
60
-
60
-
60
-
60
-
60
-
60
-
60
-
60
-
60
-
60
-
60
-
60
-
60
-
60
-
60
-
60
-
60
-
60
-
60
-
60
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
59
-
58
-
58
-
58
-
58
-
58
-
58
-
58
-
58
-
58
-
58
-
58
-
58
-
58
-
58
-
58
-
58
-
57
-
57
-
57
-
57
-
57
-
57
-
57
-
57
-
57
-
57
-
57
-
57
-
57
-
57
-
57
-
57
-
57
-
56
-
56
-
56
-
56
-
56
-
56
-
56
-
56
-
56
-
56
-
55
-
55
-
55
-
55
-
55
-
55
-
55
-
55
-
55
-
55
-
55
-
55
-
55
-
55
-
From my understanding, one thing you pointed out is part of the core of most Red Hat hate: it's basically impossible to have a Linux system that isn't full of "Red Hat projects". This next sentence is gonna seem like a bit of a jump, but I promise it's going somewhere, so bear with me. For better or for worse, systemd standardized a lot of core system components and brought them under one source tree. Some people either dislike having that much of their system under one group's control or just dislike the way systemd does things. Those people then notice that systemd is backed by a corporation, and that makes things worse since a lot of Linux users distrust corporations in general. Then they notice that the corporation behind systemd also has some level of involvement with various other projects that can't easily be avoided (including dbus, which I've noticed some hardcore systemd antis have a particular hatred for), and the paranoia neurons really start firing. This paranoia leads people to have a strong hatred for systemd, the corporation behind it (Red Hat), and anything else they see Red Hat touch. Since most Red Hat hate (at least from what I've seen) seems to be an extension of systemd hate, I'm gonna second the unixsheikh articles someone else linked to get more of a feel for what people hate about systemd.
I can't say for sure why Xorg seems to escape the Red Hat/Freedesktop hate, but I suspect it's probably some combination of people not noticing the connection and people pretending not to notice because going TTY-only doesn't sound very fun.
55
-
55
-
55
-
55
-
55
-
55
-
54
-
54
-
54
-
54
-
54
-
54
-
54
-
54
-
54
-
54
-
54
-
54
-
54
-
54
-
53
-
53
-
53
-
53
-
53
-
53
-
53
-
53
-
53
-
53
-
53
-
53
-
53
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
52
-
51
-
51
-
51
-
51
-
51
-
51
-
51
-
51
-
51
-
51
-
51
-
51
-
51
-
51
-
51
-
51
-
51
-
51
-
51
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
50
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
49
-
Being "the best" in some theoretical technical sense has always been nothing compared to having a product out there that actually solves people's problems. Case in point: the IBM PC/x8i6 platform. It was never the most elegant, the fastest, nor the most sophisticated computer out there, but it was good enough, it had widespread support, and it was easy for clone for much cheaper than any name-brand 16/32-bit computer. People do not want "the best computer", people want the computer that meets their needs, and the clone makers did it better than Commodore, Apple, Acorn, and even IBM themselves. Minix and Hurd might in some technical sense be "more advanced" but Linux actually showed up with a working system and showed up first, while Hurd has been five years away from prime-time since I was in diapers, and Minix has never even seriously tried to be a viable general-purpose OS. If there was ever an OS that threatened to eclipse Linux as "the" FOSS operating system, it was 386/FreeBSD, which was even more archaic, with a history going straight back into AT&T Unix in the '70s! And it might have indeed been better than Linux since it was much more tightly integrated, but SCO's ratfucking ensured it was not there at the right time to solve people's problem (how do we free the PC clone platform from dependency on Microsoft), and Linux was. Hence Linux is everywhere and BSD is an also-ran.
Ninety percent of life, as they say, is just showing up. Linux showed up, and the pie-in-the-sky dreams of hyper-advanced operating systems from Plan 9 to OS/2 to Minix to Hurd to BeOS--didn't. Hell, you could say that in many ways the Apple ARM platform is technically superior to the x86 PC, but since it's enslaved to Apple's strategy of selling computers as a luxury brand, it will never threaten the PC's dominance. Luxury computers don't solve people's problems, workhorse commodity computers do, even if they're slower and less energy-efficient.
See also: Motorola 680x0. It was in many ways better than the x86 (at least until the Pentium II turned the x86 into a workstation-grade RISC beast that could be had for consumer-grade prices), but it didn't matter because the computers the x86 shipped in solved more problems.
49
-
49
-
48
-
48
-
48
-
48
-
48
-
48
-
48
-
48
-
48
-
48
-
48
-
48
-
48
-
48
-
48
-
48
-
To be honest, it's not his fault or NPMs fault. It's JS fault. If it wasn't for type coercion, unsafe numbers and undefined, null, prototype, Object and Array implementation, lacking std, etc; people would've just move on. I don't see that kind of problem in Python, Go, Java, C or whatever, because these languages either have all you need implemented already, or it's easier and more reasonable to do it yourself. When I write go, I write go. If I need something and it's not in std, I write my own "std"/"utils"/"tools"/"helpers" or copy-paste it from SO. The only packages I have are huge things like zap, pgx, disgord; or something like custom errors wrapper and UUID. The go.mod file lists all the packages that are used (even indirectly) and it's usually under 20, may be 50-ish for bigger projects, but that's it. If I open up my JS app be it svelte/vue or whatever, I'll see hundreds of them... Because JavraSkip
48
-
48
-
48
-
48
-
48
-
48
-
48
-
48
-
48
-
47
-
47
-
47
-
47
-
47
-
47
-
47
-
47
-
47
-
47
-
47
-
47
-
47
-
47
-
47
-
47
-
47
-
47
-
47
-
47
-
47
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
46
-
Here in China, this kind of replacement has been taken placed for nearly a year, and mostly only on government's computers. Our linux user group has been flooded by the government employees since last year, and many of them have never heard linux before.
There are six kinds of CPU involved, like Phytium(aarch64) and Loongson(loongarch), and only one of them is x86. These computers often runs UOS, which is the 'professional' version of Deepin (just like the relationship between Red Hat Enterprise and Fedora). The UOS is definitely not for tinkers because it is not free in charge and normally you don't have root access, which requires you to apply from the distribution maintainer.
As for the paperwork, the current setup is enough, most of the user only need a browser, an OA system(usually chromium based), a wordprocessor(WPS), which are already available on these CPUs. We also teach them how to configure wine and box86 so they can run their favorite games like Red Alert 2.
Though I don't use these operating systems, it indeed helped me as a desktop linux user. Many Chinese apps now have a usable linux port, which is a rare thing just two years ago.
46
-
46
-
46
-
46
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
45
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
44
-
43
-
43
-
43
-
43
-
43
-
43
-
43
-
43
-
43
-
43
-
43
-
43
-
43
-
43
-
43
-
43
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
42
-
41
-
41
-
41
-
41
-
41
-
41
-
41
-
41
-
41
-
41
-
41
-
41
-
41
-
41
-
41
-
41
-
41
-
41
-
41
-
41
-
41
-
41
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
40
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
39
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
38
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
37
-
36
-
@BrodieRobertson agreed, but that is another problem that is specific to a steaming pileof manure that is JavaScript. If JS had actual, strict and meaningful standard, that all browsers adhere to, there would not be such issue.
You can take C code and compile it against different operating systems, heck even computer architectures (x86, x64, ARM, list goes on) and it works.
The fact that each browser has its own idea of what JS is stems from the fact that JS is poorly defined, badly maintained hackjob, that is a cancer on web development for over 20 years.
It's still better now than it was back in IE6 times. Writing a code to handle click event? Well i need to check what was the source of the event, so event.source right? "Yes" says everyone. "No, it's event.src_element." says Microsoft.
The world as a whole could be made significantly better place for everyone on the planet by making talking about, developing or running JS code illegal.
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
Can confirm Michael was a voice of reason in that conversation. I appreciate him proposing GNOME 48 for the removal of that code, hopefully that will be good enough for GNOME users and the usecases raised that need to be addressed. While it would have preferred to have been avoided, in the case of Buddies of Budgie, we have gone ahead and opted to fork gnome-session into budgie-session, so we'll have to spend a fair bit of time cleaning that up, preventing packaging and naming conflicts, and making it suitable for Budgie 10 (on top of GNOME's stack e.g. GNOME Settings Daemon, GTK, and Mutter). This is objectively going to impact our pace in moving to Wayland. We're a small team and instead of spending time on porting various parts of the codebase over to supporting Wayland, we'll need to spend time on this, only to have to rip it all out for 11 since none of it will be applicable, we'll be using our own daemon, different toolkit, non-Mutter, and so forth. But it is what it is 🤷
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
36
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
35
-
34
-
34
-
34
-
I fail to see how this a security issue at all.
Scenario A) the value is set by a user... the user could just set the value to "retry every 3 seconds" and get this same effect. The only way this could be an "issue" is if there is validation on the retry number that only sets a lower bound: retry must be more than X seconds, and now you set it to billions of seconds and you can get a value less than X secondss. Any place I can think of where I can set the retry value, I already have full control of the retry value. The value being something different than I expected is a bug, not a security issue.
Scenario B) the value is set by an application using curl as a library ... the application could just set the value to something else, and if youre running software you don't trust, you have bigger issues than "this happens at an interval I wasn't expecting".
So the only place I can think of this happening is if you A) take user input, B) don't sanitize user input, C) the request fails and warrants a retry, and D) the service you're hitting is a non-public service that someone couldn't just DDOS directly anyway (in which case, why are you taking user input to do this).
I also can't think of a single scenario where there is user defined retry periods to hit a nonpublic service. And if its app defined values, that's poor coding, not "an exploit". Can anyone explain to me even a hypothetical scenario where this would be set up in a way that might cause this to be "exploited" ?
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
34
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
33
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
32
-
Long time Blender user here. I agree with most of the things you had to say, but here's my ted talk addressing a few of the criticisms you had.
The layout is not static. If you do both 3d modeling and video editing, you can still save your startup file with the video editing workspace and create a new one from the topbar that has your 3d modelling setup. Yes, Blender stores layouts in the .blend files, but when opening one, you can disable "Load UI" to keep your layout, if that's what you prefer.
I agree, the way you have to split/join areas by clicking and dragging feels pretty weird and they only made it more difficult by default in the most recent versions. However you can easily right click on the border between 2 areas and chose wether you want to split, swap or join them. There are also faster shortcuts.
As you mentioned in the pinned, audio scrubbing does exist under the Playback popover in the timeline editor. Audio waveforms by default are set per-strip, but you can change that under Sequencer editor > View > Show Waveforms. Also if you want to modify a property for multiple selected strips/anything in blender, you can hold alt while changing it, instead of having to right click > Copy to Selection.
The reason you were getting inconsistent results with the snapping is because it depends on your cursor position. If your mouse is closer to the start of a strip, it wi snap it to the end and if your mouse is closer to the end, it will snap to the start. I guess that makes more sense when modelling, but it transfers over to the sequencer.
Finally, there is a built-in add-on for Blender called "Power Sequencer" made by GDQuest that makes video editing in Blender way easier and faster, especially for content creators. I'd suggest checking it out to anyone interested.
That being said, I agree that Blender's video editing capabilities are nowhere near comparable to other free video editing programs, let alone the proprietary software that is overused nowadays. Overall, great video, it's just that I'm a bit of an enthusiast when it comes to Blender
32
-
32
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
31
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
30
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
29
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
28
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
27
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
26
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
I'd just like to interject for a moment. What you're refering to as Linux, is in fact, GNU/Linux, or as I've recently taken to calling it, GNU plus Linux. Linux is not an operating system unto itself, but rather another free component of a fully functioning GNU system made useful by the GNU corelibs, shell utilities and vital system components comprising a full OS as defined by POSIX.
Many computer users run a modified version of the GNU system every day, without realizing it. Through a peculiar turn of events, the version of GNU which is widely used today is often called Linux, and many of its users are not aware that it is basically the GNU system, developed by the GNU Project.
There really is a Linux, and these people are using it, but it is just a part of the system they use. Linux is the kernel: the program in the system that allocates the machine's resources to the other programs that you run. The kernel is an essential part of an operating system, but useless by itself; it can only function in the context of a complete operating system. Linux is normally used in combination with the GNU operating system: the whole system is basically GNU with Linux added, or GNU/Linux. All the so-called Linux distributions are really distributions of GNU/Linux!
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
25
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
Among the numerous advantages of free/libre software such as the Linux kernel is the freedom to reduce the core kernel to a minimum and then add through the module architecture those drivers for components, both hardware and software, which you want or need. For instance, if your computer is a 64-bit Intel i5, you can remove everything supporting other microprocessor architectures and devices of no interest. Everyone else can migrate to Go or Rust or C++ while you merrily continue using C and your subset of the kernel and never concern yourself about ingratiating yourself with "but C is too difficult to learn" crowd. You never have to pull from the official Linux kernel repository again after the one-time clone before you begin the purge.
All that said, C was developed during a time when computing resources were limited by today's standards. The language was designed to replace assembly language for all except the lowest level within the kernel proper - yes, C was created specifically for systems-level programming of operating systems. In contrast, Go and Rust are much higher-level languages but whether they are better systems programming languages remains to be proven in time.
Brodie's comparison of COBOL and C ignores the changing technology landscape from mainframes to minicomputers and microcomputers which led to entirely new types of software, both systems level and application level. Mainframe dominance faded but many mainframes remain in production systems in the twenty-first century. The decline in schools teaching COBOL and the failure of employers to hire programmers into in-house training programs to learn COBOL to support their systems are two reasons for the dearth in available programmers with COBOL knowledge and skills. On the other hand, the continued watering down and chasing trendy languages in computer science degrees offered by colleges and universities has led to a rapid decline in the number of computer science students exposed to C beyond perhaps an introductory operating systems course.
In my opinion the Linux kernel architecture has grown too unwieldy such that maintaining it is overly complicated for the maintainers. Maybe it is time for Linus Torvalds to redesign the kernel, with C as the implementation language, as a new project with the objective of simplifying maintenance. Let the current kernel architecture continue until the new kernel and systems architecture reach v0.99.
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
To give a little bit more context as a Windows system developer:
Transactional NTFS was, as Brodie mentioned, introduced in Vista. It essentially moves the transactions paradigm commonly found in databases over to the file system (and technically also the Windows Registry, but we will ignore that here). Essentially, transactions work by performing multiple operations to modify the data and ensuring that either all of these modifications end up being successful (commit) or none (discard). Microsoft primarily developed it for their own Windows Update mechanism. So instead of an update ending up being "half-applied", an update would either fail, and no changes are made to the system, or it would succeed, and all of the changes are made correctly.
Transactional NTFS was deprecated: It turns out that adding transactions to NTFS caused an explosion regarding code complexity within the NTFS file system driver. The code within that driver tripled if my memory serves me well. Microsoft also found that while a transactional filesystem sounded good on paper, there were many limitations with their implementation specifically. So even Microsoft abandoned the Transactional NTFS usage for their Windows Updates shortly after, leaving it in its deprecated state.
Why would any game developer use it, then? I can only speculate. In principle, it would make sense for a game developer to ensure that the game doesn't corrupt save files. So ensuring that a save file can never be in a half-written state (because the game started writing to the save file, then encountered some error, and aborted the save procedure before finishing the write, for example) is generally a good idea. However, the same could have been ensured by creating a backup copy of the existing save file first before trying to write to it and restoring the backup in case anything goes wrong.
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
24
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
23
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
I mean, I kinda agree that Linux isn't exactly safe and secure, precisely because it's a monolithic kernel. This works fine for many things, but for mission critical circumstances I completely understand why one would prefer a microkernel, and the culture problems directly interact with the fact that it's a monolithic kernel because it means that someone can blow up the system for everyone else. That's said, I'm sure for aerospace specific needs you could tailor build software specifically around a microkernel like Hurd, Minix, or Redox, because it's not exactly software that the vast majority of people will be using or working on anyways. The problem with Linux being monolithic is it's just far too late to switch away for most general use applications of the software - no one is going to port all the drivers for all the hardware we need now to some other microkernel architecture, besides maybe System 76 who've been working on Redox, and Google with Zircon, but they'll only be making drivers for the things they need to work. Honestly, it's unlikely that the culture problem existing with a monolithic kernel will blow up in your face if you're running a stable release cycle anyways.
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
22
-
This is a side effect of having the ability to specify an arbitrary transformation matrix. The ability to specify arbitrary transformation matrices is actually EXTREMELY handy and has a number of legitimate use cases.
For example, if you are using a projector that was installed into a wall at the wrong angle, and the built-in compensation settings aren't quite getting it right (or it's too high to reach and someone lost the remote to it 5 years ago), you can do some math (or use a GUI if one exists) and have xrandr correct it for you so that the projected image is perfectly square on the wall.
Also, while I don't understand the 'why' so much for this, I've heard that some people building arcade cabinets with Linux boxes have a literal mirror or two reflect the monitor, and when you play the game you're playing it with the image literally mirrored. Arbitrary transformation matrices allow you to do such mirroring.
And of course, by far the most common use of this is to fix overscan issues, since some TVs (particularly older ones) will cut off the edges of what's displayed, so you have to scale the image down so that it all fits correctly. Especially if it's not entirely symmetrical and you have to both scale the image down and move it around a bit.
Arbitrary transformation matrices allow you to implement ALL of these features, all wrapped up into one feature. The big main desktop environments will HAVE to implement it, if they want to account for these use cases (at least internally; they might not allow you to edit the transformation matrix directly like xrandr does, but IMO they _should_).
22
-
22
-
22
-
22
-
22
-
22
-
I use Orca with the Mate desktop, as even the Gnome desktop, which is supposed to be one of the best for accessibility, has a ton of issues. If Orca were to stop working, Linux would be very much unusable. Linux is fast, simple (as in it does exactly what you tell it to), and to be forced to use Windows or Mac OS just because the accessibility isn't used by that many people would just be... Insert whatever explatives you want here. Suffice it to say, very very bad. Wayland in general should have all this fixed before it is made the default on any distros that are public facing, i.e. ubuntu and other such distros. Experimental distros are just that, experimental and forward thinking, so it makes sense that not all of these things will work. But for everything else, accessibility should be a top priority on everyone's mind, even if they don't use it. If accessibility is good, then basically everything else will also work just as expected, such as automation like you mentioned
22
-
22
-
22
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
Once upon a time, Debian put main contrib non-free into your sources.list (once it actually used sources.list by default—my use of Debian predate apt's existence), but changed that because rms said that he couldn't recommend Debian if it didn't do so. They changed that, he thanked us, and he said that he now felt comfortable recommending Debian to people.
The FSF website did not recommend ANY Linux distribution at the time, but rms did specifically mention a few times after that Debian was one of the things he could encourage you to install. Obviously Debian is no longer free enough. Y'know despite the FSF publishing a license that is no longer DFSG-free.
Oh, we weren't going to discuss that? Yeah … the FSF has relicensed the manuals of most of its core software under the GNU FDL which is not DFSG-free in application due to invariant sections. So both have a stick too far up their own arse to recognize each other as free. Except of course that Debian's standard is clearly stated, and the FSF's is subject to the cult of personality of a single individual.
I still believe Debian made the right call in its recent vote. Not because non-free software is great and wonderful (it isn't), but because if you can't use your computer without it, what's the point? Are we seriously to believe that your computer should be borderline unusable if you made the unfortunate decision to buy the most popular graphics card chipset brand in the world? According to Stallman, YES. You should have thought about that before you were so unclean as to buy the wrong hardware.
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
21
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
Feel like bringing up more examples of FOSS games, alongside everyone else who already has:
- Sonic Robo Blast 2 (3d platformer and Sonic fangame)
- Dr. Robotnik's Ring Racers (kart racer -- based on the above)
- Taisei (Touhou fangame -- would be surprised if famed weeb Brodie doesn't know of that one)
- KBounce (literally just JezzBall from the Windows 3.x era)
- Smokin' Guns (spaghetti western FPS -- think LucasArts' Outlaws or Red Dead -- made with the Q3A engine)
- Neverball (literally just Super Monkey Ball -- also made with the Q3A engine)
- Xonotic (multiplayer deathmatch FPS, made... with a modified DarkPlaces, IIRC?)
- Iliyas Jorio's ports of Pangea Software's games (Bugdom 1 and 2, Nanosaur 1 and 2, Billy Frontier, Otto Matic, and Cro-Mag Rally -- all originally Mac exclusive, and updated to run on modern OSes, including Linux distros)
- Frogatto & Friends (2D platformer)
And of course, that's not even going into all the engine ports/decompilations/reimplementations of (cult) classic games like Commander Keen, Duke Nukem 1/2/3D, Captain Claw, Tyrian, Descent, Marathon, Tomb Raider, etc.
(Certain names starting with M and Z not mentioned due to them being made by a certain company who loves litigation.)
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
Didn't notice the pointer constraint protocol, which is also very important for gaming, especially if you have a second monitor. And also, I don't beleave that "the date, when wayland will be the only option available" will happen soon, if ever. X11 still has it's usecases, which wayland is way far to cover (Like, for example, using a gui application via docker). Don't get me wrong, I'm using a selfmade wayland compositor (spent whole nigth to update it on wlroots 17, because of surprise update on arch) and I'm not an X11 prophet.
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
20
-
This:
I'd just like to interject for a moment. What you're referring to as Linux, is in fact, GNU/Linux, or as I've recently taken to calling it, GNU plus Linux. Linux is not an operating system unto itself, but rather another free component of a fully functioning GNU system made useful by the GNU corelibs, shell utilities and vital system components comprising a full OS as defined by POSIX. Many computer users run a modified version of the GNU system every day, without realizing it. Through a peculiar turn of events, the version of GNU which is widely used today is often called "Linux", and many of its users are not aware that it is basically the GNU system, developed by the GNU Project. There really is a Linux, and these people are using it, but it is just a part of the system they use. Linux is the kernel: the program in the system that allocates the machine's resources to the other programs that you run. The kernel is an essential part of an operating system, but useless by itself; it can only function in the context of a complete operating system. Linux is normally used in combination with the GNU operating system: the whole system is basically GNU with Linux added, or GNU/Linux. All the so-called "Linux" .
To chinese:
我刚才对一个时刻感到担忧。 你们提及的是,事实上,中非支助处/民解组织,或我最近发出的呼吁,即中非支助处和防污。 146. (9) 并不是一种操作系统本身,而是由中非支助处核心libs、炮击水电和重要系统组成部分为联索特派团系统全面运作的一个自由组成部分,由联索行动IX界定。 许多计算机用户每天都在对联索支助系统进行修改,但没有实现。 通过一次活动,今天广泛使用的中非支助处的版本经常被称之为“防污”,许多用户都不知道,这是由中非支助处项目开发的中非支助办事处系统。 确实有一处有防备,这些人民正在利用它,但只是他们使用的系统的一部分。 146. 防污是星球:将机器资源分配给你管理的其它方案的制度。 停车场是操作系统的重要组成部分,但本身不用;它只能在完整的作业制度中发挥作用。 146. 通常与中非支助处业务系统结合使用:整个系统基本上是联非支助处,有的是,有的是,有的是,有的是,有的是,有的是联索部队/民阵。 所有所谓的 146。
Back to english:
I have just been concerned at a moment. What you have referred to is, in fact, the Central African Support Office/FNL or the recent appeal that I have made, namely, the Central African Support Service and the Prevention of Pollution. (9) Not a system of operation itself, but is defined by UNPOS at its core,libs, artillery power and vital system components as a free component of the full operation of the UNMIN system. Many computer users are changing the UNSOS daily, but have not been achieved. Through an event, the versions of the BONUCA widely used today are often referred to as “recovery”, and many users do not know that this is the BONUCA system developed by ONUCA projects. Indeed, there is a need for preparedness, which are being used by these peoples, but only part of their systems. The fire is a celest: the allocation of machine resources to other programmes administered by you. The parking area is an important part of the operating system, but it is not itself useful; it can only play a role in the full operating system. It is often used in conjunction with the BONUCA business system: the entire system is largely the UNBIS, and it is the case that there are UNPOS/FN. All so-called 146.
hmm
20
-
20
-
20
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
19
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
There are two very serious issues that are very concerning if this model ever gets popular. You mentioned an increase in global power consumption, but did not explain its practical consequences:
- First and most importantly, it would have a huge environmental cost.
- Second, virtually no country's electrical grid would be able to handle the excess power consumption as of today. In the short/medium run, it would lead to frequent powerloss (many countries today are already struggling to meet the demand in electricity). In the long run, it would increase the cost of living for everyone because of personal electricity bills, but also taxes for the electricity infrastructures.
Several countries have started banning crypto-mining because it takes an unprecedented toll on their power infrastructures, eventhough a small minority of people mine. Imagine if everyone was constantly mining, even if it's less than the cryptobros.
Also, if you pay with your electricity bill, the website is no longer free to use. Your electricity provider just became the middleman that pays the websites you visit.
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
18
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
The thing is, Fedora shouldn't be a new-to-linux non-techy first experience, they say they are at the edge (leading, bleeding... who cares) for a reason. As even brodie said, Ubuntu LTS will ship X until 2029 (probably), other distro's as well, some will ship Wayland earlier but still only if it works 100% of the time every time...
But I agree with you, for the user what matters is the result. We used to call it Duck Theory - if it looks like a duck and looks like it is swimming like a duck, that is fine.. even if just below the water is actually a mechanical C'Thulu tangled up in Titanic with a tentacle pushing a duck-'mask'/'glove' along the water...
But for the maintainers, the sooner Wayland is the default, the fewer dragons underwater they have to wrestle with.
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
17
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
16
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
00:11 🗓 GIMP 3.0.0's release date has been continuously pushed; latest plan aims for May 2024.
01:09 📝 GIMP 3.x roadmap progress includes completed tasks like porting, but major areas like script plugin redesign are still ongoing.
02:04 🎨 Moving to gtk3 brings advantages, but complex projects like GIMP face additional bugs needing resolution before release.
03:00 📄 Documentation updates are crucial before release, contingent on completing other planned features.
04:25 🚀 Revised schedule targets a final development version next month, aiming for a release candidate a few months later, pegging May 2024 for GIMP 3.
07:18 🛠 Delay expected for the pre-release version due to pending tasks, doubting the completion of all roadmap items by December.
08:00 📅 String freeze planned by mid-January, aiming for API freeze by the end of February, leading to release candidates.
08:55 🚧 Early user access anticipated in February-March, acknowledging potential bugs once real users engage with GIMP 3.
09:49 ⏰ Firm target set for May 2024 release, aligning with the Libra Graphics meeting, though doubts persist due to the remaining tasks.
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
As the original user who added the "Waiting on CEF upstream" label to the mentioned issue at 5:43, I can confirm CEF issue 2804 is the correct issue we're (notably "I am") tracking upstream.
There are two parts to this.
The first is that CEF, like OBS, is open source with few core maintainers - notably Marshall. Additionally, OBS is not "important" in the scope of CEF. Therefore, one of us commenting "Hey this bug affects OBS" onto a CEF issue will not move the ball in the slightest - which is why you didn't see such a comment. We will occasionally poke where we feel appropriate or contribute our knowledge where we can, but you often won't see us fly-by comment with "hey fix it sooner rather than later, it affects us too." We'd rather leave it to people who know what they're talking about.
The second is the complexities surrounding this bug and CEF 2804 in particular.
To answer the first part, 3:39, browser sources use CEF's "off-screen rendering" which means they're unaffected by what windowing system you choose to use. They're rendered off-screen, and then the texture is loaded into the canvas.
The bug in question affects displaying an actual CEF window within a parent window that isn't owned by CEF (ie. browser docks and service integration login screens). You'll find a comment from Marshall directly after Tytan's where he covers two cases - Views and Ozone. Ozone is an abstraction layer beneath the Aura windowing system/GUI framework, used by Chromium. Views are a way to expose Aura more directly to apps like OBS, and required to use Ozone on Wayland - all of which is supported. However, large chunks of Ozone are still X11 only, including the ability to embed CEF windows in non-CEF windows, such as Qt. As the ticket is still open on the CEF side, there's nothing extra we can do.
Of course, even if a patch for CEF was written tomorrow, OBS currently can't update CEF until a working update to Shared Textures for CEF is released, which nobody is actively working on (the previous patch was written by an Nvidia developer, and Marshall mentioned he won't accept a PR to CEF as the change needs to be done upstream, within Chromium, before the relevant bits are done in CEF).
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
Just to mention a couple of things (I was brought here cos we've had a couple of users who evidently read your clickbait title and didn't watch the video)
TL;DW for people reading the title: No we did not drop Wayland, it's just set to prefer xwayland over wayland by default for the Appimage and Flatpack, you CAN force it if you really want to, as described in the video.
Regarding the Nvidia problems: Yes they're still a thing, one of our testers has Wayland and the whole compositor just seems to just crap itself on Nvidia, in our emulator the covers layout and whatnot just turn in to a mangled mess, it looks pretty bad.
I understand that in some cases there are different protocols we could implement to fix some of the issues, but this is part of the problem, we have to mess around doing special code just for Wayland because it dares to be different? We have universal code for Windows, Mac and X11, why should we have to mess that up and add special code paths just for Wayland? It's kind of ridiculous to have to do this.
I do agree that more developers should be screaming at Wayland folks rather than complaining on their own repos, but then you see threads like the one Stenzek linked where people are doing that, and the answer is pretty much "we don't care, your old way that everybody else does is dumb, use our new way!" and other fart sniffing responses.
Anyways, thanks for the video, even if we did have people who evidently didn't watch it.
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
15
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
In Ye Olde Dayse, you had to be pretty picky about choosing hardware for Linux. Video cards often could be made to work, and you might need to give $30 to 4front for a binary OSS sound driver. WiFi initially wasn't a thing, and then when it was there was one or two devices you knew you could get working with Linux, but only that.
Today people just assume hardware works under Linux, and that's kind of a mistake. It's why you get people like Linus Sebastian assuming it's Linux's fault his windows-only proprietary bullshit sound hardware doesn't work under Linux.
But things have changed in the decades since… A lot more hardware today is just a microcontroller with no onboard firmware, and the driver's first job is to install firmware. Nobody cared about that in 1997 because the firmware was already on the device. Now people care about it because the usually non-free firmware's got to be distributed with a free software driver. The question is, if the first step of the free software driver is "load this non-free blob and then talk to it", is the driver really free? It used to be since the blob was baked into the hardware before. So that should be good enough, right?
Except … your computer has a 486-class processor sitting on the motherboard that runs its own proprietary operating system you have no means to control or fix. That operating system runs a SECRET WEB SERVER that can be exploited. And the best we've managed to do in some properly free firmwares is quarantine that bullshit. We can't really disable it, and since it is responsible for applying security patches to our CPUs, maybe we don't want to completely disable it. But when you're got to contend with that, maybe non-free firmware is just not good enough. It's the current state of things, but I'm thinking that it shouldn't be, and we should be arguing for it for on security grounds if not on software freedom grounds.
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
As the creator of dmenufm and fm.awk, I am not really sure how should I react to this video. On one hand, I do appreciate Brodie reviewing my project and might potentially allow me to find more contributors for this project. On the other hand, I do feel being offended by your conclusion: weird projects. I believe there is no real standard for what a file manager should look like, and the reason I make this is exactly because I want to use it. I write it in a way that fits my preference, and at least I do use it daily (otherwise I won't make this in advance).
Since you've already noticed that the content of this video is quite outdated, I won't blame you saying some functions are missing. However, I will explain some details I saw in this video:
1. Awk cannot capture keyboard input. Those are captured by dd.
2. The reason why fm.awk cannot reload/refresh by itself whenever the terminal window is resized is because there is no trap function available. That is to say, as long as fm.awk is launched, there is no way for fm.awk to capture outside signals (like WINCH) and react accordingly. I wish dd or stty can trap signals, but I haven't found it now. If anyone knows how to use stty or dd to trap signals, feel free to make an issue or pull request.
3. Enter is a bug. I just implemented it. Thank you.
4. Page file manager is something just easier to implement in awk. It is simply easier to separate all entries into pages using awk array. That's why I implement in this way.
5. After implementing file preview function, I realize the problem you've described in https://youtu.be/P5nYwOcT1a4?t=324 and implement the number-typing as what you described.
6. For those file manipulation, I realize that command line is still the most efficient way to do it, but choosing files sucks on command line. So I wish fm.awk can fit into the middle ground between these two, but still be minimalist. That's why I've implemented the cmdmode and may see this function in this screencast:
https://asciinema.org/a/9YDmY7GhnV7ku2yRhGJlQa8l4
14
-
14
-
14
-
14
-
14
-
14
-
14
-
14
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
I started off on a 386, never had a 80387 and got my first taste of an FPU with a 486-DX2. I am really surprised that 15+ years after the last x86 was built, that their is a distro still supporting it. I can only imagine there being ancient dedicated servers that run some old software that hasn't been updated in over a decade that still requires a 32 bit OS, and IÄll wager that most of them sit in a corner somewhere with a "DO NOT TOUCH, EVER!" label on them. I think that it is fair to stop supporting ancient architectures after 15+ years, it must be hard to even keep systems that old online for tests (yes, I know, VMs and all that, but they are never 100%).
Also the i386 was the first 32 bit CPU with protected mode (286 was 16 bit), a game changer that really "enabled" Linus to make Linux with it's kernel controlled preemptive multitasking. So I don't think it will even be missed TBH, and those very few that need it, are more than capable, or should be, of compiling their own 32 kernels and packages from source.
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
12
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
Something that might be of interest Brodie, with kwin it has the ability to tile windows to sides and corners natively. To get a window to tile in a corner with the keyboard instead of dragging it there with the mouse, use shortcut keys to say... tile to the right side, then immediately use the shortcut to tile to the top (or bottom) and the window will corner tile. The downside of course is you can't tile more than 4 windows on a workspace, it only does side and corner tiling (and there's no gaps), but it works without having to use the new "tiler", which requires using the mouse. I never found a need to have more than 3 windows open on a workspace (and even that's rare), so it works fine for me.
You'll want to rebind the keys for sending the windows to the sides and moving between them (the defaults are, umm... odd), but it works.
Something that I think will drive you crazy is multiple monitors. Kwin doesn't do independent workspaces per monitor. If you switch from say workspace 1 to workspace 2 on the left monitor, the right one also switches. They're tied together. Pinning a needed app to all workspaces on one monitor works as an option, but it's a poor substitute for independent workspaces.
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
I used a lot of distributions, and I say they don't matter because all of them suck, but you should use Arch and Gentoo for a while at least once, for the experience, and to know that you're not too much of a scrub to do it. I used Arch for many years, but I use too many programs, so a few things would break every now and then and it was annoying. The AUR isn't much of a selling point for me, because the more unusual things in there are unmaintained and broken anyway. Personally, Gentoo was more stable for me than Arch, but nowadays I just use a minimal installation of Debian because I already have the knowledge that comes from using those distros, and now I'm lazy and don't give a shit.
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
I was involved with one of the other projects at the Shuttleworth Foundation at the time that the Freedom Toaster was developed. I know the guys who built it, including Jason who is mentioned in the video at about 4:00. Talking of which, the Freedom Toaster in that video is actually the first prototype (and I used it) - it was only later that the kiosk style was rolled out. While a great idea, the Freedom Toaster never really got off the ground, due to a couple of factors.
One of the biggest issues was keeping the ISOs up to date. It meant that someone had to copy the ISOs across, and then update the menu interface with the new versions, etc. Since they needed to be updated manually, most of the deployed machines rarely got updated. Also, most of these machines were deployed at universities, so that they were able to connect to the Internet, but they still had that manual update process. Also, at the time, most people in South Africa did not own a computer, and so there were very few people who even knew about Windows, never mind Linux.
11
-
Author here, this is a pretty fair review about the state of
things. You are spot-on with most of your criticisms. That said, the
key bindings don't work because the key configuration is in ini format
and requires the section header above the actual bindings. This
probably should be communicated better (there should be a message in
the log, though..) and yeah, the documentation is currently a bit
lacking and needs some rework. The folding was originally for
backtraces and debug info in the log viewer, but also works for the
sections in the quick actions menu (when you press 'a'). I personally
use the colemak keyboard layout and different bindings, that's where
those issues come from..
Gstreamer is a real PITA, both in terms of programming, as well as
making it work on users machines. I will probably switch to ffmpeg
eventually. There's another Arch user who reports it not working, but
I'm not sure why or which package is required. Maybe the maintainer of
the AUR package knows. I use Gentoo and have gst-plugins-{good, bad,
ugly} installed. I'm pretty sure at least the bad plugins are
required. That said, Gstreamer is only used for VIDEO (and audio)
previews. You can still get plain old image previews by compiling with
"--no-default-features --features img", which will enable image
previews only. (Add "sixel" to features if you want that).
If you'd like to help making gsteamer work or any of the stuff that
hasn't ben fixed in the mean-time (current master has a few fixes
regarding some things you mention, delete your config or start with
"hunter -u" for full effects) feel free to open an issue about it on
Github, or maybe drop by on the irc channel.
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
11
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
00:00 🖥 X11 operates in rootless or rootful mode; rootless manages windows, while rootful displays the entire X11 desktop.
01:24 🛠 Introduction of Xwayland run utilities aids in running X11 and Wayland clients efficiently, replacing xvfb for headless X11.
02:05 🖼 XVFB (X virtual frame buffer) emulates a frame buffer using virtual memory, primarily used for server testing and various X11-related tasks.
03:01 🖥 Xwayland run emerges as an alternative to XVFB, specialized for headless X11 operations with added features like setting virtual screen sizes.
04:23 🎮 Xwhand run allows running X11 clients within their own rootful X11 window, aiding in managing applications like games with fixed resolutions.
05:44 🎮 Gamescope offers an alternative to rootful mode for gaming-related issues, solving many problems effectively.
06:13 🛠 Three utilities emerged: Xwhand run for X11 clients, WL headless run for Wayland clients (supporting various compositors), and Xwfb run as a hybrid of the two, replacing XVFB.
07:22 💻 Discussion arises on whether these utilities should be included in the Wayland namespace or kept separate due to compositor-specific code.
08:45 🧩 Consideration is given to integrating these utilities into existing Wayland utilities or X server repos, discussing concerns about compositor-specific scripts.
10:25 🤔 There's debate about these tools' placement: whether they should reside in Wayland utils or within each compositor project for better integration.
11:19 🌐 Community involvement might determine the shift of these tools into a more communal namespace; it's suggested that active contribution could drive their integration.
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
Yes OBS is probably one of the worst nightmare applications you could throw at a sandbox-type environment like appimage, because it wants to capture whole system audio / video / inputs including from other applications, across multiple windowing and audio APIs, etc. It doesn't want to just stay in its one box / window, its whole purpose is to interact and "spy on" the entire system. And then because you're trying to be an appimage that supports multiple distros, you have to make these features work for not just a single Ubuntu release for example, but all across different distributions with possible different versions of software, graphics APIs, available audio, GPU drivers, kernels, etc.
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
It was the 90s, nobody was really sure how versions were supposed to work yet. Semantic versioning, like we do now, wasn't common. Everyone did versions differently back then. Emacs did their versions like this 16.60, 17.36, 17.43, 17.46,17.48...17.63, 17.64, 18.41, for example.
People worked that stuff out in the late 90s, early 2000s.
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
I Brodie, my name is Márcio, I'm 36 years old and I live in Brazil, I am the user who created the forum post Lutris with the nick of m2a, thanks for your video, but I was really banned for using lutris as I've been doing since 2018, playing BF3, BF4, BF1 and V until EA ban, a lot of people are speculating that it's really developing some cheat, which was not, I was just banned on BFV, playing on Island, not just me, as other users Linux, unfortunately EA will not assume this error, as Blizzard did with Overwatch, as I said in the forum after EA's replies, I do not buy anything from them, so angry already deleted the games I had installed. Anyway, what pisses me off, that the real cheats are out there ruining the game of many, and these are not banned properly. Thanks
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
Thanks for the interesting video. I use Arco Linux (Arch) with Gnome as my daily driver. I see that tiling WM are very popular these days, but they should not be so useful for me, as I use all my apps in full screen (maximised). I also much appreciate to use Gnome as an extension called "Maximised by default" opens all my apps directly in full screen that avoids some floating WM issues. I only use half screen (set with a shortcut) for the file manager and sometimes other apps. If I need quarter of screen, I also have shortcuts, but the window comes very small and not so useful. I mainly stick with Gnome as it has a fantastic built in blue light filter and a Gnome extension "Night Light Slider" allows me to very quickly and easily adapt the temperature of the screen to the current room luminosity. Another extension called "Change PC brightness" allows me to scroll on an icon in the top bar in order to quickly and precisely adjust the screen luminosity. I used to have all my apps in dark mode, but I now much prefer to be able to quickly adapt temperature and then the luminosity. I am not sure if I could set all this up as easily in a tiling WM, and, as I found a setup that fits perfectly my workflow, I don't see why I would take time to test something else. With Arco, I can quickly install 16 DE or WM, but Gnome works so good for me, so I am happy to stick with it. I also use a lot my keyboard as I open and switch to all my apps with keyboard shortcuts, thanks to "jumpapp". The last extension I wish to share is "Unite", it gives my window much more space on the screen, as the buttons and title are moved to the top bar. What is fantastic in the Linux world is that we have such a large choice that everybody can find what he needs. I then fully respect and understand the fact that many users love to use tiling WM for all its advantage, it's just not the case for me for now.
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
10
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
Brodie, thanks for another informative review. I agree that nnn is excellent, thanks Arun!
Three strong points for me:
- nnn is so fast to move around in (with hjkl) that I rarely feel constrained by the bookmarks being limited to 10 choices - I just make those choices good starting points
- I use the contexts about every second time that I use nnn - excellent for moving stuff from one location to another, while checking what's in a third, and so on
- a killer feature for me is rename (r) which opens the current location as a text file in your editor, which in my case is nvim, and which then allows me to do insanely fast bulk renames - I've never before seen anything as efficient as this
So yeah, I'm sticking with nnn: relatively easy to configure, intuitive, once you get Arun's thinking, and fast fast fast.
I did look at lf, after watching your two videos on how it's a good competitor for ranger, and looked into it for two use cases, listed here (and how I STILL decided to stick to nnn):
- working with a tree of image files - solved for me with vimiv
- looking through a tree of configuration files - tmux with nnn on the left and with nvim using NerdTree on the right
So nnn wins again...
Thanks Arun, you've created a utility that's revolutionised my GNU/Linux life.
And thanks Brodie. Keep it up!
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
I would like to point out that the linked Ariadne article states "partially mitigated through a microcode update."
"Mitigated" already means - less likely but still possible, and I can only interpret "partially" as "even less effective than normal mitigation".
So I don't think it is fair of them to say that a CPU without this microcode is "broken" in another paragraph of the same article.
There is no way to really fix specter and meltdown without physically building a new CPU, because these vulnerabilities are at an architectural level.
They can be made harder to exploit through software both at kernel level and microcode level, though at least kernel patches have significant performance price.
So, the real choice users face here is: use the processor with a known issue, or load a proprietary microcode blob that might help slightly, but might also introduce new problems at the same time.
I would say that decision is not as cut and dry as the linked article makes it seem.
By the way, if reverse engineering CPU microcode was achievable with reasonable effort, we would have free replacement for those blobs right now, and this whole thing would be a non-issue.
There has been tons of reverse engineering done to support devices on Linux which the manufacturer never supplied drivers or even specs for.
So while I am not an expert, I think it is safe to say the reason we do not have any free microcode alternatives out there, is because creating one is very seriously hindered either technically, legally, or most likely both.
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
When I log into my Xenix system with my 110 baud teletype, both vi and Emacs are just too damn slow. They print useless messages like, 'C-h for help' and '"foo" File is read only'. So I use the editor that doesn't waste my VALUABLE time.
Ed, man! !man ed
ED(1) UNIX Programmer's Manual ED(1)
NAME
ed - text editor
SYNOPSIS
ed [ - ] [ -x ] [ name ]
DESCRIPTION
Ed is the standard text editor.
Computer Scientists love ed, not just because it comes first alphabetically, but because it's the standard. Everyone else loves ed because it's ED!
"Ed is the standard text editor."
And ed doesn't waste space on my Timex Sinclair. Just look:
-rwxr-xr-x 1 root 24 Oct 29 1929 /bin/ed
-rwxr-xr-t 4 root 1310720 Jan 1 1970 /usr/ucb/vi
-rwxr-xr-x 1 root 5.89824e37 Oct 22 1990 /usr/bin/emacs
Of course, on the system I administrate, vi is symlinked to ed. Emacs has been replaced by a shell script which 1) Generates a syslog message at level LOG_EMERG; 2) reduces the user's disk quota by 100K; and 3) RUNS ED!!!!!!
"Ed is the standard text editor."
Let's look at a typical novice's session with the mighty ed:
golem> ed
?
help
?
?
?
quit
?
exit
?
bye
?
hello?
?
eat flaming death
?
^C
?
^C
?
^D
?
Note the consistent user interface and error reportage. Ed is generous enough to flag errors, yet prudent enough not to overwhelm the novice with verbosity.
"Ed is the standard text editor."
Ed, the greatest WYGIWYG editor of all.
ED IS THE TRUE PATH TO NIRVANA! ED HAS BEEN THE CHOICE OF EDUCATED AND IGNORANT ALIKE FOR CENTURIES! ED WILL NOT CORRUPT YOUR PRECIOUS BODILY FLUIDS!! ED IS THE STANDARD TEXT EDITOR! ED MAKES THE SUN SHINE AND THE BIRDS SING AND THE GRASS GREEN!!
When I use an editor, I don't want eight extra KILOBYTES of worthless help screens and cursor positioning code! I just want an EDitor!! Not a "viitor". Not a "emacsitor". Those aren't even WORDS!!!! ED! ED! ED IS THE STANDARD!!!
TEXT EDITOR.
When IBM, in its ever-present omnipotence, needed to base their "edlin" on a UNIX standard, did they mimic vi? No. Emacs? Surely you jest. They chose the most karmic editor of all. The standard.
Ed is for those who can remember what they are working on. If you are an idiot, you should use Emacs. If you are an Emacs, you should not be vi. If you use ED, you are on THE PATH TO REDEMPTION. THE SO-CALLED "VISUAL" EDITORS HAVE BEEN PLACED HERE BY ED TO TEMPT THE FAITHLESS. DO NOT GIVE IN!!! THE MIGHTY ED HAS SPOKEN!!!
?
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
9
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
Brodie, you also missed a couple points, or perhaps just didn't emphasize them, about the advantages of source-based distros and Gentoo in particular.
1. Security: It can be argued it's easier to hide malicious code in a binary than in source, nad therefore compiling from source yourself is safer. However, at the end of the day, you almost certainly have to use some precompiled binaries: gcc and rest of the toolchain as well as the live environment you are installing from. You have to trust someone at some stage. Sakaki's EFI install guide on the Gentoo wiki explains some of this and goes through a full bootstrap recompile process. Very interesting and fun stuff =).
2. USE flags: From the little I understand, another advantage of compiling from source is that you have the oportunity to include or exclude optional portions of the code. For instance, if I am running X and don't care about wayland, I can tell gnome I don't care about wayland support by removing the wayland use flag. Then the wayland portions of gnome are not compiled. I am not sure how well this works. It also relies on modular code that is marked as such for Portage.
3. Patch sets: You can apply patches to any package automatically using Portage. If you write some bit of code to improve a package, it is possible to have that code change be applied on every update of that package. Obviously this could break at any substantial package upgrade, so is not for the faint of heart, or those with precious little time, but this is still a really cool feature.
8
-
8
-
8
-
8
-
8
-
8
-
I received the list of questions for my Written Interview a day ago. I was shocked! As Brodie mentions, there are embodies sub-questions within each bulleted question. The total # of questions is close to 30. This is for an SDR Team Manager role. In the past, I have gone through interviews that require a psych assessment and don't have a problem with them. I have also been asked to put together a "final presentation" in the form of a PowerPoint. I didn't mind that either. But this situation and hiring practice with Canonical seems ludicrous!!!! You are expected to invest a LOT of time in these essay questions only to get you to the interview rounds! And with ZERO guarantees! So, I emailed the recruiter back and offered to submit the questions IF she could FIRST answer some of MY questions. In the job post, there are ZERO mentions of compensation, base salary, commissions, splits, benefits, etc. Only mentions a few perks that seem ridiculous and that I could care less about. So, I want them to answer ME first about what I can expect (or at least a range) for how much this role pays. Show me the finish line and what it's worth for me to do hours/days worth of dissertation writing before I start writing it. If they are unwilling to answer a few of MY easy questions, why would I answer all of their extremely detailed ones?
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
Good video, got a few comments.
vim use <C-w> as a prefix for windowing system, like C-a in GNU Screen or C-b in Tmux, so it's probably to keep uniformity with the rest of those commands. <C-w><C-{h,j,k,l}> also works as well. Not that <C-w>q will close the current window, and <C-w>o will delete all other window and keep only the current windows, hence o. Perhaps C-{h,j,k,l} was not see as to not conflict with the control characters that might have caused problems in the past.
set relativenumber without set number will show only 0 at the current line, without the current line number.
Maybe you could have init.vim have only the plugins and keep other setting in another file, then have init.vim source that other file at the end. Then, when you save one file, it source that file.
Can't wait to hear about your plugins!
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
> Simulate a conversation between a C zealot and a Rust zealot arguing about the addition of Rust in the Linux kernel
C Zealot: "Why are you trying to push Rust into the Linux kernel? C has been the standard language for kernel development for decades. It's tried and tested, and there's no reason to change it now."
Rust Zealot: "Well, I understand that C has been the go-to language for the kernel for a long time, but Rust offers some significant advantages. For one, it has built-in memory safety features that prevent common bugs like buffer overflows and null pointer dereferences. This means that Rust-based kernels would be more secure and less prone to vulnerabilities."
C Zealot: "But C has been around for so long and has a proven track record. It's not like Rust is a new language that hasn't been tested. Plus, there are tools like Valgrind and static analyzers that can help find these types of bugs in C code as well."
Rust Zealot: "That's true, but Rust goes a step further by eliminating the possibility of these types of bugs from occurring in the first place. And while C has a long history, that doesn't mean it's the best choice for every situation. Rust has gained a lot of popularity in recent years for its performance and reliability, and I believe it would be a valuable addition to the Linux kernel."
C Zealot: "I see your point, but I'm still not convinced. C has been the backbone of the kernel for so long, and I think it's a risk to switch to a new language. What if something goes wrong or there are compatibility issues?"
Rust Zealot: "I understand your concerns, but Rust has been thoroughly tested and has a growing community of developers. Plus, there are already a number of projects using Rust in the kernel, so it's not like it's completely untested. And if there are any compatibility issues, we can work through them as a community. I think the benefits of Rust outweigh the potential risks."
C Zealot: "I see your point about the reliability and safety features of Rust. And I understand that it has gained a lot of popularity in recent years. But I'm still not convinced that it's the right choice for the Linux kernel. I think C has served us well for so long, and I'm not sure if it's worth the risk to switch to a new language."
Rust Zealot: "I understand your hesitation, but I think it's important to consider the long-term benefits of Rust. While C has been the standard for a long time, it's not perfect and there are still security vulnerabilities that can occur. Rust offers a solution to these issues by eliminating the possibility of certain types of bugs. And while it may be a risk to switch to a new language, I believe it's a risk worth taking considering the long-term benefits it could bring to the kernel."
C Zealot: "I see your point. I can understand why Rust might be a good choice in the long term. And I'm open to hearing more about its benefits and how it could potentially improve the kernel. But I think we need to carefully consider all of the potential risks and implications of switching to a new language before making a decision."
Rust Zealot: "I completely agree. It's important to carefully consider all of the potential risks and implications. But I think the benefits of Rust far outweigh any potential risks. And with the growing community of developers using Rust, I believe it's a language that we should seriously consider for the Linux kernel."
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
Ok, here comes my wall of text which you probably expected :v)
I personally also have system and vim clipboard synced. Don't know anyone who doesn't use this. I feel like it just is a lot more comfortable also it is a nice binding in the way that it doesn't actually remove any normal vim functionality, you can still copy to and from other registers like before using ", ctrl+r etc
I like how you skipped over some of your more obscurre bindings like the normal mode ctrl bindings that might work in your opinionated setup but will seriously harm someone planning to get properly into vim.
exeptions here are kind of the window switching bindings. They might not be the most popular bindings but I tend to use C-h as backspace all over my terminal as it's standardized and ctrl+j is newline, though I don't use that one that often and instead use ctrl+m for enter. Those are the actual sequences sent to your terminal when you press things like enter or backspace btw. You can check with ctrl+v.
Personally I use alt+hjkl to switch windows.
Btw the reason why it's this weird emacsy kind of keychord is probably just for lack of keys. All the window management shortcuts are started with ctrl+w. So you have ctrl+ws for a split, ctrl+wv for a vertical split, ctrl+wo to close all windows besides the current one, ctrl+wq to close the current window, ctrl+wK moves the current window to the top. The list goes on and on. It feels kind of like a WM almost.
rebinding S is risky business since it's a single letter in normal mode but it can be fine I guess since S does the same thing cc does. capital I as a substitute flag does disable case insensitivity, that is correct. By the way you can also manually disable or enable case insensitive searching inside of a normal search with \c and \C
also scrolling works in nvim without setting mouse for me at least. I noticed that your st build seems to be kind of broken with it's scrolling in general though, when you made the video about it pasting characters, when you want to scroll. Maybe you don't need that anymore now though since you now use alacritty.
The reason why you set both relative number and number is that with only relativenumber you just have a 0 at the current line. With normal numbers you then have the actual line number at that position instead.
Also another binding that tends to be pretty popular amongst people is binding j and k to gj and gk instead. Maybe also interesting since you have your gj and gk remapped.
What this does is move the cursor a visual line instead of an actual line, meaning that if the text wraps around because of a too small window, you will be able to navigate the wrapped line like it was an actual line
Overall a really good video, I approve!
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
Discord is not the best thing ever and it has a god darn bunch of issues, like this one, but I actually really find specially annoying the open source people who love to point out how some proprietary software is bad and you shouldn't use it, while not providing any value. OK, it is shit in a bunch of ways, usability-wise it still is more concise and featurefull than any of the alternatives (Matrix and XMPP clients are really half-baked and missing features, IRC is inadequate, whatever other stuff is really finicky to use), and, most importantly, it is a chat program, not using it means I'll cut contact with all the people and groups there, and I'll not go DMing every single contact I have to annoy them into using something else just because of me.
I fucking get it, proprietary bad, stop repeating the same phrase over and over again without doing anything and expecting things to change just because you're angry, want people to use open source counterparts? work to make them as good or preferably better than the proprietary ones, don't just settle for some half-baked alternative and then pretend it is perfect and harass people to not use the proprietary thing
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
When people do screenshots of “linux” they are usually showing you the desktop environment. You can also run most of those on other Unix and Unix-like operating systems like FreeBSD, and you likely won’t notice any difference, at least until you do system admin tasks. FreeBSD is a bit like MacOS, and Android, where you update the base system and the packages separately, whereas on Linux, the package manager generally does everything.
BSD Init is very different to SystemD, and I find it a lot easier to work with, And in FreeBSD, we have variants instead of distro. Generally the base image is different, but the ports/package collection is the same.
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
Yeah, this has the problem so many attempts to strip Windows down have: It goes too far. For me to recommend this thing, there's four things that need to be done:
1. UAC, re-enable it. It's not the end-all, be-all, but it's an important thing for both security and privacy that uses system approaching "LOL moving your mouse cursor takes more". Leave it be.
2. Windows Defender. It's not the best AV out there, but it's basically effective and it's the lightest-weight always-on antivirus that exists. Just use it.
3. Mitigations. Ask during installation, default yes, and caution people against no. If your system has hardware mitigations, the software mitigations speed up your system. If you do not have hardware mitigations, new ways to abuse that ARE being created all the time. However they're the biggest performance impact of the four, and if you're never going to have the machine on the Internet (offline MAME cabinet or something) you can safely turn them off for a sizable boost in performance.
4. Windows update. It needs to work. Not automatic. Bonus if I can schedule it to do its thing at a specific time, but I don't want it always churning in the background.
Without that, I just can't recommend this thing. And it's too bad, because Windows telemetry, bloat, and bullshit are just unacceptable these days.
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
It feels like it's getting easier as Wayland becomes default for more users. It was easy for Wayland to say "well behaved apps shouldn't do that" when it was only a relatively small number of developers and early adopters asking for these features, but now that the unwashed masses are using it, it's undeniable that, yes, we really do need screen capture, global shortcuts, screen tearing, absolute window positioning, cursor warping, and always-on-top windows. After 15 years of slow-as-molasses development, the last year feels like a huge shift.
And as much as I hate to admit it, some good things have come out of Wayland's reluctance to accept these features, like support for sandbox engines, so your compositor can let you opt into an app recording your keystrokes in the background.
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
FWIW, this isn't MATE's first mention of migrating to Wayland. A year or two ago they had something experimental if you wanted to use Weston or some other compositor … it was experimental but most things probably mostly worked. They did have a tracking age for what needed to be done much like XFCE has had but that was all developer-level stuff, not really for end-users.
Neither XFCE nor MATE seemed to have the developer manpower to move forward with any speed, so they kinda just chugged along in the background working on it where they could without much fanfare. It'd actually have been really cool if the Cinnamon, XFCE, and MATE folks would pool resources a bit—but I know that's not totally practical.
Cinnamon is Gnome 3 based, for all its faults for what the Linux Mint team wanted from a DE. It was a hard fork and they haven't tried to integrate with Gnome anything so they have to do everything the hard way now. Now that MATE is GTK3-based they can use MATE apps to fill in holes in their own suite but they don't have much to share with MATE's Gnome-2-forked codebase. And XFCE was always its own thing.
I ran XFCE, but some bugs caused me to need a solution, so I switched to MATE. The MATE window manager marco did not suit well, so I switched that out for compiz. Buggy Disaster, because compiz always has been. Wound up with a weird hybrid of MATE's session and panel running on xfwm4 with a mix of both MATE and XFCE apps. But then it looked like XFCE had fixed my bug (they had) and I could get what I wanted moar bettar going back to XFCE and oh yeah they were making more progress on Wayland … so I went back to XFCE with just a combination of apps.
I don't even know if I much need a DE, honestly. I need working font scaling. A menu thingy. A bar with a few convenient mousey-clicky buttons and a grid pager on it, maybe input mode and volume. No clock—I do that with a conky. Huh. Wayland conky … I should investigate what exists. edit OH YAY, it just works, neat! Now, font scaling… That's the big one.
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
Well, in terms of backwards compatibility, Rust isn't really different from C
With C, you have several standards available. Like C99, ISO C, C11, etc. Compilers support them simultaneously and you can just stick to the standard you originally picked and that is supposed to work
Now when it comes to Rust, it is very similar. There are several editions of the language and you can pick any one of them for your code. They aren't (backward|forward) compatible with each other in the sense that valid code written in one would be valid in another, but they are fully compatible when it comes to importing whole modules. You can use libraries following different Rust edition than you've chosen
Currently there exist Rust 2015, Rust 2018 and Rust 2021. What's more, all editions will be supported in Rustc indefinitely
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
I fear I'm not going to get a warm response with this comment, but oh well. For context: I am a packager for Fedora, and was involved in getting "orjson" packaged. That said, I have nothing against projects preferring to ship their software themselves as flatpaks, and I use several flatpaks myself.
HOWEVER. The kind of integration work that happens when doing distro packaging doesn't happen at all in the flatpak case, and in my opinion, this might cause problems for the ecosystem in the long run.
For example: The issue that originally caused the Fedora package for bottles to get out of date - the missing package for the "orjson" Python package - mostly occurred because the orjson code was doing unsafe things that crashed Python on some architectures we support. I actually worked with upstream projects (PyO3, orjson) to get things sorted out, and got them fixed properly. That took some time, but now that releases of these projects are out which contain these fixes, every user of these libraries benefits (yes, including the bottles flatpak). Now, I'm pretty sure that bottles developers didn't even realize that one of their dependencies was broken in this way (we only knew because we run test suites for packages we build!). So, Fedora packagers trying to keep the package alive very likely caused those bugs to be discovered and fixed in a way that everybody benefits.
I'm not saying our processes are perfect, but you do get benefits from caring about stuff like this that you dont get if you just bundle all the libraries you need and yeet flatpaks at your users. It's not always pretty, it sometimes takes longer than you'd like, but the end result are improvements for the entire ecosystem.
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
I actually applaud Canonical* for aiming for a default installation that targets providing 90% of users with 90% of the software they need. IMO, the default suite should contain, beyond the base "server" installation of systemd, apt, snap—fine, because we're Canonical—etc
- That flavor's desktop environment with its preferred settings configurator
- Graphical "software store"
- File browser
- Terminal emulator
- bash, Python, curl, wget, git, nano and vi[m]
- graphical plaintext editor (gedit/kate)
- Web browser
- PDF reader and CUPS printer / scanner suite
- Image viewer (*not* a photo "manager")
- Simple music / video player
- Font manager
- LibreOffice because since the majority of people still don't export documents and slide decks to PDF, users still need a way to view (if not edit) those files without uploading them to GSuite.
But that's probably it. No IDE for sure, no GIMP, no Shotwell, no iTunes clone, no Steam, WINE or Lutris, and no email client, because who the heck uses an email client in 2023?!
*wow, I don't say that every day
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
No, that absolutely is escaping, explicitly so. The shell uses three types of escape sequence:
The backslash escapes one single character following it. It only really works (or is needed) on a few otherwise special characters, however.
The double quote escapes every character after it until it hits another double quote, while allowing a handful of special characters to still be parsed (particularly the backslash and $).
The single quote escapes every character after it until the next single quote, with no exceptions.
Note that the quotes act more like toggles than containers. They are simply processed left to right. Escape on, escape off, Daniel-san. (Exception: some nested constructs like $(...) create their own parsing contexts, and any quotes inside them are treated separately from the line outside it.)
bash and some other shells, also have $'...', which acts like single quotes, but any special backslash pattern inside it is instead expanded into the actual character it represents.
See "man bash > QUOTING" for a more detailed explanation.
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
it depends. Generally, if you contribute some code to my GPL project, you retain the copyright for your contribution, which means I need your permission to relicense.
That is because the code I wrote myself is licensed from me to whoever gets it, but the code you contributed is licensed from you to whoever gets it, including me. I am allowed to use your code so long as I obey the license you put on it, which presumably would also be the GPL so everything is fine.
If I want to change the license later to something that isn't compatible with the GPL, I can do it as the copyright holder for my own code, but I can't do it for yours. You'd have to either legally transfer ownership of your code to me (as in, sell or gift the copyright) or license your code to me again under a different license that is compatible with my new license. Alternatively, I can rip out all the code that isn't mine, rewrite it myself, and then I can relicense when I'm no longer distributing code that someone else owns. This is a key feature of Free Software/copyleft licenses that distinguishes them from permissive licenses that are "just" open source.
If the project were open source but not free software, say under the MIT or the "do whatever the F*** you want public license" (which save for the asterisks is a real license some projects use), then it might be easier. Licenses that allow for use of the code in proprietary software, so if you contributed under MIT, I am allowed to just yoink it and publish it proprietary without asking you or ripping your code out.
The final scenario is when a CLA exists: this is a Contributor License Agreement, which basically outlines the terms under which you the contributor provide your code to me the project owner. In that case, I could require that even though the project is GPL, you license your contributions to me in a permissive way, such that I am allowed to relicense, or even that you just give me the copyright altogether, at which point it's my stuff and I can do whatever with it. So long as I only accept contributions from people who agree to the CLA, I am covered.
It would appear in the case of DuckStation, there was no CLA, but the owner went around asking everyone, some people said yes, some people said no, he took out the parts contributed by the people who said no, rewrote them, and then published the new version under a corresponding license, with permission from everyone whose code was still in the project.
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
Note that many early computer keyboards did have a control or escape key left of the 'a' key, including the keyboards for the terminals on which ed, ex and vi were written on. It's only once the IBM PC became dominate did such radically different keyboard layouts die out.
As people have said before, setxkbmap(1) can also remap keys, but at a higher level. If you less /usr/share/X11/xkb/rules/xorg.lst, you can search for options and it gives you a lot of options for remapping the various modifier keys. The downside is that, unlike xmodmap(1), I believe it is a lot more difficult to write you own rules if you want to or need to.
For one of my machines, I have (among other rebindings) remapped caps lock to Control_L and used xcape(1) to bind pressing it alone as Escape. This works surprisingly well, as Vim has a fair number of keybindings that use Control, and even without xcape, you leave Insert mode with Ctrl-[ (which, in terminal-land, is equal to ESC). EDIT: Also, I have set pressing both shift keys will toggle caps locks, which is good compromise between having a dedicated button and throwing it completely away.
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
My most comfortable controller is also the PS5 controller, but the Steam deck is the most comfortable handhand I've ever used, even if not perfect (which mostly I think is down to the weight). I definitely found a marked improvement after applying the JSAUX PC0103 Valve Steam Deck Silicone Soft Cover Protector.
The bumper buttons, I honestly never noticed how loose they are on the Deck but now you've mentioned it, its very obvious, though its never caused me any problems as I've been playing Pinball FX on mine and not had any false triggers. Though my OLED hasn't arrived yet so no idea if its the same, but I suspect it is as they just made the button higher quality, not changing the plastics.
As for the refresh rate vs frame rate cap, its actually not just that reducing the refresh rate might be a problem, by having it at a higher refresh rate and duplicating frames it actually reduces the latency between the frame being ready and it being output to the screen.
There's definitely something off with your local transfer, I've tested it on the LCD model and its the same speed as over my Gigabit broadband. When using it between two desktop PCs, its often faster as all my Windows boxes are connected at 2.5Gbit. Although its possible if a game downloads large files then extracts them to lots of small files, it would be slower over LAN as its then having to copy those small files not the big archive.
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
I was really thinking that this could be real, many mention Fedora totally understandable, I was thinking of nixos or freebds or even more deep Alpine Linux as a desk. I understand all the comments about Arch, I use it since I decided to enter the world I use only Arch and absolutely nothing more (although I do nothing to use Arch, use should be exactly the same but I do not like APT for any reason. ..).
In fact I want to try different things such as Gentoo, Nixos, Alpine, BDS System, Void, Slackware, etc. For example, mainly to learn and explore other systems other than ARCH and the mainstream (Arch Linux in my opinion is quite mainstream not as much as Debian)
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
Thanks for the inspiration, I might add some of those to my config. Here is some of my polybar setup, maybe finds that useful as well:
- icon for battery (charging, full, discharging, discharging+low), click: percentage notification
- icon for ethernet connection (only visible if connected)
- icon for wifi (disconnected: gray, connected: white, click: network manager GUI)
- 2-character-code for keyboard layout (click switches german/us layout)
- icon for bluetooth (disabled: gray, enabled: white, connected: blue), polling every 10s, udev/ipc is on my TODO
- icon for volume (muted: gray, three speaker icons for low, normal, high, click: mute, scroll: +/- and percentage notification)
- icon for package updates (only visible if there are updates)
- and the obvious ones: workspaces, date and time
Regarding location and style:
I have it at the bottom, so I only really see it when actively looking at the bottom of my screen. Most of the time I look at the top or center of my screen, so a bar at the top would be more intrusive. Also I use less visual effects: dark background, two levels of gray (normal/inactive) for fonts and icons (fontawesome). Then a light red and light blue font color for active/alert. No underlines, no module background (just bar background everywhere). My background color is the color of my monitor bezels, which makes it even less intrusive.
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
I actually took speed reading in school. The very first day, the teacher put up a screen and turn the film on. All we saw was flashes of light. She turn off the film and ask us what we seen. Nothing but flashes of light. But teacher swear there was readable text on this film. We use this light machine to read our books. The book is prop up with a metal wire in the center of the book. So the book will stay open. Then this thin line of light goes down the page one line at a time. We dial that light to our speed, so we can read and understand the book that we are reading. We was taught to skip works as we read. The common words that you use to make a sentence. Words like; like, to, a, an, the, can, will, that, do, go, get, make, know, take, see, come, look, use, find, give, tell, now, how, also, not, as, here, so, very, them, him, me, my, it, many, more, new, any, down, or, if, in, for, by, into, about.....well you get the picture. Because you can read the other words in the book while skipping these words and put that sentence together pretty easily. Then we was taught how to read words differently. Can you read this?
Aoccdrnig to a rscheearch at Cmabrigde Uinervtisy, it deosn’t mttaer in waht oredr the ltteers in a wrod are, the olny iprmoetnt tihng is taht the frist and lsat ltteer be at the rghit pclae. The rset can be a toatl mses and you can sitll raed it wouthit porbelm. Tihs is bcuseae the huamn mnid deos not raed ervey lteter by istlef, but the wrod as a wlohe.
These exercise actually taught me how to speed read. Slowly we speed up that one line light to our books to go faster. And still able to read the book easily. Couldn't believe it. After our 6 months of being taught how to speed read. We all watch that same film on that first day of class. And guess what, there was readable text on that film. That we thought was just flashing light. We was taught by a great teacher.
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
Like or not web applications are the future and will likely replace native options for most stuff
With the increasing fragmentation of the OS space developers are forced to develop 4 different native apps to reach everybody (windows, Mac OS, iOS android android) all with similar but slightly incompatible ways to do things.
In Linux land things are even worse as the choice of a framework to build against will get you screamed at (I’m being overly dramatic here but you get the point) by at least half of the community.
Don’t even get me started on distribution which is a complete dumpster fire between having to appease Microsoft, apple and google to get digital signatures and approvals for stores (and that is the easy part). For Linux is again even worse, with many popular packaging formats (deb, rpm, snap, flatpack, and countless others) with constant flame wars between them that are bound to get you yelled at.
You could deal with all of this, or just develop a PWA, package it with electron and call it a day.
As a developer/business owner what would you pay for?
The burden would be on the community to make the developer life as easy as possible, however it is much more important to yell at each other above which of the two 100M+ corporations with heavy NIH syndrome is the evil one…
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
6
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
I have tried a ton of text editors, including typora (a wysiwyg sort of thing), and still haven't found one that recognizes footnotes which is something I work with everyday (social sciences), so I just use geany (doesn' t either btw, maybe some developer can cash on this, because a lot of us on social sciences are running away from windows/word and all that environment) because it is lightweight (it runs in a pentium with antix linux, that I rescued and use it as my typewriter) and pandoc does the magic turning it to pdf, that's my input right now, useful vid Brodie, nice content as always, thank you
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
I doubt if there is any OS that doesn't require troubleshooting (I trouble shoot Windows for my friends over the phone -- not without stress), but I have found Mint to be the easiest for me. There's a different version of Mint called Linux Mint LMDE, which is not based on Ubuntu, which might get round your problems. Or if you have a very new computer, it's possible that you need the latest and greatest to work with your hardware, so you could try Linux Mint Edge. It might well be that another distro would be better for you, but I haven't found anything that is easier to get going and keep going than Mint.
It's also the case that, with normal luck, getting stuff going properly is a one time thing. Try entering a search with the name of your hardware, your distro, and a description of the problem, and see what Duck Duck Go or Google brings you.
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
As a Brit, the Type G plug is the safest in the world, when in or near the socket.
On the floor Brodie,13:20, is correct. I've died many times, but I'm getting better.
As for USB-C, I love it's plethora of standards, most of which are ignored by cheap Chinese manufacturing. Saying that I'm careful these days over which cables I buy, and insist on USB-C charging, unless it's something I want from 10 years ago.
Even then Valve had the foresight of this upcoming battle of the USB connectors 10 years ago, and sensibly decided the Steam Controller should use AA batteries /s
5
-
5
-
5
-
3:50 That's a preference thing I suppose.
I personally greatly prefer the Deck's light triggers.
But I also use them in weird ways, e.g. this very comment was written with my Deck's onscreen keyboard, using trigger clicks so I can type faster with it.
Which would be fine for (even lengthier) normal texts, if the darn thing wasn't as buggy as it is.
With my LCD Deck, I'm completely fine with the emulated rumble from the trackpads, after reverting the default change of increasing their intensity.
I don't quite recognise the trigger haptic issue, but I do know that this setting can be changed in game specific layouts I think.
Outside of specific game layouts, I use the actual D-Pad over the buttons.
The pads still work surprisingly well for all kinds of weird things.
Perhaps the over-saturation when streaming has to do with the colour profiles?
Uncertain on that, last time I used it on my LCD Deck it was fine, but game streaming often gets broken in updates.
Along with the onscreen keyboard and closing non-steam programs in game mode, though the latter seems fixed at last.
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
I don't because I find the standard file handling utilities (ls, mv, cp etc.) sufficient for the things that I need to do.
I find the ability to use wildcards to handle many files at once quite powerful and I'm not sure how a file manager could imitate that.
Furthermore, the only 'file manager' I use at all is netrw (even that is strictly for opening files in Vim) and when I do, I find the need to move between directories one at a time quite cumbersome for anything more than 1-2 layers of subdirectories in a project folder. By contrast, cd (complemented by zsh's excellent completions) allows me to move to where I want to be in a single step.
I'm sure file managers probably have the ability to type a directory name to move there, but, at that point, why not just use cd?
As for file previews, I can often make a good guess at what a file contains based on its name.
Edit: Granted, I've never used a ranger-style file manager (my experience is limited predominantly to vifm), so maybe I just don't know what I'm missing out on.
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
This is a response on github from the maintainer to an issue I raised:
Huh, well for the past few weeks I've been working on performance to the point that hunter is now faster than anything else out there as far as I can tell. It's almost twice as fast as nnn at loading obscenely large directories (think 1M files), plus it doesn't block while loading (major design goal). Note that I hold nnn in high regard, it's been the benchmark I've been single mindedly trying to beat for the past few weeks. If you know anything faster, please tell me :). Work on performance and CPU efficiency is still ongoing, however... I can probably make it another 50% faster by using the right low-level APIs and other optimizations. Maybe more by utilizing a persistent on-disk cache or something.
Things I plan on adding:
Additional graphics backends (urxvt, iterm2, w3mimage/überzug as fallback)
More advanced input options (custom layers/multi-key maps and macro recording/playback)
Better minibuffer with shell script syntax-highlighting, better editing and using the preview pane to list all (matching) executables
Also more interactive directory traversal
Browsing into archives like folders, maybe http/ftp as well, maybe using FUSE and/or other external helper utilities
Turn the subprocess viewer into somewhat of a terminal multiplexer like tmux/screen (especially to get colored output and to get input to subprocesses to work)
A new README, since the current one is pretty outdated...
A new screenshot, the current one is terrible, even the previous one was much better lol
A TODO that explains all this in a bit more detail
Lots of bugfixes and smaller QOL improvements here and there
Even faster!!
It's true that I've not been promoting it a lot. I only posted it once on HN almost a year ago and that's it. I also stopped working on it for almost half a year after getting a bit burned out, only to come back later to improve on and fix what I started. I guess the downtime hurt it a bit in terms of popularity, but currently I'm still working on getting it to a state that will make people go "WOW!!". Or at least that's the plan ;).
Thanks for your support! I appreciate it!
Maybe could u put some issues on his github that aren't already on his radar? I dont want to point him to this vid in case it bums him out!
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
"They just picked a mineral and went with it." - Yep, pretty much. Sounds cool and there are green opals, with green being the Budgie official primary branding color, but honestly I would've figured a way to shoehorn in something anyways..
All that said, I agree that the naming should be "Fedora Atomic X" like "Fedora Atomic Budgie", "Fedora Atomic GNOME", and so-on. Fedora seem hesitant to go back to the Atomic name but I can't say names like Silverblue really do the branding any justice either. Your "average user" isn't going to know the difference between Fedora and Fedora Silverblue, just as they wouldn't between Fedora and Fedora Atomic, but at the very least it would help communicate it with those that are at least familiar with atomic, immutability, etc. and keep the naming more consistent.
Until then, I just continued rolling with the mineral naming and avoid shaking the (OS)tree too much with a proposal to rebrand all the things :D
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
I find the way they say that Libadwaita "don't support [Linux Mint, XFCE, MATE] users" to be rather strange. It's not like these apps don't work on those desktops. The only "issue" is that they have different design languages. But this same attitude doesn't happen with Steam, Discord, OBS, Electron apps, etc that all have their own themes.
I get that Linux Mint wants a cohesive desktop after an install, so I support their creation of Xapps, but I don't like how much they're making it seem like Gnome is doing something bad here with phrases like "Libadwaita is for Gnome and Gnome only". Just because the apps don't match your theme doesn't mean you cant use them, and again, apps like Steam don't get this same criticism.
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
5
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
I bought an Arc 750 in late 2022, and while the early driver experience was ... rough (as i expected, looking at reviews beforehand), it has improved at a frankly astonishing pace ('46% FPS uplift in game X' being a common thing to read in update notices), and is, at this point, downright decent. In some areas, Intel even has a lead over at least AMD, particularly frame generation (XeSS is has always given me better image quality AND performance than FSR 2.0).
Sometimes you run into software which assumes any Intel GPU to be an iGPU (giving me a certain Y2k vibe), but that's not exactly Intel's fault.
I'll certainly take a good look at Battlemage, and upgrade either to it or Celestial once that comes around
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
hey Brodie! longtime plasma user, occasional tiling enjoyer, I just wanted to mention 2 things that you might wanna try if you plan on using plasma a bit longer. the first one is a Kwin script called Polonium. it's still beta, it's kinda finicky, i stopped using it after it crashed kwin, and i had trouble getting windows that were already open to go into the tiler when i first enabled Polonium, but it does act like a basic tiling window manager and adds a dozen or so settings to the kwin keybinds for you to use.
second, is the quick tile functions. this is just basic plasma stuff, i believe it has the 4 half-screen ones set by default but not the 4 quarter screen ones. i have the quarter screen ones set to meta + h,j,k,l, and it's just keyboard shortcuts that automagically set a window to one of the basic 8 half-screen and quarter screen zones you normally set windows onto by dragging them to an edge. it's a far cry from what you get from a real tiling manager or even from using polonium. thanks for the videos!
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
I was going to talk about 1000 (ISO notation) and 1024 (IEC notation) "megabytes", but it was already commented, so I'll just leave this 2 complementing articles
https://chrisdown.name/2018/01/02/in-defence-of-swap.html
https://wiki.archlinux.org/index.php/Power_management/Suspend_and_hibernate#About_swap_partition/file_size
48GiB is a effton of swap
Also, you mentioned btrfs and 5.0, I'll change the wiki later, the LTS kernel in arch is already 5.4, so even if you install the LTS kernel, you can have support for swap in btrfs
And about hibernation ... if you have 32 gibibytes, you probably could "Suspend", instead if saving to the storage, it saves your state in the memory.
Nowadays, the fastest NVM-Es are about the speed of DDR1/2
Crude example: DDR1 400MHz, (were processors 32 bits when memory was ddr1?)
400 x 1000
x 32 transfer rate (because of the CPU)
x 2 (DDR = double data rate)
÷ 8 (ammount of bits in a byte)
3200 Megabytes per second
Also, you mentioned "a line in your fstab" for swap
I worked along a guy to bring to the archwiki more information about XBOOTLDR, which is a systemd specification - "Extended Boot Loader" - with systemd-boot, specifying the partition type for root, swap, home, boot and efi, I managed to made an arch machine boot without fstab (systemd can auto mount based on the partition type), it didn't got screwed up by a major windows update, and it even recognized swap. It would be cool if you could give more visibility =) ... https://wiki.archlinux.org/index.php/Systemd-boot#Installation_using_XBOOTLDR
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
While Plasma is really powerful and feature-rich, you can always feel that they lack people with UX design skills. They tend to add loads of features and make things as configurable as possible so users can set up their workflow. But issues like this make this almost impossible. As a user, I see all the options and I think I'll make it work for me, but then I'm stuck at 80% with a broken workflow. This is frustrating.
Better design the experience first, decide which features are needed, how they fit together, implement those and discard others. You're absolutely right: You should be able to save tiling configs, cycle through them with the windows following, have different configs per virtual screen and monitor, and have the quick tile shortcuts respect the tiling areas. Just to name a few. But floating areas which can be tiled and are not floating after that, this just doesn't make sense at all.
What would work is keybindings to pop a tile out so it becomes floating, or push a floating window into a tiling layout. Other useful/essential keybindings would be maximizing/restoring the active tile, shuffling windows around in a layout, and changing layouts.
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
I've been developing software with Rust for 9 years, so all of Rust's static checks are ingrained into my brain. Yet the compiler still catches mistakes here and there with helpful compiler error messages to guide me to fix them. It's impossible to write perfect code without the help of the compiler. There will be times when you are in a rush, or working late at night, perhaps you are sick, tired, or there are distractions around you. No matter what shape I'm in, Rust fixes my mistakes, and therefore you will never see those mistakes when running a production or development build. That means QA does not waste time with silly mistakes, and I can get on with my life working on more meaningful tasks.
Also, I often work with other people in the real world. Sometimes they are less experience in Rust and programming than I am. Yet the compiler ensures that any contribution they make to projects I maintain are at least free of these same common mistakes. Mistakes that are still rampant in most GNU C software today. Just recently this month, there was a severe security vulnerability in GNU libc because of a silly mistake in the hash algorithm making a root escalation exploit possible.
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
Sometimes, you just have to scratch that itch on your brain, you know?
As an example, I was having some issues with UTCs, where if I transfer media from my camera's SD Card to my computer, the time on the media files would have an additional 8 hours, because the camera's internal clock gets reset if the camera was stored without the battery for a long time, and I take out the battery out when I store it to eliminate the risk of leakage.
So, instead of setting up the time zone on my camera every time I use it, I made a bash script that I have linked to a desktop shortcut, so that after I plugged-in my camera's SD Card, all I have to do is click on the shortcut, and then it would first change my RTC time to local, mount the SD card, copy all the media files from it, then separate them to images and videos inside their dedicated ingest folder, then change the RTC time back to UTC.
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
Once you enter the realm of highly regulated industries, banking, credit cards, insurance, health care you will see 10y systems all the time.
Getting software certified and compliant is an annoying process … so vendors just don’t do it
You buy their tool, it supports RHEL and Oracle of that year and that’s it, they will only support something new with the next major release in X years.
We could run deploy it on a recent OS but that is not supported by the vendor, so no support, so it won’t happen.
You get the software, the matching os, and hardware combo all with 5 or 10y contracts, and after that time you migrate to the new.
Btw Canonical … there has never been any discussion on RHEL or Canonical, my auditors know rhel I get ready to run cis hardening tools and benchmarks on the other side I will not find any certified Ubuntu engineers when I have to hire somebody.
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
I am not tied to anything windows. When i last used if before, all utilites i used were third-party apps. like notepadd++, mplayer, firefox, etc. And everything on windows is just too slow from launching apps to boot etc. and oh, a malware or something delete all my files or at least turned them to zero bytes and I dont game much either. Everything in linux has been great for me. from install size (i never went past 30GB for everything i use on system), its fast (msys2 or wsl just too slow on windows than native linux, yeah i compiled and code alot of stuff), a lot of opensource stuff, its customizable to even your specific needs, etc. 3 things i believe user stay on windows because of two main reasons, Familiarity and Games.
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
Three pillars of good software: code readability, speed (@ runtime) and memory footprint (resources used @ runtime - although extremists also count sloc in source).
I interpret minimalism as low memory footprint. Since "minimalism" has become more of a meme, there have been more interpretations of its meaning. There are many who take "minimalism" to include both memory and speed (refusing to compromise on any trade-offs), to them "minimalism" could mean the same thing as "space-time efficiency" or "value" (best bang for your buck). Others interpret minimalism as the lack of unnecessary features in a program. Lots of these definitions are associated with each other, but the lack of consensus on which definition is the most important has led to many futile forum debates and confused junior devs.
My biggest concern about "minimalism" is how the topic seems to infinitely generate proxies of the editor war (thanks to the reckless passion of developers). Instead of just vim vs. emacs we have: arch vs. debian, wayland vs. xorg, dwm vs. bspwm, urxvt vs. st, nnn vs. ranger, screwdrivers vs. drills etc. Instead of increasing productivity, minimalism has become a sort of wild goose: causing devs to perpetually hop across git repos in search of the most minimalist solution / combination of solutions (https://en.wikipedia.org/wiki/Knapsack_problem ).
The real shame is that there are times when minimalism just isn't practical. Imagine the person who has finished the chase and has achieved the perfect "minimalist" dev environment on their *nix machine. Now imagine that this person gets a new job as an engineer and all the company laptops run Windows. What should they do? Should they decide to use VSCode and IntelliJ like the others, or in the name of "minimalism" should they run *nix in a docker container and copy their dotfiles over (or install an X server on WSL / or download the Windows versions of their dev programs and use mingw to patch the holes / or etc.)?
There is almost never a right answer, but there is almost always a best answer. Letting "minimalism" dictate choices is almost always a wrong answer.
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
00:00 🏗 Building a Wayland compositor from scratch takes a lot of time and effort, usually suitable for reference implementations or significant projects like KDE and GNOME.
01:37 🛠 Louvre is a new C++ Library for Wayland compositors, promising high performance with a strong emphasis on ease of development, simplifying complex low-level tasks.
02:29 📚 Wayland compositor development is more intricate than X11 due to broader responsibilities encompassing a display server, window manager, and more, requiring understanding of various APIs.
03:38 ⚠ Louvre, while promising, is still in its early stages and faces development issues like lacking support for XWayland and Wayland protocols, impacting practicality.
05:51 🚫 Louvre's absence of some Wayland protocols, like session lock, tearing control, and fractional scaling, may hinder broader adoption, though not fundamental design problems.
07:27 🌐 Louvre's potential impact lies in becoming a significant voice within the Wayland ecosystem, collaborating across desktop solutions and addressing existing limitations.
08:23 📊 Benchmarks showcasing Louvre's performance compared to Weston and Sway indicate promising results, but lack clarity regarding tested hardware and direct comparability between library and compositors.
12:31 🕒 Louvre has the potential to simplify Wayland compositor development for many, but its current stage requires observation as it aims to compete in the growing Wayland landscape.
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
Let’s see what I can find in this video ...
* on Arch, /bin is symlinked to /usr/bin, thus /bin/zsh is the same as /usr/bin/zsh.
* there are ways to modify the environment without .profile scripts: rc scripts (not with sysD what Arch sadly uses), on OpenBSD the file /etc/login.conf, on Arch the ENV_PATH variable in /etc/login.defs, ...
* even if you change the `shell` variable in st, it uses a few other things to determine the shell used, one of them $SHELL, which gets exported by login(1)
* there should not be any problems running scripts (if $PATH is set correctly and stuff) after changing shells, because they are started as ordinary programs and the interpreter is only determined through the shebang (the #! line at the start)
* on Arch, /bin/sh is a symlink to bash, I changed it on my machine to dash
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
@BrodieRobertson
Harsh criticism ahead, proceed at your own risk.
the intro is awful in my opinion and very repetitive (same words every time), the song is bad, I never liked tech video intros. I can say this definitively: this very video can be compressed into a two minutes video without losing anything useful. Luke's videos, for instance, don't span more than they have to and they always have so much to them that you feel like taking notes or rewatching 3 times. your videos, on the other hand, are 10% useful information, 90% filler. I don't have to watch you go to some dir in lf. have it prepared already and let me figure it out from the prompt, don't even tell me what dir I need to go to.
In my opinion, you really have to let the intro go, it sucks (it feels as if you have three intros, btw. you, the animated one, then you again wishing me a happy birthday or something).
Just think of how you turned a two-line (script not included) entry in ArchWiki into a quarter-hour video. This is bad, and you should feel bad.
I realize I'm being too harsh. As for the beard, I could not give a shit to save my life.
Love ya
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
Always good to keep it simple. I like the scratchpad for Xscreensaver, Timeshift, Firewall, Character Map (mostly when I add items to my persistent list in Parcellite). My puter's too slow to do serious video, though I have one ws for kdenlive for when I feel like torturing myself of my handful of subbies, one for LibreOffice, one for VLC and Rhythmbox (I float those windows with big gaps), one for Meld, one for Google Earth, one for Nemo and Atom, one for Cherrytree, and one for Firefox, i.e. most apps are full screen. I like tabbing mode when I feel like clustering 4-5 windows when I have stuff to juggle or hop around in.
Minor suggestion: raising your laptop so the webcam is closer to eye level but still a touch under, so we're looking up to you less :) And really cool the subs you grabbed overnight. You're earning them. peace
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
There is a video here on youtube with a madman who installed the latest Gentoo at that time on a 486, in 2018.
And another video from this year, if I remember correctly, installing ... Gentoo (of course it's Gentoo) on a Pentium 133 MHz.
And yeah, they boot really slow 5-10 minutes to boot, another 5 minutes just to shut down. Though you can actually connect to the internet and see or download stuff safely. But overall, there's little reason to install a modern kernel. Though, regarding the speed, the one on the Pentium doesn't seem like he went full optimisation, so it actually might run much better.
I'm with Linus too here. Part of me hates to see support like this dropping, but in reality, it keeps the code more maintainable, without an everlasting list of things to check or maintain compatible. And the devices who get to be dropped support truly are obsolete and also don't really benefit from having the latest kernel.
Now please excuse me while I stash a nice whiskey bottle for 2032 when 586 support will be dropped.
4
-
4
-
4
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
It is very simple, let them set their sane defaults as they like.
But there is a nice video by JayzTwoCents that is titled: "Motherboard Default settings could be COOKING your CPU!"
That video showcases the problems with the "limits" as vendors / add in board partners have the mentality of "Bigger numbers go brrrr and win benchmarks" and sets them a lot higher than needed on both sides.
For CPUs you have to really know what you are doing to get them back to AMD or Intel's specs pending on the BIOS, but it is the same with GPUs, well at least GPUs where they aren't being forcefully controlled like Nvidia.
I like an ecosystem more open like AMD, even to board partners where they can play around with the hardware, and if we are to have this, then we must also have the ability to fix their "brrrr" moments ourselves as users of said hardware.
Either that or full Nvidia, and it is a shame Alex is a bit short sighted here.
But as you said Brodie, if I get an AMD GPU, I will be looking for a kernel with Alex' stuff snipped out of it, he may have a hissyfit knowing that such may end up existing, but that is his problem.
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
I'm pretty late, so maybe you won't see this, but check out image viewers like fbv and fbi. You can use them to view images in the TTY, so maybe you can make them work on terminal emulators. I haven't experimented with that yet, because I'm mostly interested in experimenting with getting rid of X11 and window managers entirely, at least for most things, and using software for the framebuffer and multiplexers instead. I thought that you couldn't watch videos like that, but someone pointed out that mpv and mplayer both can do that (and fbff), and then I realized that you can do most things that way. Images (fbv, fbi), video and pdfs (fbpdf) all work, and you have dvtm and fbpad as dwm replacements. The only things that you can't do that way are games and bloated web browsing.
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
If you missed the old days, you did miss a lot, though not as much if you looked at all the old stuff while it still existed. Sure, the internet was slow, but it was really something special back in the day. I remember when I became aware of the possibility of watching videos online, watching short videos (because long videos were not a common thing in the days of dial-up). Anime OPs without the actual anime because that was too big. Also, the early days of peer-to-peer, using software like Kazaa and eMule, that made downloading bigger files actually viable, and then when I started using torrents, everything changed forever.
I remember trying to download some anime episodes from websites really early on, and I only succeeded a few times, because it took all day to do and if the phone rang, it was over. It was great to be there and see that kind of thing slowly become possible. Also, downloading games and playing them on emulators was viable really early on, so entertainment post-scarcity was already becoming a thing. Emulators like Nesticle and ZSNES were there in the early days. Then later on Visual Boy Advance became a thing, and that changed things for me a lot, because it was the first new system that I wanted, couldn't have, and then replaced entirely with an emulator and stopped caring about buying things. I remember when the Mew glitch in the Pokemon games appeared online, and I was the only cool kid that knew how to do it. The clone glitch in the second generation games was a big deal as well. There was also a shiny Ditto glitch that not many people know. I remember messing up with HTML as a kid, because having a website or a blog was just the thing to do back in the day. Growing up along with the internet was really cool, man. They don't make them like they used to. sip
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
Pinephone daily user here, so I will have comments!
1:38 Initially I was like why base it on Droidian instead of Mobian, only to find out from Droidian's blog that their downstream from Mobian...
3:15 Honestly Despite the fact I'd rather use XCFE, or KDE instead of Gnome, Posh was the best most functional GUI on Pinephone's screen with it's cheap SoC, and from polls on pine64's forums I was in the majority, easily. I'd love to see the other GUIs improve though, plasma wasn't that bad, just very slow, and the Gnome apps where better.
3:50 So as a Pinephone user (Mobian). Text and MMS work, there is a bug on Version where the modem gets clogged with texts though. I haven't bothered to figure out. And the speakers hardware wise just suck so calls are a bad idea in general, they do sometimes work.
10:50 Ah waydroid, It's great, minus the issues caused by the Pinephone's Soc, also do not launch if your not planning on turning off the phone soonish, it's a battery hog.
12:00 So on the topic of it's arm so less packages.... I honestly did not notice that much missing (Dub, LDC, SDL2, GIMP, thunar, most of GNOME, Firefox, Sublime text along with random cli apps, all just installed with an apt install..). I think the only thing I wanted on my phone that I couldn't get a Linux version of was Discord. And I had to use Waydroid to sign up for telegram, despite telegram desktop coming preinstalled!
13:00 Posh has a Favorites section to group a small set of apps at the top of the list. Just long hold on the icon.
13:55, "Mobile apps made by Desktop devs." Yep, that sums up the experience very very well.
21:07 "It's no longer a phone it's a pocket computer" That's what I wanted! (This is how you end up daily driving a pile of beta software and cheap hardware)
22:22 Once I found out the Pinephone does display out, and I could have a pocket Linux laptop. I bought. This is my top list complaint (above even the old kernal for android drivers) on the Furiphone, I spent more on a lapdock, wireless mouse, and laptop bag for the phone than on the phone itself just because of the DP over usb-c.
23:10 I don't trust a gui app on my phone to update anything. APT (or pacman when I was running manjaro...) failed a system update once! The SD card ran out of write cycles and went read only. Don't use an SD card as boot drive. Learned that the hard way twice.
24:45 ouch.. The pinephone (if nothing stupid happens in the background), has worked as an alarm clock for me in a pinch, and has held charge overnight. In use the battery life is bad. That is definitively a software issue. The pinephone got much better with updates in terms of standby. I do still turn it off overnight and charge it every afternoon, though.
28:45 The price problem, I've seen it solved once (without just supporting some random android device), and it involved shipping 4 in order, old fab process, no SMT arm cores, and a very weak GPU. Looks with both anger and happiness at my pinephone.....
32:42 "Let me know your thoughts down below", Hopeful this wasn't to many thoughts....
Also to any reading, I've been daily driving a pinephone for years now, so if your curious ask away!
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
Hi!
Maybe you already prepared everything, but still ...
Yes, put the swap (I prefer swapfile) on the nvme, and maybe even render on the NVME and then move to the HDD, writing speed of HDDs suck, configure your swappiness value to have a little more performance during swapping.
About swap on flash drives, arch wiki is your friend - "If using an SSD with TRIM support, consider using defaults,discard in the swap line in fstab. If activating swap manually with swapon, using the -d/--discard parameter achieves the same. See swapon(8) for details." - https://wiki.archlinux.org/index.php/Swap
About the network card
I hope you didn't had the same issues as I did, I spent quite a while trying to make my wireless work (another wireless card, so, check the wiki)
Archwiki helped me - https://wiki.archlinux.org/index.php/Broadcom_wireless
Basically I installed b43firmware and broadcom-wl-dkms, ran "rmmod b43 b43legacy bcm43xx bcma brcm80211 brcmfmac ssb wl && modprobe wl" and added "blacklist b43legacy" to /etc/modprobe.d/blacklist.conf
Motherboard for Intel, AMD processor
LOL
Glad you went full AMD. They are wiping the floor with Intel from what I'm seeing
Wireless mouse
I had one, didn't had a good experience, now I barely use while on my linux with DWM
I use more when I'm gaming on Windows
ASUS monitor
No opinion, never seen. I can recommend BenQ, I also have an AOC monitor which is more than 7 years old, I don't even remember when I bought it
NZXT tower
Can recommend, I have one, I don't have a single complaint
Sticking around to this point
I did =)
It's your channel man, you can do whatever you want. But you are surely better in "software hacking" (meaning discovering everything about it)
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
Used brave to view this video haha
I fell for the meme...
It is faster, I noticed it too, probably because so much of the ads and scripts are blocked, I totally agree with you, Google can redefine the web at will =(.
To put simply, Widevine is the thing that allows the browser to play DRM (digital rights management) content, for instance, watching a movie in Netflix
Besides DWM, I also use KDE because I like the project and I know one of the main devs ... in plasma, I don't have the brave icon, so I cannot give you any info about "being an Arch thing"🤔
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
While I agree with you that in retrospect those CPUs were flawed, that's the problem, "in retrospect". As an engineer, when I design something that's within the design parameters, I do that by the knowledge that is available at the time, and if something is proven to have been wrong in the future, I can't be held liable since I followed all the standards at the time. I would, however, be expected to learn from those changes in the standards, and follow the latest recommendations in the future. It would be one thing if it was known that speculative execution done wrong could be a security problem, however if that was not a known thing, I can understand why it was not concidered
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
You're point would be valid if the examples that people use to copy iso files to a thumbdrive (usually for a an installer) actually made use of these features, but they don't (at least I've never seen any instructions for creating installers to use them).
Don't get me wrong, dd is far superior in certain very specific tasks than cat is, but simply copying an entire iso file to a disk isn't a specific task, it's a general one.
Now, I could agree with you in the sense that you might not learn these extra features if you're not exposed to dd in the first place, and that can make some sense. However, dd is at best equivalent to cat when copying an iso, but is more verbose, more complicated, and less readable. Simple solutions (in my opinion) are the best, and that's why I've been using cat in this way for several years, in fact, I actually use cat simply to copy large files because it can be faster than even cp if the file is sufficiently large.
As for "efficiency" I'm not sure what you're talking about. You can find multiple benchmarks for dd vs cat online--or even perform them yourself, and they usually conclude that cat faster than dd even when you set a huge block size like 64M. Now, sure, there are probably niche cases where setting the block size to a specific number might cause dd to perform better than cat, but in general that just isn't the case from my experience, benchmarks I've personally run, and results from other benchmarks.
In short, using dd for copying an iso is like using a scalpel to cut a birthday cake. Sure, you can do it, and you might be able to cut more "efficiently precise" pieces for your guests, but then I would ask whether that level of precision is actually necessary for that task (where I wouldn't question the scalpel if it were being used for surgery).
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
Personally, what matters most to me is perceivable performance and organisation.
What I mean by "perceivable performance" is whether it feels slow to interact with a program. It absolutely doesn't matter how many lines of code there are, not to mention that not all lines of code perform equally.
And what I mean by "organisation" is that everything has to be easily identifiable and accessible. I absolutely hate the one-liner trend. It makes code much harder to understand for absolutely no practical reason. All my scripts have a maximum line width limit of 80 characters. If a line is getting too long I'll rather slash it \
and continue on a new line. If nothing else, it reduces eye strain because I don't have to move them as much.
When it comes to choosing what programs to install, I ask myself a very simple question:
Can this be done effectively with a terminal application?
If the answer is yes then perfect. A GUI isn't needed for a huge amount of tasks. But, if it's something that requires a pointer input, like video/image editing, drawing, sculpting, etc. them even if for some of them there are command line alternatives, a GUI is the natural way to go that doesn't require you to learn some convoluted way of interacting with it, and which won't even bring you a perfect result in the best case scenario.
At least for me, minimalism is about using something concrete that solves a specific problem without touching anything else. Sadly, not everything can be done within the terminal, but you can at least get 90% there and not lose out on anything.
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
I had to take over two libraries that were completely abandoned, one with completely disappeared developer.
The D library, std-experimental.xml was supposed to be the DOM compatible replacement of std.xml, but its original developer disappeared, and as std.xml got deprecated to "incentivize developers to create a replacement", which did not happen, as most projects used JSON for config files instead, or SDLang, which is a de facto standard with D projects. However, I needed a MARKUP language for a text format, since I already was using SDLang for configuration, so first I tried to fix the library (primarily for its SAX parser), but seeing how needlessly complicated it were at parts with its army of templates for everything (it was configurable in theory for nogc use (not in practice, since D unfortunately lacks attribute templates) and what kind of unicode it's using), I threw out most of them, and replaced them with simpler solutions, also made a lot more things more compatible with the actual DOM, and making it work for reals instead of having 500 different configurations. Only things the DOM currently lacks are namespace URI handling (not a popular feature anyways), and some error handling (which is not always triggered).
The other library I took over was an X11 binding, which had issues with getting excluded from Windows when building with dub, because "um, ackchually, windows is poopoo, also write command line apps because they're less memory hungry and you can write scripts for them to automate complex tasks". (Interestingly, I almost never saw such scripts, only people who think they're amazing.)
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
the main problem is that, thorium isn't the fastest browser on earth. it's fast, yes,, but not the fastest. that award easily goes to Chrome.
maybe it's different on x86_64 on Linux, but on my machine, which is a M1 Max Mac,
chrome 113 is 18% faster than thorium 110 in SpeeDOMeter, and 5% faster in basemark.
(the old version of chrome is because the google chrome updater also updates other google software, and google made the macos google drive client completely unusable for many use cases in an update).
all this for a browser that crashes very often
also, the icons in thorium being bright colours looks dumb
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
@BrodieRobertson It's a matter of optimising your resources use. People mostly freak out about Electron because of RAM usage. I've heard from friends that are required to use several electron apps on work, that they basically have no available ram because of it, albeit they have not the best machines to work with. Any electron app, even the most basic one will use 50-150 MB of RAM, depending on optimisation from the developer, because they need separate version of browser running.
PWAs solve this by using a kinda shared runtime of your installed browser.
Electron gives developers huge possibilities, transforming website into a frontend for an app with native functionalities (like fs access and so on), but chrome is pushing new web APIs to give regular web apps same functionality.
And here lies a question: if you have two ways of transforming a web app into a native-like experience, what would you choose: electron, giving you more options but requiring you to support runtime as part of your package, or more limiting PWA approach, that imposes on you less dependencies.
If you ask me, Electron is a bit misused solution. It's so convenient that literally everyone and their dog jumped on it to easily provide users with "native" versions of their product, even when all they need to do is actually make a PWA.
2
-
2
-
From the perspective of a 49(!) years old who started with Linux about a year ago, and who discovered a passion for going under the hood, this is simply the best youtube Linux channel. A natural progression would perhaps be something like Switched to Linux>Chris Titus>Distrotube>Brodie. I don't count Luke, I find his non-Linux content a bit immature. Perhaps you're in that small niche where general interest/non IT professional viewers are near the end of their youtube cycle, before they stop watching channels altogether and move beyond youtube for Linux content. Or alternatively decide to stay at that "level" and keep watching youtube. That's a small, but very important, niche. Your content hits directly home to me, there are no other directly comparable channels, with the same quality. Keep it up, young man! With your passion and obvious intelligence, I'm pretty sure you find a way to grow and maybe even go big if you want, while still catering to the type of Linux users described above!
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
Ah, okay. Stallman actually has a valid point here, though I don't even think he realizes it. He's saying that Debian, Fedora and Arch play a vital role as gatekeepers who audit the software that's included, even in their nonfree repos. The reason PPAs and the AUR exist is that the distro maintainers (leaders in the free software movement) don't deem that software as an essential part of their experience.
Stallman's concern is that the rise of Flatpaks means that people won't be installing the "Debian-approved" build of Blender or Firefox, but the developer-approved bundles, and that this will encourage bad habits. Audacity is a good example. They're FOSS, but they threw in telemetry on a whim, and there was little stopping the Flatpak version from immediately going out to users.
To be clear: while I would agree with this stance if distros had all the staffing and funding in the world, they don't, and the amount of time and the effort spent reviewing and repackaging software across a dozen distros is squandering precious resources.
I also think Flatpak's sandboxing-by-default and dependency isolation features more than outweigh this lack of review.
Especially because, ultimately, these are our systems, we're going to install the software we want, and having to apt-add a PPA has never deterred me in the past.
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
I've seen a few people recommending netrw; if you want to try it, I would advise you to use tpope's vim-vinegar, which is an enhancement of it that adds various features that you can look at on the GitHub page: https://github.com/tpope/vim-vinegar
Also, :x and ZZ are great ways of closing files/splits for lazy people.
Next, I would give you a list of commands for buffer management and other things but instead, have some RTFM:
:h buffers
:h scrolling
Edit: My personal recommendations for buffer chads are :ex, :enew, :ls, :b, and CTRL-6 (This one is my favourite).
Also, if you ever have file names in a file (or want to put them there), such as the css files imported at 6:39, I would strongly recommend gf and i_CTRL-X_CTRL-F.
:h i_CTRL-X is good reading.
Edit: I'm going to be editing this comment until the end of time.
q: is the command you've probably hit a thousand times and wanted to find forever.
q/ and q? could also be pretty good, but I don't tend to use them.
In conclusion, the closer you are to being able to just edit a file in ex, the better.
Once you have reached this godly plane, press Q and edit like a true hackerman
(then use :vi when you get scared).
Edit: Please see the first line of my previous edit.
CTRL-O and CTRL-I (capital i) are great for users of jump commands such as {, [[ and (.
I'm going to turn off my computer now.
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
What the? (not blaming you brodie)
I am pretty sure that krans doesn't mean funeral wreath in Swedish, I am not a native Swedish speaker, so please correct me if I am wrong, I believe krans just means wreath in Swedish, funeral or no funeral, wreaths are especially common also during Jul/xmas, no funeral needed. If I am not mistaken, in Swedish a funeral wreath is begravningskrans, notice the whole other word tacked on at the beginning of the word krans. Oh and krans coming from the old norse word kranz, which comes from the proto germanic word kringaz, which means circle or maybe ring idk.
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
Just a few points, not exactly to correct you, just some information
Ubuntu 20.1
Ubuntu releases every april and october, the version is based on the release date, so we'll have 20.04 and 20.10 this year, 21.04, 21.10 next year and so forth and I think the Budgie Edition is better than what I spent 5 minutes trying to configure on Arch.
Archlabs being lighter
(yes, I read all of the comments of this video) and if someone told you it was lighter, the person was probably wrong. Basically it is arch with some extra repos and a more automated way of installing than pure arch. But, to its defense, I only remember wallpapers from that repo (I might be wrong), so there is not to much bloat/heavyness in that
What I think you might have heard is Artix. It is basically an Arch with another init system [besides several extra features (245 added HomeD), systemd doesn't use so much more memory than other init systems nowadays, computers can handle it, and I live in a 3rd world country, here there are a lot of old/crappy hardware ]
*BSD
I prefer FreeBSD... nuff said
Manjaro being Arch-based
Yes... but also no
I talk to some arch devs/maintainers, almost daily. Some of them really doesn't like Manjaro. They get a package from Arch, modify, put on their own repo, from what I read, not giving credit to the Arch maintainer and also make few to none contributions upstream.
Another thing to support what I just commented above, gnome 3.36 is already present in arch, working and stable, on manjaro it won't come out for a few weeks because it isn't working as intended (I think it was posted by the Gnome team on Twitter, replied by Manjaro - I don't have the link, I saw on the group that I'm in that some of the Arch's devs are)
Liked the content (and the video), keep'em coming.
Cheers Brodie
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
@BrodieRobertson Still me but different account lol. You should install lineage OS btw, it's great. I try to use as many FOSS apps as possible, but I haven't been able to eliminate all proprietary apps.
gcalendar -> simple calendar
chrome -> firefox/brave
pocket casts -> antennapod for podcasts
gmail --> k-9 mail
google search--> ddg/startpage
I'm using orgzly instead of google keep right now but I can't say I'm happy with it and so far I haven't found a simple and convenient foss to do list app that doesn't require me to write markdown and stuff.
I also use the simple gallery. I have tried using open street map instead of gmaps, but I still use gmaps quite a lot. OSM is decent for navigation but doesn't have as detailed information about local shops and stores and stuff. I try to use newpipe instead of youtube and it has great features that you can't do with the regular youtube app but sometimes the latter one just works better. I also use the reddit app, wikipedia (is that not foss?), ebay, imdb, amazon, DB navigator and the spotify app. I use bitwarden as a FOSS password manager. IMO it's just not possible to go all FOSS unless you're happy with just the basic functionality of a smartphone (messages, calls, taking pictures, calendar) atm. Luckily, LineageOS gives you fine-grained control over what data these apps can access.
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
@BrodieRobertson Yeah, that's more like a matter of preference, I guess. What I was referring to is that you say you don't believe Vim makes a good IDE, with that, I have to disagree ;-) It makes a great IDE if you invest your time and effort into making it so. By itself, Vim is not an IDE at all, but that doesn't mean it doesn't let change this fact - just throw in a few plugins and there you go. And that's what's so powerful about Vim.
Also, ad the matter of preference - if you can't cope with something in other IDEs, you're basically screwed, but if you don't like something about Vim, you can simply change it to your liking.
Btw, here is my current Neovim setup: https://github.com/proxict/nvim-dotfiles
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
Hi, Brodie. I've been enjoying your videos for a long time now, especially how you differ from what I call the linux "cultists". How do you differ? A. You try to look at all sides of issues reasonably. B. You have some wit. C. You actually GASP do research on a topic BEFORE you make videos about it. So it is kind of too bad that my first comment here to you is a LITTLE negative. You speak of people of a certain age, and arthritis, re single or double clicking. I'm 70, been using linux for more than two decades, and here's the thing. Due to a little too much American football and baseball, I have quite serious arthritis, tendonitis, and bursitis in my hands (don't even ask about my shoulder, the list goes on and on, lol.). But even with these hands, at my "certain age", if someone tries to switch my default from double click to single click, I'm gonna put a 9mm in one of their kneecaps. Well, metaphorically, but you get the photo.
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
@thewhitefalcon8539 That's not how software works, and given that Wayland has features that Xorg doesn't have it's not very complete. At this point the wayland clients mostly are at parity, or slightly worse in most areas, there are areas where wayland supports something Xorg does, like HDR, though currently it's looking like the first compositor to have it(for the average person) is going to be kwin with KDE 6. And even if Xorg was "complete" with one of the biggest maintainers dropping out, it's not going to get the fixes that it needs to get to remain a real option. Which is why I want this to exist, while I have little to no interest in using anything X11 desktop myself, I want it for other users. And you yourself have mentioned a way in Xorg is not complete, there is no rootless way to summon a wayland window within Xorg. And security problems are also an issue when a project looses a lot of its maintainers, and problems with compatibility with the current linux stack. Software can't just not change without things breaking due to a changing user space, while the linux kernel itself doesn't change the userspace, all the surrounding programs do change, ever so slowly. Xorg isn't dead, and I don't see it truely being dead for maybe 5-10 years, but I'd say after around 2-3 years, it might be dead enough for the average user to find Xorg unusable, either with newer gpu drivers not supporting X11, or less testing done throughout the stack due to people moving to wayland, for the desktop software, DEs, and the general linux stack. While I'd like to see Xorg maintained, it's just not happening. We need a solution that actually fixes the problems in a way that is going to work, as in improving Xwayland rootless mode to work in the same way as Xorg, letting X11 desktops still live on with a more thourghly tested stack
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
@BrodieRobertson Would love these, as well! There's plenty of videos about installing Arch, but definitely not start-to-finish installs and configuration videos for various WMs and programs. Would be interesting to watch you go through all the Suckless offerings, as I feel those are the most intimidating but often recommended. My wishlist: BSPWM, Polybar, ST, Surf, DWM, Rofi, dmenu, pass (& rofi-pass), w3m, mpd/mopidy, ncmpcpp, and kitty.
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
I love, how Wayland finally seems to mature more and more and distros jumping on actively deprecating X11. *However*, there are still some issues with Wayland that can be annoying.
On my Office-PC (KDE 5.27, nVidia propriatary), I cannot right-click on my "taskbar". No context menu opens. This hurts my workflow so much that I switched back to X11.
On my home PC (KDE 5.28, AMD), it struggles in multi-monitor, multi-GPU mode: My primary monitor is fine (on RX 7900 XTX) while my secondary monitor (iGPU of X7800X3D). As soon as I move my cursor to the iGPU screen, this happens:
- Mouse cursor disappears, when on 2nd screen.
- FPS drop down to around 1-3
- AMDgpu driver spams some framebuffer issues in kernel log
Workaround: Disable the hardware cursor by using EnvVar KWIN_FORCE_SW_CURSOR
But that workaround kills re-draws of some effects like blur, whenever the cursor moves over it.
All together: Yes, I am a big fan of the Wayland implementation in KDE. But yet, there are some more showstoppers. Crossing fingers, most will be fixed in KDE 6.
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
I never used anything other than Linux at university, except during "Lab sessions" in the first two years where we were forced to use Windows. I'd say 95% of my programming and other studying was done on Linux (since that's all I had on my home laptop). Now I'm in work, I was given full freedom to install whatever on my system, as long as it met certain requirements (autolocking screen, encrypted disk etc). At my latest job, my laptop didn't even come with an OS, I was given a heads up and brought an install USB stick with me on my first day.
I spent quite some time trying to optimize developing for Android without Android Studio. Gradle and adb cli tools, uiautomater python plugin, various vim plugins, ranger, tiling window manger etc all help a lot.
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
@BrodieRobertson Since you're genuinely looking for feedback, I'll try to make the case for shorter videos.
First, do you yourself like to watch long videos that you feel can be cut down in half at times? The algorithm now promotes 8 minute videos, so although that was a good argument for your financial well-being, it's no longer applicable.
As a certain bald fuck put it, you got to put the meat first and get the viewer hooked early on. I automatically skip 30 to 2 minutes of the videos I decide to watch just because you're going to tell us about the use case before the actual thing. Which seems sensible but, in hindsight, very unsatisfactory. Let the thing speak for itself.
Consider doing 3 videos of 4 minutes of length instead of a single video of 12 minutes. It's not like there's a shortage of topics/FOSS. Just relating to torrents, I can think of webtorrent/webtorrent-cli, instant.io (P2P file sharing,) btfs (https://github.com/johang/btfs -- this is actually cool.) (Also, since you're a bloody weeb, try twist-dl [git clone, not npm install.])
You give the example of tig in the video and you're right. To properly look at something as big as a GUI for git, you need time. But man, all your videos are 12 minute long. At this point, I expect there to be unnecessary ranting no matter what the topic is.
By thought process and, to a larger extent, discovery process I meant when you want to show something but, first, you show the absence of it. Like running a command without a flag just to show us what magic the flag will do. (God help me when you do this with config files.)
You have to make viewing your videos a breeze not a chore. Don't make people feel impatient even if it means sacrificing a bit of quality for accessibility.
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
Replace Ctrl+P with FZF, it's the fastest. You can also incorporate FZF withing your terminal. Just look at this video to see how amazing it is: https://www.youtube.com/watch?v=qgG5Jhi_Els
gi goes to implementation, for example in case of interfaces in Java. I guess the braces are where the implementation starts in typescript (?).
To navigate back to where you came from after using gd, use Ctrl+O and Ctrl+I to move forward again. I think this is the way to traverse the vim-internal jumplist.
Also I find key-sequences (like when using the leader key) much more usable than all the Ctrl/Shift/Alt shortcuts. So I map some command to key-sequences like ,m (toggle nerdtree), which is basically the same as using space for leader key.
You need to try to use kj less, you basically scroll up/down with it, but there are more efficient ways to navigate, like M,H,L or Ctrl+u/Ctrl+d or w/b.
Also build yourself a dactyl keyboard so weird shortcuts won't bother you anymore.
Glad to see your channel growing. I really hope you won't incorporate some shit which doesn't belong here (or anywhere for that matter) like the distrotube guy ("help black people are mean to me") or luke with all his subtle hints...
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
If you want a funnier story, you could cover FreeBSD, code migrations. They moved from CVS, to SVN, to Mercurial, and I think currently to Git, all over the course of about 20 years. (I could be wrong about mercurial though...I think by that point I no longer had much reason to grab the code directly).
I know currently the view is "lol someone isn't using git?" but at the time this was happening things were a lot less clear cut, there were tradeoffs with a lot of them, and some projects had reasons for choosing certain systems. Most of the discussion was done in the open (for both FreeBSD and Mozilla at least, because I remember looking over the arguments back and forth) Personally I liked Mercurial a lot. The difference between grabbing a lot of code with mercurial vs SVN and CVS was really dramatic. It was also really easy to use, at least for what I was doing.
That period for open source stuff was really exciting. It's a bit of a shame that the kind of insanity from the big players basically died off. There's small projects still doing nutty things, but none are really shaking things up the same way. It might be worth a video diving into some of the history of the era, there were like arms race type competitions in basically every corner of open source for a while.
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
When micro USB was made to become EU standard, it took a bit for everything to change, because, of course, people were not throwing away their equipment to buy brand new ones with micro USB. So, by the time you could "safely" borrow a charger to charge your phone, USB C was becoming the new standard and those who needed to buy new devices might have gone for that, hoping their devices would last long enough to make it a good idea to follow the new standard, rather than the old official one.
Basically, I won't be surprised if very soon they will introduce a new connector to the market and make things messy again.
I also won't be surprised if most of what the EU does, even when is a good thing, is marketing and is intended to push you to buy something "new to you" (and already somehow "standard" in the market) at the time they're planning a new standard, so that you can't jump from micro USB to the next USB after C, but you have to buy yet another device every time they decide the market is prepared to accept a change.
I've seen enough changes "for the better" that were actually inspired by profits for the fat cats and I don't think this one is any different.
Having said that, I still consider having all the same standard, where safe, a good thing (as long as you don't use a low power rating extension to connect a high power charger and a high power device passing through a high power cable that you want to make longer to position your device where you need it).
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
I'll literally repost part of my original comment to explain the things you didn't bother to:
[...] It can also open specific "sections" based on what's in parenthesis next to the manual name. The most important one to know is that the "p" denotes the POSIX version of the manual. I often times find the POSIX manuals to be much more useful because they describe a lot more about the program, and the information they provide is more standardised since it comes from one source.
Keep in mind that newly installed programs with man pages tend not to appear right away because the man database hasn't been updated yet (though you can still just open the manual manually from the terminal, or if you know the exact name of the manual you can just type it into dmenu and it will open it, do keep in mind that dmenu will select whatever result is highlighted so if something is matched press SHIFT+ENTER to bypass that). Depending on the system, it may be updated with a cron job, during boot, etc. but if you want to update it manually, run:
sudo mandb
To understand what the numbers in parenthesis mean, read the initial part of the man page for "man-pages". And to get more on sections, type "intro" into this script. I wish I knew these things when I was starting out, instead I found it by just randomly typing "intro" into my own script. To this day I don't see anyone teaching how to properly read the fecking manual. [End of repost.]
So basically, there ARE cases where there are manuals with the same name, the most important examples are those marked with a "p" after the section number. This is the POSIX version of the manual. This also explains why awk does what it does in the script. Without it, you would not be able to open the POSIX version even if you selected it in dmenu, it would just open the normal one.
2
-
2
-
2
-
2
-
2
-
Almost four months into my graduate certificate program and I have noticed some disturbing realities about the program itself; the transition from in-class to remote learning only exposes these issues. The total lack of integration across the curriculum between courses, which are interrelated unlike say a typical college or university, results in lost opportunities to (1) reinforce how these courses fit together, (2) provide more realistic hands-on assignments and projects which could be leveraged across multiple courses, and (3) inspire and motivate students / learners struggling in some of the courses through no fault of their own. The reason I enrolled in this program is due to its uniqueness combining courses teaching subject matter expertise and courses teaching the information technology skills that are ultimately why most students might have chosen to attend. There are 6 courses this semester of which only 2 are being taught well. I learn a lot each week in those courses, while 3 other courses are a frustrating experience and the sixth courses is akin to lukewarm water - neither good not bad. I do not know if the other 4 instructors would be any different in a normal in-class teaching situation in which they are more comfortable.
The refusal of the education system to break out of the industrial-age model of teaching courses as separate discrete subjects is, in my opinion, hampering society. Does a carpentry apprentice learn about various tools in isolation or within the broader context of the workplace and workshop? No, it would be foolish to think building codes, tools, and projects are not highly integrated into most construction projects. The trades provide a better model and it is unfortunate this style has not been adopted by all educational institutions.
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
one thing that always annoys me is that people can't agree on what argument to use to show the help arguments. Here are some that I have seen:
-h -help --h --help -?
and I have even seen ones that expect it to be when you pass no arguments.
and sometimes I try to use something that should be the help menu, but it's something else, and it does something very unexpected.
also, people disagree on what should be included in the help menu. some show a short usage thing, while others essentially include their entire manpage, so I either have to scroll for a while, or run it again through less
same thing goes for arguments to show the version, so for example: python -v does not show the version, --version does
also, what is shown in the version message can vary a lot. some show just the version, while others (gcc for example) show a lot more
(in the case of gcc, -v and --version both are for showing the version, but they give entirely different outputs)
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
There is a number of reasons, which do make some economic sense.
1. YouTube is the biggest target audience, and pays the best. If you release on other platforms, your audience will become fractured among various platforms, whereas if you can keep them all on youtube, you will make more money, as youtube pays best.
2. The amount of exclusive users on alternative platforms is too small to justify the uploads. If your content is good enough, you most likely can convince alternate video host users to watch your videos on youtube - thus increasing your views/subscribers/algorithm engagement on youtube, and making more money.
Those are what I see as the two major reasons. There are others, such as not wanting to be associated with fringe or unorthodox content (very much the case with Bitchute, which has unfortunately has made a name for itself as primarily a host for edgy content) - if you are interested in maintaining a family-friendly image, then uploading to bitchute might be damaging to that image.
Don't get me wrong though, there are plenty of good reasons to use alternative platforms despite these issues, I just don't entirely think that at this stage the economic argument is the best arguement.
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
All you psychopaths (Linux folks included) ought to make your way back to OS/2 warp, lest you want to spend New Years in the lowest depths of the Ambassador Brig. All jokes aside, if I were likening software choice with 'psychopathy', I'd need one hell of a looking glass for self re-evaluation.
My bone to pick with Linux is package management. You have distro-repos, standalone packages like deb and RPM, appimages, Flatpack, I won't even mention Snap because it and Ubuntu deserve to rot in a fiery pit. Sandboxed standalone apps from the likes of appimage certainly streamline the process of getting programs working across multiple environments with little effort, but then present the whole "Why is galculator 300 mb?" question. Then of course, you run into dependency failures across the board, with libraries being incompatible with each other should you wish to use an older package of x or y. Or is the need to use specific versions of packages under circumstances also psychopathic?
I can agree that Proton/WINE give Windows a run for its money with respect to gaming. I applaud Valve and the Deck for bringing a non-Windows replacement to the mainstream.
-Psychopathic Windows NT 4 Workstation user
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
Hi there! I opened that Merge Request to bump Hyprland to 0.32.3. It's funny: I've originally found out about Hyprland with one of your videos; now I've contributed a feature and helped upgrade the package, so I feel like it has come full circle with this video :)
Addressing the problem you raised about packages getting out of date and maintainers going missing: when I originally set out to upgrade this package, MRs were still not publicly available, so I had to shoot an e-mail to the maintainer, like you mentioned. Long story short, I don't think they ever received it, because I didn't get any response. However, a few weeks later, an ArchLinux Admin let me know that MRs were not longer exclusive to Arch staff, so I opened one and, sure enough, it didn't take long for it to get merged.
I do think this situation was an exception to the rule: the maintainer was busy right around the time changes were made to the build system; even if the changes were for the better, the maintainer probably didn't have time to double-check them.
The good news is that, now that the Arch Repos are hosted on GitLab and MRs are available to everyone, packages are less likely to be kept outdated for too long. This workflow feels very natural to developers (both in OS and proprietary work), which in turn makes it more likely that people will try to contribute. At least, that's how I feel: it took me very long to even start working on updating this package because I did not know the workflow that was going to follow it. Now, however, I will always be looking forward to updating the packages I use the most.
Additionally, this workflow also makes it so that package discussions are no longer primarily held in e-mails or IRC chats, but, instead, in the repositories themselves. This also means that, if a maintainer was to be missing, it's easy enough to ping an Admin to get things resolved. This is important, because you mentioned that maybe some sort of automation could be done, but I feel like the problem was with the process, and not the people.
Great video! Cheers!
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1