Youtube hearted comments of p11 (@porky1118).
-
19
-
17
-
15
-
12
-
9
-
8
-
7
-
1:20 That's something I recently started to do more organized.
I always wrote down my ideas, but only in order to go into detail. First I only used paper, then I used text documents, but only recently I started to write down some of my ideas in a central place, which is synchronize using git. I'm mostly talking about stories. Often I just write down the idea itself, sometimes just one title line, often also a bit of context. Basically everything I have in mind mostly unfiltered.
Some time ago, when I had an idea, I tried to create something out of it. Instead of just writing down the idea, I strated writing a story. And I thought about an introduction. I wrote the introduction, and sometimes I just wrote an introduction without even arriving at my core idea. Sometimes the intro even went somewhere else.
6
-
2:00 That's basically how I'm writing now. I get an idea, I only write down what I have in mind, I'm not required to finish this scene yet. This really helps me to get started, especially since I sorted my scenes, so I know which scenes are unfinished.
Sometimes it's just a single line of text, sometimes it's a few key points, sometimes it's a pretty long key point list describing the scene very detailed, just without caring about the exact wording or if something necessary is missing between the key points.
Sometimes I write down the first half of a scene, or the core part of a scene which still needs some proper introduction and ending, sometimes I get more ideas during writing and I write everything down completely.
When I'm not really inspired to write, I can still finish some of these drafts by converting key points into real text, which is often pretty straightforward the more detailed it is, or just rereading one of the basically finished scenes and maybe adding some introduction, fixing typos, adding important lines inbetween, or improving some sections I don't like anymore.
5
-
4
-
4
-
4
-
3
-
3
-
9:55 When I was new to Rust, I didn't fear using references, even if they had lifetime annotations. Now, years later, I try to avoid explicit lifetimes alltogether and only use them if there is no chance to avoid them.
But there are a few cases where I really want lifetime annotations, especially in structs. Normally when I want some kind of accessor to some data, which should not be modified or go out of scope as long as the accessor exists.
For example I used this to call some function safely. Calling the function is only safe if I specify a valid parameter. So instead of calling the function directly, I created a struct, which lists all valid parameters, and in order to call it, I only specify the index of the parameter I want to call.
But iterators are also a good example for this.
3
-
3
-
@NoBoilerplate
After stopping using Lisp, I switched between Rust and Scopes for some time.
I've been the only user for some time besides the developer.
I ended up with only using Rust for now though, mostly because it's cleaner (like orphan rules; in Scopes I can just import a module which overwrites some of my methods without letting me know. I don't feel safe anymore).
Scopes is still the most powerful language, I think. It has all the features of major programming languages, most interesting for you, it's S-Expression based, has Lisp macros, and a Rust borrow checker (version 2, only one level of borrowed data, no structs or tuples containing borrowed data).
It's even more low level than C (by default values are not even on the stack but in the registers and immutable).
And more high level than Lisps. Even a new kind of macro, which has access to type information and is able to do partial evaluation.
It's the only language, where I was able to write a compile time dimension and type generic automatically typed geometric algebra library with versor tracking.
And probably the only language where this is even possible theoretically, besides C++ maybe.
It's difficult to keep it short when talking/writing about Scopes ;)
3
-
3
-
3
-
3
-
3
-
3
-
3
-
2
-
2
-
2
-
2
-
9:00 One important thing, which helps me to go to bed earlier, is having a laptop.
It seems contraproductive at first, but when I didn't have a proper laptop for some time, I knew, I would not be able to do anything on my device, when I turn it off.
So I knew, I had to watch this video now, or finish writing this segment, or make this program work, or whatever I wanted to do, because I wouldn't be able to until tomorrow.
And since I want to get enough sleep, I didn't expect to have time in the next morning before work to finish it.
So when having a laptop, especially one I can set to standby mode over night, makes it easy to lay it away, so I think, even if I can't sleep, I will be able to do something instead. And when I already lie in bed and put my laptop away, because I'm getting tired, I can just start to sleep. I don't have to get into bed, maybe also get dressed or whatever, which makes me awake again.
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2:44 No, that's not true. It doesn't matter too much if my data is stored in plain text or in a binary format. If the data is complicated, it's always better to use a tool.
And as long as you know the structure of a file, you can write new programs to use the same binary format. I've done that.
I've written a project called pns (published on gitlab), for simulating petri nets, which uses a binary format. I already had created a similar tool before, which was already more powerful. So I used the old tool to create the petri net, saved it using the binary format, and then loaded it from the binary format into the new program.
If you know the binary format, it's much easier to create new tools to read and write the data. No need to write complicated parsers yourself or use serialization 4lugins.
4:22 Okay, it seems to be about text heavy data. In that I agree. These should always be stored in a non-binary format (markdown, latex, html, etc.).
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
@NoBoilerplate
>better ... in some areas
Feature-wise it's better in almost all areas.
It just isn't as clean as Rust, no focus on safety, etc. (also no orphan rules)
The main reason, I don't use it anymore.
>I note there's no macros
There are macros in Scopes.
Three different kinds of macros even.
The documentation is probably still missing a lot.
inline, which is a function partially evaluated at compile time, more like templates than inline functions
sugar, which is like lisp macros (transform one expression into another one, just syntactic sugar)
spice, which has access to type information and evaluated constants at compile time.
It's the only langugae where I was able to implement geometric algebra, generic (at compile time) over dimensions and types.
It also uses llvm vectors internally, important for addition and is able to track if a a multi vector is a versor at compile time for further optimizations.
2
-
2
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1