Youtube comments of ⃠ (@U20E0).
-
361
-
313
-
193
-
191
-
167
-
123
-
109
-
80
-
78
-
69
-
68
-
65
-
48
-
45
-
44
-
42
-
39
-
39
-
37
-
37
-
37
-
37
-
36
-
32
-
31
-
28
-
27
-
26
-
26
-
26
-
26
-
25
-
24
-
24
-
24
-
24
-
24
-
23
-
22
-
20
-
20
-
20
-
19
-
19
-
17
-
17
-
17
-
16
-
16
-
16
-
15
-
15
-
14
-
14
-
14
-
14
-
13
-
13
-
13
-
13
-
13
-
13
-
13
-
12
-
12
-
11
-
11
-
10
-
10
-
10
-
10
-
10
-
10
-
9
-
9
-
9
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
8
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
7
-
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
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
( do not quote me on this )
In the US, two-phase 240V AC is used.
This means that each house gets two 120V AC lines, which are 180° out of sync.
Between those two lines you have 240 AC, and between either line and nil you have 120V AC.
Outlets will use one line and nil, but with different wiring you can get a 240V circuit for ovens and such, and those use an outlet/plug with one L-shaped prong, and the other prongs at an angle.
As for why, mostly historical reasons. Changing the grid to anything else is practically impossible due to it’s size.
In Europe, three-phase 400V AC is used, with three 230V lines 120° out of phase.
These are split such that one household gets one 230V line ( you can get a direct 3φ400V connection too, but a regular household does not need that )
Both use three-phase in the grid itself for longer range transportation ( iirc, but that makes sense, 3φ is very efficient for that ), but at much higher voltages.
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
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
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
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
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
a pointer is literally just 4 bytes ( 8 on 64 bit ). It is not expensive at all, and since it is on the stack you don’t need to allocate anything )
You probably got confused. Allocating the space for data on the heap ( what malloc() does ) is expensive, but, you don’t have other options, if you allocate too many large structures on the stack, you will eventually run out of space in the stack ( which is significantly smaller than the heap ) , and get a segmentation fault.
Plus, the more important thing about the stack:
the stack holds “local” variables. If you allocate something on the stack, that data will become invalid as soon as you leave the function where you allocated it.
As for the rest of your comment, i am not sure i understood it correctly, could you elaborate?
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
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
@pmberkeley not what I think about it, we are arguing about two different perceptions of the same phenomenon. You need small particles to properly taste them, and metals like to collect into very solid crystals. So you need the metal to either be finely ground to create some free particles, or you need it to be a part of a smaller molecule, which mostly happens in salts, where the metal is an ion. Or you need electrical current to rip out some small quantity of particles, in which case, there will also be electricity involved. Metal connected to arguably small-ish molecules also happens in blood, but blood has a lot of salt that partially covers the iron taste with its own sodium. All of this may or may not result in the taste being associated with salts, or if you can also tell the taste of the other component of the salt and/or have put wires in your mouth for no reason — with electric charge; instead of metals.
Don’t know about you, but for me this was a very informative conversation, especially this conclusion.
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
-
1
-
( the backslashes are here to stop YouTube from consuming asterisks an making random text bold )
it is just a pointer pointer, that is useful when you want an array of strings for example
( that would be “const char** A” or “const char \*const\* A”, the latter is if you want the entire array to be constant, not just the strings themselves )
As for char \*x and char\* x, those are identical, the latter is considered to be better, because of things like “int x, *y”
(char*) x may not compile because it might be interpreted as a cast, but i’m unsure of that.
char (*x) will compile, parentheses can be used to indicate priority, which is needed in a function pointer declaration:
int *f(int x); is a function returning an int*, but
int (*f)(int x); is a pointer to a function returning an int, the parentheses tell the compiler that the asterisk applies to ‘f’ itself, and not to it’s return type.
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
Physics being relative to perspective is confusing as always. “observe” is just a convenient word to use.
Example system: you have some things, A B, and C,
which have numbers, and can interact by averaging their numbers.
A has a superposition of 2 and 4 as it’s number, B has 7, C has 9.
If B interacts with A, from the perspective of B the following happens:
A collapses to 4 and both become 5.5
But from the perspective of C the particles become entangled:
A remains in superposition, and B becomes a superposition of 5.5 and 4.5.
After all, how can C know that A became 4 and thus B 5.5 without observing the result?
The superposition only collapses from your perspective when you “enter” it by observing it, so from your perspective your observation and your observation only can collapse a superposition.
And it only gets more confusing from here by my understanding:
If C were to now interact with B, from B’s perspective:
B itself is 5.5, C is 9, so both become 7.25.
From C’s perspective, B can collapse to 5.5 ( and A to 4 ), making both B and C 7.25.
OR
B can collapse to 4.5 ( and A to 2 ), making both B and C 6.75.
In the latter case C now exists in a different timeline. But that in not detectable in any way because everything makes perfect sense from all perspectives, other than that of the omnipresent reader of this comment
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
-
( warning: length )
I found 2 interesting ones:
You can attempt to modify a string literal in C:
“abcdefg”[1] = ‘x’;
you can deference a void pointer in C, and cast the result of that to void.
And an obvious one:
C allows implicit casting of void* to any integer type, C++ requires an explicit cast
And a few practical ones:
In C++, only one “member” of a union can be “active” at a time, which imp defeats the purpose of a union.
this:
union u
{
float f,
int i
};
u.i = 3;
return u.f;
is valid C but undefined behavior in C++
C allows designed initialization in any order, does not require you to initialize any member, and allows initialization of subobjects and array members.
You can redeclare a global variable in C
global variables always have external linkage in C unless else specified. In C++ constant globals are internal unless else specified.
C allows you to buffer overflow yourself in dumb ways:
char c[2] = “abcdefgh”;
The size of a character literal is that of an int in C, but that of a char in C++
And a few less practical ones:
You can call main() in C
You can define main() with any return type in C.
in C, for a function declared as
int function() {}
calling it with arguments will not result in an error, unlike with
int function(void) {}
you can declare a constant without defining it in C
And a few obscure ones:
C allows a type definition anywhere where it allows a type:
void f ( struct s { int x } s );
C++ would give an error.
you can use prefix ++ and — on a bool in C
you can goto over a declaration in C
given struct S { int i : 1; } S;
sizeof( 1, s.i ); is valid C but invalid C++
Someone else can explain why
you can declare something as auto without defining it in C
you can put way too many const modifiers on a type in C.
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
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
@LowestofheDead fuck. After indecisively editing my comment for like 30 minutes trying to make it not sound offensive &c i just confused myself and made highly ambiguous nonsense.
( also i use Focus as a synonym for Comment, as is sometimes done )
(( also long warning ))
((( also I can’t see your comment or my comment or your other comment or most of what i’m writing because mobile, so expect issues caused by that )))
What i meant, is that you can’t logically have a conflation of Subject and Topic.
in “the dog will not eat that”. SVO word order overrides T-F order, so which is which is unclear ( although it’s more likely to be S )
in “that, the dog will not eat”, T-F order overrides SVO word order, so “that” is clearly the Topic and the rest the Focus, and since can’t have a Subject standing alone in English, “that” must also be Object.
You can’t really argue here that Subject is a conflation with Topic when they are clearly separate here.
As for why i said what i originally said, i failed to find a suitable expression for “you can’t put Topic into a word order” ( as it is most often an abstract thing, not a clause or word )
You also can’t put topic in a word order because that’d imply that none of SVO can be T since T is separate
(((( Note that linguistic terminology is as bad as terminology can get, so we may be using terms with different definitions in mind ))))
“Would the concept of Subject/Object be necessary if we had not invented them ( but still had, […]?
Consider this:
“i see the dog”
- Agent: “I”
- Patient: “the dog”
- Topic: could well be “the dog”
- Focus: could well be “is seen by me”
“the dog is seen by me”
- Agent: “I”
- Patient: “the dog”
- Topic: probably “the dog”
- Focus: probably “is seen by me”
You can’t distinguish the two using Agent/Patient alone. Topic/Focus are useless for describing a sentence out of context because they are defined by context.
And yet they may have completely different meanings.
You need to have Subject and Object to accurately distinguish these.
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
-
The reason is quite stupid:
int* x, y;
looks like it's defining two int*s, but y is actually not a pointer. To declare two pointers you would use
int* x, *y;
int *x;
int *x, y;
makes that "feature" more obvious
In my opinion the better solution is to just never use multiple declaration except for coordinates like char r, g, b; or float x, y, z;
The community seems pretty split on this, and ultimately it's just aesthetic preference:
const int *const *x;
int const *const *x;
const int* const* x;
int const* const* x;
To me the last looks the best since it's the most consistent-looking, but (very) many would disagree.
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
-
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
-
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
-
@s1r_m3ga84 it’s exactly as i was suspecting:
that’s not a bone… that’s either cartilage ( which can be hard, soft, yellowish white, blueish white, and anything in between; and is found at the end of a bone, or in some cases in place of a bone )
or a tendon ( which is grayish, connects bone to muscle, and can be found running parallel to a bone ).
A few fish do actually have “skeletons” made completely of cartilage ( sharks, rays, skates, sawfish, chimaeras )
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
-
1
-
1
-
1
-
1
-
@caschque7242
Making such a study would be very difficult, since you can't just get analytics about old versions of search engines unless you anticipated something like this and collected them beforehand. I wonder if someone has actually done it.
I don't see people blaming AI for everything (here anyway), and that is not what i was trying to do. I should have clarified that i was trying (with questionable success) to explain why the reaction is what it is, not to agree with it
( My personal opinion is that a: i don't care about this because i don't use Google; b: this is too recent of an event to form concrete opinions it; and c*)
What i see is two things:
1: people finding all possible arguments against everything Google does because people dislike Google; and
2: people being sick of AI getting shoved absolutely everywhere, including where it has no place.
Of course it requires time for companies to figure out how (and if even) to apply a technology this new, but that's not what most companies are doing.
That's not what most companies are seeking. Most companies are focusing on the "shiny new thing" aspect of AI.
This argument doesn't apply here, since this is (potentially) an actual good use case, but most times AI is introduced somewhere it ends up being to the product's detriment, and so that is the default assumed.
I also appreciate you appreciating an argument you don't agree with, that's rare on the internet nowadays.
*c:
The person who accepts the AI answer would also accept the first thing on the page, which is just as likely to be garbage.
The person who doesn't accept or accepts it after more research remains unaffected except being slightly annoyed.
1
-
1
-
@caschque7242 Update:
My opinion has shifted, and yours should too.
This turned out to be a disaster.
My justification for saying that this doesn't really affect things was that people who will just accept it would also just accept the first thing they click, and the quality would be comparable, since it just summarises the content of the results..... or so i (and i assume you) thought.
After seeing this thing a bit more in action i think this should be removed immediately until Google fixes it.
It cannot analyse the context of information.
It has no restriction on when to activate.
It has no restriction on where to pull information from.
It does the standard AI thing of trying to not infringe copyright by mixing things up a bit.
It does not verity any information at all, especially that embedded into it from training.
When you combine these factors together you get a disaster.
It will readily take information from a couple of good sources, then blend it together, and add some more information from complete garbage places or its own knowledge which it assumes to be completely true.
One of the places it can pull from, a place also contributing significantly to it's training data, is social media. A place known for information that should not be taken out of context (jokes, troll responses, etc). It takes this information out of context and embeds it into the responses, often without saying it did so.
This is worse than misinformation sites, because misinformation sites are usually pretty obviously misinformation if you don't have confirmation bias to it, and are often written in a way that doesn't sound particularly authentic.
this on the other hand takes the misinformation (or incorrectly interpreted information) and puts it in completely well-formed and formatted language next to a bunch of good information, and it also can seem way more credible to people who don't know what's happening, even though it isn't.
Then even worse, it triggers on questions that absolutely should not get answers from experimental technology, like medical and financial advice.
This isn't just theory either. Some of the things it is saying could lead Google into legal issues. It has literally told people to *** themselves when they asked what they should do as they are feeling depressed.
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
-
(this is my opinion and is probably not entirely correct)
It's more about the message than the actual behaviour of the code ( unless the struct is huge )
When passing by value, like you would a number, you are conveying that the data is indeed like a number.
It is short-lived and not in and of itself tied to anything else, you can copy it freely. You combine it with some other data, look at the result and then throw everything away.
Or: you are passing it by value, like you would a pointer, and the data is indeed like a pointer ( ie it's a handle )
It just identifies the actual data you want to work on, which is stored elsewhere (and maybe some metadata), but is in and of itself useless, and you can still copy it freely.
You should also name your structure appropriately if you intend to mean the latter.
But A also somewhat applies. You can't pass a structure by value if it is too big the compiler will convert it to "pass by reference and then copy"
C also somewhat applies to large structs.
If you write is as pass-by reference, you can now choose whether to copy or not individually every time you use it. This might also make the function more flexible and contribute to D.
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
-
@Goryus i think you missed a very important word in my first comment:
"theoretically" ( which should have actually been "hypothetically" but whatever )
This won't ever happen, especially not unintentionally. And maybe it can't, but none of this conversation actually matters anyway:
You highlighted the actual issue with the comment on for loops. All processes are equally deterministic, the entire universe is just an infinite loop with some equations and updating a state, with some randomness thrown in. Is the universe conscious? Does that question even make sense?
Consciousness is a completely arbitrary philosophical concept, and there really isn't any reason to even mention it when talking about AI, although we will, because discussing subjective things is what makes humans human.
I personally believe the duck method is most reasonable:
If it walks like a duck and if quacks like a duck, then it is a duck.
If it acts like it acts like it has opinions, then it is conscious ( and current LLMs generally don't, although i think the only reason is training methods )
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1