00:00:08 <int-e> salpynx: And that would potentially allow total languages to be TC. Of course we know that this doesn't actually work, but it requires some theory.
00:00:23 <int-e> Not just definitions.
00:00:40 <korvo> int-e: Indeed, without any further constraints, Set + 1 ≈ Pfn. This is because Pfn is defined over *all* partial functions, not just some computable universe.
00:01:29 <Raoof> korvo the main reason is that I'm bored and as I said I have a hunch that it is possible not in the finite case but in the infinite case
00:02:02 <korvo> Raoof: Understandable. I'm afraid your hunch is wrong but it certainly is for an interesting reason.
00:02:35 <Raoof> salpynx you can think of Ar as unrolled BA
00:03:07 <korvo> Like, if nothing else, I think http://tac.mta.ca/tac/reprints/articles/15/tr15.pdf is quite interesting. If you have a category-theory allergy, https://arxiv.org/abs/math/0305282 is delightful and has many more examples.
00:03:20 -!- mtm has quit (Ping timeout: 244 seconds).
00:05:46 <salpynx> Raoof: well, if you are simulating BA's main loop with an infinite repetition of the code, and all other operations and abilities are equal, I'd consider that equivalent, and therefore TC (based on trust of BA)
00:06:12 <Raoof> korvo I've looked at those links before they all comes down to d(n) = 1+u(n,n) and that is just not a problem in Ar or Ar2 or Ar3
00:06:31 -!- mtm has joined.
00:08:35 <korvo> Raoof: That's not the issue; the issue is the transformations without fixed points. For example, Ar2 features negation, and negation has no fixed point.
00:09:15 <korvo> Like, looking again at n and 2 ** n, the problem is not n at all! The problem is 2, and specifically the transformation 2 → 2 which has no fixed point because it sends true to false and vice versa.
00:09:23 <salpynx> int-e: I guess "basically" was doing some extra work in my "by definition" claim :) The proofs and wisdom we have make Total and TC languages safely mutually exclusive, I can't imagine someone finding some example that is both.
00:09:50 -!- X-Scale has joined.
00:11:07 -!- Lord_of_Life_ has joined.
00:12:45 -!- Lord_of_Life has quit (Ping timeout: 276 seconds).
00:12:45 -!- Lord_of_Life_ has changed nick to Lord_of_Life.
00:17:16 <Raoof> korvo I didn't get what you mean and maybe you didn't get what I mean
00:18:47 <salpynx> i;m not following some of the function and diagolisation arguments, I'm just trying to get TC claims straight
00:20:27 <Raoof> salpynx it's just a claim I don't have any proof [yet]
00:23:41 <Raoof> korvo can you explain your understanding of diagoanlization and fixed point a little more ? maybe it will resolve my confusion
00:29:16 <b_jonas> Raoof: the diagonalization argument is similar to last time near https://logs.esolangs.org/libera-esolangs/2024-07-02.html#lRc , it's just more complicated here because you are composing functions with an unlimited number of arguments
00:30:46 <Raoof> salpynx my goal with Ar and Ar2 and ... is to use it as a practical language and a total model of computation so if I can show that you can program very easily using it then I can confidently say it is TC
00:32:58 <korvo> Raoof: I'm basically with Lawvere. In a Cartesian-closed category (a simply-typed lambda calculus), we can write a combinator that looks like (lambda (f x) (f x x)).
00:33:08 <Raoof> b_jonas you weren't convinced then that diagonalization does not work for Ar ? Ar is just unrolled BC
00:38:30 <salpynx> As I understand things, Ar is either total or TC, it can't be both (maybe I need proofs here). If it is TC, it's not because of inc, sub, mul, div, rather it's the ability to simulate a loop with infintely repeated code (which I don't see in the spec but if it is part of the construction it should work), and having enough locations to store and manipulate data to make decisions.
00:38:43 <salpynx> 'unbounded program length' is a pretty big advantage, and I think that uncontroversially gives you a loop equivalent, which could make something otherwise not TC, TC (this is literally the Z3 glue-the-program-tape-together trick). I don't think that's too controversial, if we're allowing that sort of thing.
00:40:13 <Raoof> korvo I'm with https://kar.kent.ac.uk/88974/1/turner.pdf_nocoversheet : "A non-recursive function is logically possible—because Church’s Thesis might be false
00:40:29 <salpynx> I'm inclined to believe that if Ar really is unrolled BC, then it's TC, but that's not a controversial result at all.
00:42:11 <korvo> Raoof: Lawvere's setup is constructive. Choose the category and an object O which will play the role of 2, and we can look at the arrows O → O to see whether all of them have no fixed points.
00:42:59 <korvo> Yes, there *might* be ways which go around it, but you
00:43:06 <korvo> *you're using Z, which obeys these rules.
00:46:09 -!- Guest15 has joined.
00:47:16 <Guest15> i want to change my username on the wiki, is there anyway to do so?
00:47:40 <Guest15> if this is not the correct place.person to ask anyone know the correct palce/person?
00:48:14 <esolangs> [[Subleq]] M https://esolangs.org/w/index.php?diff=134240&oldid=124660 * TheCanon2 * (+91) /* External resources */ All of Mazonka's links are dead
00:49:11 <Raoof> korvo I'm not super familiar with category theory, can you explain diagonalization like I'm five :D ?
00:50:27 -!- X-Scale has quit (Ping timeout: 256 seconds).
00:51:18 <Raoof> salpynx people will argue that because all Ar programs are bounded by a polynomial function no Ar program can compute 2**n or ack function
00:53:34 <salpynx> Raoof: BA has 6 registers, does Ar have registers or equivalent data storage?
00:55:10 <Raoof> korvo of course I can learn category theory and read the papers with a lot more attention myself but I want to read your explanation
00:56:29 <Raoof> salpynx you can think of function arguments as input registers and functions values as output registers
00:56:38 <korvo> Raoof: The second paper I linked, Yanofsky 2005, avoids category theory and explains it using sets. If you're already doubting Cantor's theorem, though, then that's not going to help.
00:58:00 <korvo> The thing is that, right now, you don't have *evidence* against these classic theorems, just a hunch. And so Newton's flaming laser sword, or even Hitch's razor, says that your position is bogus.
00:58:39 <korvo> So I could explain all day and night using every metaphor I've ever heard, and it still might not help you.
01:07:10 <Raoof> korvo I think theorem 2 of the second paper is true but that does not contradict my claim that there is a total and turing complete language
01:07:53 <salpynx> Raoof: in Ar2.py I think you're missing a i(a, b) def
01:10:24 <Sgeo> Is there a changelog for uxn?
01:11:05 <salpynx> yeah, I'm willing to believe that maybe Ar is TC with some infinite program trick, but then it's not going to be Total, so I also doubt your overall claim. I see it as either TC or not, and either way no existing theorems or proofs are shattered.
01:15:33 <korvo> Raoof: I agree. So, let's focus on that claim, because it does fall between the cracks, definitionally, and it doesn't appear related to diagonalization. So, first, what *is* a total language or a TC language? I could guess, but I can't think of any guesses which satisfy both.
01:15:59 <korvo> We're not talking about the language, but the *programs* written in the language, yes? The language just gives us codes for those programs.
01:16:02 <salpynx> sgeo: i think this is the official repo for uxn: https://git.sr.ht/~rabbits/uxn (it was a little hard to find)
01:16:46 <salpynx> commit log: https://git.sr.ht/~rabbits/uxn/log
01:16:58 <Raoof> salpynx that is not important I created Ar2.py online and I didn't copy that function from my computer
01:17:31 <korvo> So, "total" means that every program has a well-defined mapping from inputs to outputs, and "TC" means that there is a program which implements e.g. a universal TM.
01:17:33 -!- Noisytoot has quit (Remote host closed the connection).
01:18:41 <Raoof> korvo by TC I meant that it can compute all total computable functions
01:19:37 -!- Noisytoot has joined.
01:19:51 <Sgeo> salpynx, I don't see the spec in there. I'm interested because I saw some old stuff talking about instructions that trap, but the current spec never traps
01:19:55 <Sgeo> except brk maybe
01:19:57 <korvo> Raoof: Okay. Then this is very easy: Halting is not computable, so the total function TM → 2 which tells us whether each TM halts is not computable either.
01:21:00 <Raoof> korvo yes we are talking about the set of programs that you can write using only 4 or 1 or any other number of total functions
01:21:25 <korvo> It's not really possible to run away from this because of what I mentioned earlier: your TC language must have a program which universally interprets TMs, and we can feed never-halting TMs into that program to make it run forever.
01:22:06 <Raoof> korvo halting is not a total computable function
01:22:36 <korvo> Raoof: Okay, so "total" means something more limited too.
01:22:42 <salpynx> Sgeo: you've probably found https://100r.co/site/uxn.html and https://wiki.xxiivv.com/site/uxntal_reference.html .. I don't really know uxn, but it has been on my radar for a long time
01:24:13 <Raoof> korvo why do you mean by "more limited" ?
01:24:45 <korvo> Raoof: I mean that there are plenty of total non-computable functions, so "total language" must mean something that is much more limited.
01:26:04 <zzo38> I do know uxn, and I had written an implementation
01:26:22 <korvo> "Turing-complete" means "complete for TMs"; it means that any task a TM can do, your language can do. But that means that, like a TM, your language must be able to run forever, so it can't be total.
01:26:36 <korvo> The only way around this is to automatically detect the TMs that run forever, which is Halting.
01:26:44 <zzo38> There is also a IRC channel #uxn on Libera for discussing uxn, too.
01:27:49 <salpynx> korvo: I think you are saying what I was trying to say, but better. I'll stay silent on this and just listen for now.
01:28:40 <korvo> salpynx: Honestly, I feel like you and ais523 already put it best, and I worry that I'm like Mojo Jojo, repeating myself endlessly and unhelpfully.
01:29:23 <zzo38> I do not know of any change log for the official specification of uxn/varvara, although I wrote my own documentation which I intended to match the official specifications, and those documents are in a version control so it might be possible to look at that for some idea of a change log of the specification, although I cannot guarantee its accuracy or completeness, and it does not go all the way back.
01:29:52 <Raoof> korvo if you have a total and turing complete language then you don't need to worry about the halting problem because everything you write always halt so I don't understand why you are interested in non-computable functions ?
01:31:36 <korvo> Raoof: What does "Turing-complete" mean here?
01:32:11 <Raoof> korvo being able to compute all total computable functions
01:33:27 <korvo> Raoof: Okay. Without loss of generality, those computations are going to be given to you as Turing machines. (You can pick *any other* Turing-complete language, of course. That's why TC is such a powerful concept.)
01:35:32 <Raoof> korvo why do you want to speak TM/gibberish when you can speak Ar/English
01:35:44 <korvo> Raoof: You haven't proven that Ar is Turing-complete, so you can't pick it.
01:36:37 <korvo> Computer scientists usually use this in the other direction: build a language, implement an emulator for something Turing-complete, *then* claim that the language is also Turing-complete because the emulator is faithful.
01:43:26 <korvo> Raoof: There *is* a natural Turing-complete problem which is adjacent to Ar, but it can't be used to save your hope of a total-and-TC language. Ar implements Diophantine equations for *one* parameter, the inputs. Searching for *all* parameters is TC.
01:45:29 -!- X-Scale has joined.
01:48:38 -!- amby has quit (Quit: so long suckers! i rev up my motorcylce and create a huge cloud of smoke. when the cloud dissipates im lying completely dead on the pavement).
01:58:59 <salpynx> Not sure how helpful this is, but this is my attempt at a infinte loop counter in Ar (as I understand it), with notation to specify a left and right infinte string: "inc("<sup>ω</sup> . "0" . ")"<sup>ω</sup>
02:01:02 <salpynx> I think that's a trivial example of non-Totalness.
02:03:26 <Raoof> salpynx what do you mean by "example of non-Totalness" Ar is a total language. and I didn't get your infinite loop counter can you clarify?
02:03:26 <salpynx> Like I mentioned the other day, I'm using <sup>ω</sup> to mean repetions of the string by the first infinite ordinal number.
02:04:37 <salpynx> it's a program that doesn't terminate. It's infinte copies of "inc(" followed by a "0", then terminated by infinite copies of ")"
02:06:20 <salpynx> that will increment for ever. At first I thought you were unrolling BA by allowing "<contents of loop program" x infinitely repeated, but it looks like there is no external storage, so everything has to be nested
02:07:44 <salpynx> Raoof: from the Ar readme "you can see that in the limit (unbounded program length) these four functions are turing complete", this is what I think I'm allowed to do with it
02:09:50 <Raoof> salpynx I meant that if you allow unbounded program length then obviously any total computable function is easily computable in Ar you just write the definition of the function in semi-English
02:11:04 <salpynx> is that what you meant by "unbounded program length"? I think allowing left infinite strings is weirder than just right infinite strings, but it's all a bit weird. I'm exploring infitem omega-words in other projects, so I'm kinda interested in infinite strings.
02:14:23 <salpynx> It's possible I have misunderstood you, but I took 'unrolled BA' and this "unbounded program length" mean program input could be a infinitely long string of source code.
02:15:07 <Raoof> salpynx my goal with Ar,Ar2,Ar3,... is to came up with a total model of computation that is also a practical language for programming and teaching not as an esoteric language I thought this community maybe interested in a total and TC language.
02:15:49 <Raoof> salpynx program length can be infinitely long but that turn Ar into an esoteric language not a practical language
02:17:14 <salpynx> Well, I think the _only_ way to make Ar TC, in the absence of loops is by allowing infinite source. I thought that is the argument you were making in comparing it to BA.
02:18:46 <salpynx> BA has a handful of registers, and a loop (implicit or specific? there is a loop at any rate), those are quite important for its TCness. I don't think the 4 functions are that relevant to TCness.
02:19:29 <salpynx> There was some very misleading stuff in the Z3 article which I think I've tidied a bit (that was my main concern)
02:24:26 <salpynx> "total model of computation that is also a practical language for programming and teaching" seems like a worthwile goal, but I think you'll have trouble demonstrating a language that is really both total and TC. I think you'll find it is one or the other.
02:26:47 <Raoof> salpynx I haven't studied BA that much but at first glance it's like a "while True: f(n)" where f is a total function using only add,sub,mul,div that can crash when division by zero occurs. if you unroll the loop you get iterations of f(n) using only four operation. now comes the question: is there an UNrollable function ?
02:28:18 <Raoof> salpynx *is there an UNrollable but computable function ?
02:42:32 <salpynx> Not sure i understand the question sorry. The only way I can think of 'unrolling' the function with the tools you have provided in Ar is by repeating source code infinitely like I did above. That's not going to contradict anything krovo said tho.
02:48:06 <salpynx> my apologies for being too esoteric, I took "it's TC because of infinite source code" at face value, and it made sense to me, I'm laughing now because I wrote infinitely long code, probably based on a misunderstanding (mine)
02:48:59 <Raoof> salpynx what about Ar2 and Ar3 ?
02:53:48 <Raoof> salpynx I meant to say can you write ackermann function with Ar2 or came up with a way to loop or iterate easily in Ar2 or Ar3 ?
02:58:05 <Raoof> salpynx I haven't said that there is no need for apology I didn't thought your comment was too esoteric
03:01:55 -!- X-Scale has quit (Ping timeout: 256 seconds).
03:25:23 <salpynx> well, ack = lambda m, n: inc(n) if isNonZero(m) else ack(dec(m), 1) if isNonZero(n) else ack(dec(m), ack(m, dec(n))) works, but uses Python `if` and recursion
03:26:26 <salpynx> ack = lambda m, n: If(m, inc(n), If(n, ack(dec(m), 1), ack(dec(m), ack(m, dec(n))))) runs into recursion problems using the current Ar2 If, I haven't tried yet to shotcut the evaluation if it's not needed
03:27:30 <salpynx> .. the logic seems inverted from what I'd expect isNonZero(0) == 1 , but i can work with that
03:30:47 <Raoof> you can think of u in Ar2 as a universal database and function using u as queries of the database, now without using the recursion of python and using only u can you write a query that computes ack ?
03:35:52 <salpynx> yes, but it involves a while loop, which i guess i could simulate by allowing infinite arguments or some other infinite repetition of source code
04:07:40 <salpynx> lol, i was doing well until i came to implementing 2**(n + 3) - 3 , and i think that was mentioned earlier as being hard/impossible. I already used up one infinity for faking a loop. I'm going to stop for the day on this, but i'll probably pick this up later because I already have a partial PMMN ackerman function that's going to need similar effort
04:11:37 <korvo> Raoof: Okay, I'm starting to get a little worried. Please notice that there is no value of e (the first argument to u) such that (lambda x: u(e,x)) computes the following sequence starting at 0: 1, 3, 5, 1, 2, …
04:12:11 <korvo> I generated this in the traditional fashion: [u(x, x) + 1 for x in range(20)]
04:12:42 <korvo> I recognize that you disagreed with this earlier, but we never got to *why*. So, please explain why this isn't a diagonalization.
04:16:11 <Raoof> korvo if you believe that you can compute any total computable function using u then g(x) = 1+u(x,x) = u(u(x,x)) is also a total computable function but u does not have an index for it
04:16:37 <korvo> Raoof: So u isn't a universal database, even if you believe it is.
04:18:36 <Raoof> korvo consider an ordinary mysql database if your tables does not have a summation column does that mean you can not write a query that does summation ?
04:19:35 <korvo> Raoof: Please don't be a Gödel crank. You can just choose to not do it.
04:20:19 <Raoof> korvo I accept any other alternative
04:20:44 <korvo> Raoof: u doesn't have an index for g. Therefore u doesn't have indices for all total computable functions. Done, yes?
04:25:05 <Raoof> korvo yes but do you think u is not a universal operation ?
04:25:09 <korvo> Anyway, sorry for being a jerk about this. You gotta understand that anti-Gödelian folks regularly vandalize wikis and derail philosophy/maths forums on a regular basis. There's one crank who's passed away and we're still finding his sockpuppets.
04:25:57 <korvo> Raoof: It's clearly not a universal operation, no. This is why I'm worried; it seems like you see the evidence and yet aren't accepting it.
04:28:02 <Raoof> korvo can I compute all total computable functions using u even if u does not have an index for all total computable functions ?
04:29:15 <Raoof> korvo I have to say that Ar not Ar2 has an index for g at least there is not proof that it does not have
04:30:05 <korvo> Raoof: No, because there aren't other ways to compose multi-parameter functions besides Kleene's operations. So, if there were some func that u could compute, but u didn't have an index, then it would have to be composed from other funcs that u doesn't have indices for either, and there's no acceptable base case.
04:30:21 <korvo> Raoof: I was using Ar2.py from your repository. I had to remove the tests because they weren't importable.
04:33:47 <korvo> If you pick a code for Ar, then the same construction will go through. Diagonalization is an algorithm.
04:34:30 <Raoof> korvo let me ask an uncontroversial question what is the set of functions that inc,div,mul,sub in Ar and u in Ar2 computes ?
04:35:26 <Raoof> korvo not that div in Ar is different than Ar2. div in Ar has a gap value meaning n/0 = halt and return n
04:35:48 <korvo> Not sure. I think it's a subset of primitive recursive functionals, but I'm not sure exactly what can be computed.
04:36:34 <korvo> It's kind of an awkward basis. The traditional basis is zero, succ, and higher-order primitive recursion (zero, succ, and pr in Cammy) but I suppose that your choice is workable.
04:37:06 <korvo> This channel's got a weakness for silly bases, which is in your favor. I bet somebody would help you work it out.
04:40:58 -!- chiselfuse has quit (Remote host closed the connection).
04:41:15 -!- chiselfuse has joined.
04:41:49 <Raoof> korvo does the sentence "Godel's sentence is true and not recursively provable but non-recursively decidable" make sense to you ? if it does you can see that my choice of basis is not that awkward
04:43:08 <korvo> The first two parts make sense. The third part sounds tricky but not important right now.
04:43:54 <korvo> Fix a first-order arithmetic like PA and a Gödel coding. The Gödel sentence G is a true fact about *the* natural numbers, but it's also unprovable in PA, which means that PA doesn't know everything about the natural numbers.
04:46:53 <korvo> I don't see what the choice of basis has to do with it though. If you choose a basis that isn't strong enough for Gödel coding (as suggested by sub and div) then you can't prove a bunch of standard facts about natural numbers, e.g. that there are infinitely many even numbers or infinitely many primes.
04:47:29 <korvo> (This is just because weaker theories tend to ratify modular arithmetic, which has only finitely many natural numbers for any chosen modulus!)
04:56:06 <korvo> Okay, done being ranty on IRC. Have a good night.
04:57:49 <Raoof> korvo I have one more question if you have time
04:59:04 <Raoof> korvo in Ar2 you can define an anti-diagoal function that u does not have an index for it, how does this match in your model ?
05:01:15 <Raoof> korvo and if you let me ask another one, you didn't say anything about the gap value of div in Ar, how does that match in your model ?
05:02:01 -!- salpynx has quit (Remote host closed the connection).
05:02:11 -!- salpynx has joined.
05:02:43 <Sgeo> An esolang is being sold on Steam.
05:02:47 <Sgeo> I... cannot process this.
05:03:07 <Sgeo> I guess it's like a Zachtronics game?
05:04:52 <salpynx> which one? I have seen the various kinds of games that involve effectively coding for exotic machines, yes, Zachtronics is what i've seen
05:06:15 <salpynx> Raoof: I'm running out of steam, but arbitary powers of 2 will be hard. po2 = lambda *a: (mul(a[0], sub(2, isNonZero(a[1]))), If(a[1], 0, dec(a[1]))
05:06:49 <salpynx> and then an infinite run of po2(*po2(*po2(*po2(*po2(*po2(*po2(*po2(*po2(*po2(*po2(*po2(*po2(*po2(*po2(*po2(2, dec(5))))))))))))))))) will let you compute 2**n
05:07:17 <Sgeo> It calls itself a "Zacklike" in its description
05:07:46 <salpynx> ... but i'm being silly with infinities, and i think i'm cheating with mulitple return values, and encoding multiple things in one value is made hard in Ar2
05:10:15 <Raoof> salpynx if you allow infinite length program in Ar or Ar2 you can define all computable and non-computable functions very easily using nested If and eq as condition
05:11:58 <salpynx> Raoof: can you implement a function that returns 2**n without infinities?
05:14:32 <Raoof> salpynx that is the thing that is controversial because to prove that 2**n > n^a for large value of n you have to take the limit so it is not clear who is cheating mathematicians or programmers
05:17:01 <Raoof> correction 2**n > n**a where a is constant
05:17:32 <salpynx> Sgeo: I've just looked at A=B, it looks like the language is Thue / string rewriting
05:25:58 <salpynx> well, i think i created a 2**n function in Ar2, but only by faking loops and recursion out of infinite source code, so it's basically still just loops and recursion. I could return a single value by using fixed width cells of any finite size. To prime encode unbounded registers i'd need to use arbitrary exponentiation, which lack of loops makes hard.
05:30:03 -!- Sgeo has quit (Read error: Connection reset by peer).
05:38:31 <Raoof> salpynx infinite length program is a good joke but I don't think it is a fun practice
05:40:38 <salpynx> are you saying there is another way to implement 2**n in Ar2? I'd be interested to see it
05:44:24 <esolangs> [[User talk:Ais523]] https://esolangs.org/w/index.php?diff=134241&oldid=134198 * BoundedBeans * (+791) /* C-INTERCAL with Funge-98 bug */
05:44:35 <Raoof> salpynx if you believe in a breakthrough in computing that allows us to transcend space and time then it is possible but it might take an infinite amount of time to figure it out :D
05:50:12 <Raoof> korvo are you still here ? because I have to go too
05:50:57 -!- salpynx has quit (Remote host closed the connection).
05:51:21 -!- salpynx has joined.
05:57:10 <korvo> Raoof: My approach to Ar2 doesn't say anything at all about those properties. It treats u as a black box. Have a good night.
05:59:34 <Raoof> have a good night everybody bye for now
05:59:40 -!- Raoof has quit (Quit: Client closed).
06:00:09 -!- Guest15 has quit (Quit: Client closed).
06:00:55 <salpynx> do we we agree that a TC language can compute 2**n? (i think i've lost track of what we're trying to prove or disprove at this point)
06:07:06 <esolangs> [[SLet]] https://esolangs.org/w/index.php?diff=134242&oldid=134190 * ZCX islptng * (+136)
06:21:46 -!- tromp has joined.
06:22:46 <esolangs> [[Turtle just want to dig]] https://esolangs.org/w/index.php?diff=134243&oldid=134225 * Gggfr * (+665)
06:23:22 <esolangs> [[SLet]] https://esolangs.org/w/index.php?diff=134244&oldid=134242 * ZCX islptng * (+291)
06:28:25 <esolangs> [[Turtle just want to dig]] https://esolangs.org/w/index.php?diff=134245&oldid=134243 * Gggfr * (-6) /* how it works */
06:28:50 <esolangs> [[SLet]] https://esolangs.org/w/index.php?diff=134246&oldid=134244 * ZCX islptng * (+80)
06:32:11 <esolangs> [[Language list]] https://esolangs.org/w/index.php?diff=134247&oldid=134193 * ZCX islptng * (+11) /* S */
06:32:22 <esolangs> [[Turtle just want to dig]] https://esolangs.org/w/index.php?diff=134248&oldid=134245 * Gggfr * (+8)
06:33:27 <esolangs> [[Turtle just want to dig]] https://esolangs.org/w/index.php?diff=134249&oldid=134248 * Gggfr * (-8)
06:33:35 <esolangs> [[Talk:Pendulum Instruction Set Architecture]] N https://esolangs.org/w/index.php?oldid=134250 * Sgeo * (+171) /* Broken links */ new section
07:00:25 <esolangs> [[User:EvyLah/ideas]] https://esolangs.org/w/index.php?diff=134251&oldid=133766 * EvyLah * (+895) stringcell starting
07:00:40 <b_jonas> int-e: wait a moment, re https://logs.esolangs.org/libera-esolangs/2024-07-28.html#lTc how do you construct RgRgRgRg:------Sb:------Sc:--Cy---- that way?
07:01:01 <b_jonas> (I haven't written a program to try.)
07:02:37 <b_jonas> I'll probably eventually have to write a program for all this shapez stuff, just because I still have some general questions that I'd like to know the answer for, not just this thing
07:05:45 <esolangs> [[SLet]] https://esolangs.org/w/index.php?diff=134252&oldid=134246 * ZCX islptng * (+15)
07:09:08 <esolangs> [[Special:Log/newusers]] create * Quantum logic circuit * New user account
07:09:51 <esolangs> [[Esolang:Introduce yourself]] https://esolangs.org/w/index.php?diff=134253&oldid=134042 * Quantum logic circuit * (+162)
07:11:08 -!- X-Scale has joined.
07:11:44 <esolangs> [[User:Quantum logic circuit]] N https://esolangs.org/w/index.php?oldid=134254 * Quantum logic circuit * (+112) Created page with "'''Quantum logic circuit''' is an important concept used in quantum computing. [[Category:Quantum_computing|?]]"
07:12:32 <esolangs> [[User:Quantum logic circuit]] https://esolangs.org/w/index.php?diff=134255&oldid=134254 * Quantum logic circuit * (+19)
07:12:54 <esolangs> [[User:Quantum logic circuit]] M https://esolangs.org/w/index.php?diff=134256&oldid=134255 * Quantum logic circuit * (+9)
07:14:28 <b_jonas> https://xkcd.com/2965/ => this one would have been funny ten years ago, back when we were making jokes about that tobacco store that burned down here in Budapest. it's less funny now after the 2020 explosion of unclaimed ammonium nitrate in Beirut, and since the parliament forced tobacco products out of ordinary supermarkets so now they're all sold in dedicated tobacco stores that sell almost nothing
07:14:34 <b_jonas> but alcohol and tobacco, so there's now so many tobacco stores that you do expect some to burn down, plus one actually burned down close to where I live a few years ago.
07:18:30 <esolangs> [[SLet]] https://esolangs.org/w/index.php?diff=134257&oldid=134252 * ZCX islptng * (+238)
07:23:26 -!- tromp has quit (Quit: My iMac has gone to sleep. ZZZzzz…).
07:50:55 -!- X-Scale has quit (Ping timeout: 256 seconds).
08:12:01 -!- X-Scale has joined.
08:26:03 -!- X-Scale has quit (Ping timeout: 256 seconds).
08:35:29 <esolangs> [[SLet]] https://esolangs.org/w/index.php?diff=134258&oldid=134257 * ZCX islptng * (+261)
08:41:45 -!- X-Scale has joined.
08:52:07 -!- X-Scale has quit (Ping timeout: 256 seconds).
09:02:18 -!- cpressey has joined.
09:04:19 -!- X-Scale has joined.
09:07:20 -!- __monty__ has joined.
09:09:07 -!- X-Scale has quit (Ping timeout: 256 seconds).
09:18:02 -!- wib_jonas has joined.
09:19:20 <wib_jonas> ok, I rescind my objection. RgRgRgRg:------Sb:------Sc:--Cy---- is just impossible, and RgRgRgRg:------Sb:------Sc:----Cy-- can be built by just joining a zigzag on top.
09:22:48 -!- X-Scale has joined.
09:43:07 -!- X-Scale has quit (Ping timeout: 256 seconds).
09:48:33 -!- joast has quit (Ping timeout: 248 seconds).
09:49:07 -!- joast has joined.
10:15:37 <int-e> wib_jonas: yeah the quadrant-supporting-opposite-quadrant is a fairly big class of shapes that can't be made
10:16:01 <int-e> So anything like Cc------:----Cc--
10:16:18 <int-e> with optional slices above and below
10:16:47 <int-e> wib_jonas: I can share my code if you like but it's also a fun programming puzzle/exercise.
10:17:26 <int-e> (I like working with bit masks)
10:18:00 <wib_jonas> sure, this is easier if I program myself than if I directly use someone else's code. I may want to compare *results*, to verify that we're computing the same thing, but not the code.
10:18:40 <wib_jonas> anyway, one interesting case is RuRuRuRu:------Sb:------Sc:--CgCy-- , which can be built, but only using a top support layer that disappears
10:19:48 -!- amby has joined.
10:21:12 <int-e> Ru------:------Sb:------Sc:--CwCw-- is slightly more awkward to make
10:27:23 <int-e> wib_jonas: that "parrot" shape is similar in concept too... it requires attaching a floating zig-zag shape with a 5th layer
10:28:22 <int-e> (I mean the --Wb----:Cg--Cg--:--Sr--Cy:Rw--Rw-- shape)
10:30:04 <int-e> wib_jonas: In any case, I find that convincing myself that shapes are impossible really hard.
10:31:02 <int-e> and I'm sure that I've given that question more thought than most
10:51:00 -!- esolangs has quit (Ping timeout: 258 seconds).
10:51:38 -!- esolangs has joined.
10:51:38 -!- ChanServ has set channel mode: +v esolangs.
11:04:36 <wib_jonas> int-e: I'm fine if the impossibility proof is hard since we can brute force it. what I will want to get is two things to prove that the possible shapes are easy: (1) a human-understandable algorithm of how to make any possible shape, even if the proof isn't necessarily easy to understand, further (2) an algorithm of how to make any possible shape
11:04:37 <wib_jonas> that's also not too hard to translate to a shapez factory, as in ideally I don't want a huge ROM, but I don't require that the factory must be synchronized like yours is. In both case I'd like a construction that doesn't use much more joins than necessary for the hardest shapes. For (2) I'd like to have not more actual joiners placed in the factory
11:04:37 <wib_jonas> than necessary (for some reasonable fixed throughput, don't loop everything through one joiner, in fact don't loop anything period). I don't mind if the second makes easy shapes in too many joins. For (1) I'd like one that uses as few tricky joins or cuts as necessary for every shapes, but I don't care if it doesn't optimize the number of
11:04:38 <wib_jonas> non-tricky joins or cuts for easy shapes such as those you can just build layer by layer.
11:06:28 <int-e> wib_jonas: Yeah that's slightly different goal from mine. I wanted a (mechanically) uniform recipe without many special cases. I pay for that with the huge ROM.
11:07:04 <int-e> The synchronization wasn't a goal initially... it's just something I could relatively easily do on top of those requirements.
11:07:27 <int-e> Plus it allowed me to reuse a good chunk of my existing MAM design.
11:08:34 <int-e> The ROM "only" stores ~6k shapes btw.
11:08:35 <wib_jonas> You can reuse a good chunk of an existing MAM design anyway, like at least the shape farms and pigment farms and selecting 16 quadrants with specific shapes and colors.
11:11:20 <wib_jonas> (I get the irony that the MAM that I built can't actually do that, it can only select 12 quadrants with shapes and colors given in the input, not 16.)
11:11:23 <int-e> Anyway. I'll be interested in what you come up with (if anything, no pressure).
11:11:57 <wib_jonas> yeah, when I feel like writing this sort of program I don't know if I'll want to do this or instead continue the polyomino/polyform counting
11:12:47 <int-e> (But can produce 2 belts worth of output for "simple" shapes, so it's not a complete waste.)
11:14:59 <int-e> What's the worst case, hmm. I think the lightest non-simple shape is Ru------:--Rb----:--Rc----:----Rw-- which would only utilize 4 out of the 32 selection outputs.
11:15:56 <wib_jonas> in any case, here my goal won't be to actually build a shapez true MAM factory, nor to figure out how I'd optimize one, I'd just like to understand the tricky parts enough to know how I could build one if I wanted one
11:16:16 <int-e> Though, actually... I think my MAM handles that and produces two belts worth of output for that.
11:17:44 <int-e> Ru------:--Rb----:--Rc----:----RwRw will only utilize 5 out of 32 though.
11:18:15 <int-e> I think that's the actual worst case for that design.
11:18:59 <wib_jonas> int-e: Ru------:--Rb----:--Rc----:----Rw-- => I don't think so. that looks like it's easier to build than the rocket, because you can just build Ru------:--Rb---- from cutting an easy two-layer shape, then --Rc----:----Rw-- by cutting another easy two-layer shape, then joining the two
11:19:46 <int-e> wib_jonas: And yeah, actually building the true MAM is much more daunting than designing a recipe. I mostly had that part done last year.
11:21:00 <int-e> wib_jonas: Sorry. I actually have a *definition* for "simple shape", which is that it consists of two halves that can be stacked together.
11:21:59 <int-e> and here neither left and right half nor north and south half do the trick.
11:22:47 <int-e> But it's still just composed of two half-shapes stacked together, with a tilt. And I detect that and produce two belts of output.
11:23:15 <int-e> Ru------:--Rb----:--Rc----:----RwRw actually requires stacking three parts.
11:23:56 <int-e> And that means I can no longer produce two belts of output.
11:25:19 <int-e> (it also, incidentally I think, requires a 5th layer)
11:32:12 <int-e> Yeah Ru------:--RbRbRb:--Rc---- works for that as well and doesn't require a 5th layer. Still talking about the worst case utilization of my TMAM design.
11:50:54 -!- X-Scale has joined.
11:56:51 -!- X-Scale has quit (Ping timeout: 256 seconds).
11:56:54 -!- cpressey has quit (Ping timeout: 252 seconds).
12:03:34 -!- mtm has quit (Ping timeout: 260 seconds).
12:05:50 -!- mtm has joined.
12:21:36 <esolangs> [[Turtle just want to dig]] https://esolangs.org/w/index.php?diff=134259&oldid=134249 * Gggfr * (-2) /* the printing extension */
12:29:11 -!- cpressey has joined.
12:31:57 <wib_jonas> Ru------:--RbRbRb:--Rc---- => yes, that might be the worst case I've seen in three layers
12:53:31 -!- wib_jonas has quit (Quit: Client closed).
13:00:44 -!- lutherann has quit (Ping timeout: 260 seconds).
13:02:48 -!- lutherann has joined.
13:04:12 -!- cpressey has quit (Quit: WeeChat 4.3.0).
13:04:40 <esolangs> [[Esolang:Sandbox]] https://esolangs.org/w/index.php?diff=134260&oldid=133768 * Unicodes * (+42) /* Tests */
13:09:08 -!- X-Scale has joined.
13:21:37 -!- Noisytoot has quit (Quit: ZNC 1.8.2 - https://znc.in).
13:21:57 -!- Noisytoot has joined.
13:44:48 -!- X-Scale has quit (Quit: Client closed).
14:07:35 -!- salpynx has quit (Remote host closed the connection).
14:55:31 <esolangs> [[Collabi]] https://esolangs.org/w/index.php?diff=134261&oldid=133979 * Qawtykit * (+388) /* Added commands */
14:57:30 <esolangs> [[Collabi]] M https://esolangs.org/w/index.php?diff=134262&oldid=134261 * Qawtykit * (+25) added User Edited category
14:58:38 <esolangs> [[Collabi]] M https://esolangs.org/w/index.php?diff=134263&oldid=134262 * Qawtykit * (+2)
15:01:44 -!- visilii has joined.
15:30:03 <esolangs> [[SLet]] M https://esolangs.org/w/index.php?diff=134264&oldid=134258 * PythonshellDebugwindow * (+51) Categories
15:37:43 <esolangs> [[2 bytes B)]] M https://esolangs.org/w/index.php?diff=134265&oldid=110038 * PythonshellDebugwindow * (+103) Categories
15:42:20 <korvo> Well, I'm mortified at my behavior last night. Gentle reminder that anybody can ask me to leave if I'm being disruptive.
15:44:50 -!- tromp has joined.
15:53:39 <esolangs> [[Esolang:Sandbox]] M https://esolangs.org/w/index.php?diff=134266&oldid=134260 * Tommyaweosme * (-78) typo
16:35:38 <esolangs> [[Esolang:Sandbox]] https://esolangs.org/w/index.php?diff=134267&oldid=134266 * Int-e * (+78) Undo revision [[Special:Diff/134266|134266]] by [[Special:Contributions/Tommyaweosme|Tommyaweosme]] ([[User talk:Tommyaweosme|talk]]) Please stop!
17:18:09 <fizzie> What is it with crawlers these days? Someone (single IP) has been crawling logs.esolangs.org at a not-crazy rate (2-3 req/s, though since some pages are quite chunky that's ~2 Mbps), but it's been going on for 5 hours, and there's not *that* much of them. Apparently they've re-fetched some of the full-month pages >100 times in the last 30 minutes.
17:18:34 <fizzie> You'd think "don't fetch the same page repeatedly" would be like the first thing you implement in a webcrawler.
17:27:45 -!- X-Scale has joined.
17:44:38 -!- tromp has quit (Quit: My iMac has gone to sleep. ZZZzzz…).
18:06:24 <b_jonas> fizzie: so it's not just that they're fetching the daily pages instead of the monthly?
18:11:32 <korvo> It's often up to small site admins like you to choose to teach them this important lesson about crawling. I think you've been very polite about the whole thing so far.
18:14:09 <zzo38> There is a crawling delay command that can be added into the robots.txt file but I think not all programs will recognize that command. And, I don't know if there is a HTTP response header that can be added to tell it that the file will not be changed in future (or for a specific amount of time).
18:14:34 <int-e> fizzie: it's likely stupid but the pages are not recognizable as mostly static; there's no Last-Modified or ETag to send back with If-Modified-With or If-Match
18:16:04 <b_jonas> fizzie: if they're fetching dailies and all three formats then five hours with 2 requests per second is less than twice as much as you have pages
18:17:19 <int-e> err, If-Modified-Since
18:17:43 <esolangs> [[Ichi]] M https://esolangs.org/w/index.php?diff=134268&oldid=134207 * TheCanon2 * (+768) Added a Hello, World! program
18:24:05 <esolangs> [[Ichi]] M https://esolangs.org/w/index.php?diff=134269&oldid=134268 * TheCanon2 * (+9) added code to show spaces instead of newlines
18:38:35 <fizzie> They were also fetching the daily pages, but only one format. Besides, 5 hours * 3600 seconds/hour * 2 requests/second = 36000 requests; surely that's more than the number of pages? The logs start from 2002 (barely), and 22 years * 400 dates/year (to round up) * 3 pages/date is only 26400 pages.
18:39:28 <fizzie> Oh, less than twice as much, but still more. Sure.
18:40:22 <fizzie> I feel like I at least thought about adding one of those headers, but maybe didn't actually do it.
18:40:53 -!- X-Scale has quit (Ping timeout: 256 seconds).
18:41:55 <fizzie> The content of all but the last day/month is technically static, though I guess it'd also need to account for any changes to the formatting.
18:45:04 <fizzie> Maybe I could generate an ETag that's a combination of the raw data file size (since the files are either frozen or append-only) + a format version number I increment for any change.
18:45:49 <int-e> similarly, Last-Modified could be the maximum of the logical page timestamp and the last time you updated the format.
18:47:08 <fizzie> I'll add that to the list of ideas.
18:50:53 <int-e> wow, this is mainstream news? "Website-Betreiber beschweren sich über gierige KI-Bots" -- "Website operators complain about greedy AI bots"
18:51:45 <int-e> One source is https://www.theverge.com/2024/7/25/24205943/anthropic-ai-web-crawler-claudebot-ifixit-scraping-training-data
19:05:10 <int-e> Claude-Web is now known as ClaudeBot. One wonders why (not really, it's absolutely obvious that this subverts robots.txt.)
19:07:12 <fizzie> Wonder if that's the explanation for consistent traffic from "facebookexternalhit/1.1" too.
19:08:19 <esolangs> [[User:Fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff]] https://esolangs.org/w/index.php?diff=134270&oldid=133721 * Fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff * (+12)
19:08:33 <esolangs> [[Rnadom]] https://esolangs.org/w/index.php?diff=134271&oldid=134238 * Fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff * (+4) /* commands */
19:08:40 <esolangs> [[Rnadom]] https://esolangs.org/w/index.php?diff=134272&oldid=134271 * Fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff * (-4) /* commands */
19:08:51 <esolangs> [[Rnadom]] https://esolangs.org/w/index.php?diff=134273&oldid=134272 * Fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff * (+4) /* programs */
19:08:57 <fizzie> Meta says at https://developers.facebook.com/docs/sharing/webmasters/web-crawlers/ that "facebookexternalhit/1.1" is when content "was shared on one of Meta’s family of apps, such as Facebook, Instagram, or Messenger", and that they use "meta-externalagent/1.1" for "use cases such as training AI models", but esolangs.org was seeing a constant crawl of all pages using "facebookexternalhit/1.1"
19:08:59 <fizzie> (from Facebook's IP ranges).
19:09:09 <esolangs> [[Rnadom]] https://esolangs.org/w/index.php?diff=134274&oldid=134273 * Fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff * (-4) /* polyglot */
19:09:30 <esolangs> [[Rnadom]] https://esolangs.org/w/index.php?diff=134275&oldid=134274 * Fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff * (+240) /* polyglot */
19:09:43 <esolangs> [[Rnadom]] https://esolangs.org/w/index.php?diff=134276&oldid=134275 * Fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff * (-230) /* polyglot */
19:09:48 <int-e> so maybe the crawler has started sharing its links on some internal Facebook app ;)
19:09:59 <esolangs> [[Rnadom]] https://esolangs.org/w/index.php?diff=134277&oldid=134276 * Fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff * (-10) /* polyglot */
19:10:13 <esolangs> [[Rnadom]] https://esolangs.org/w/index.php?diff=134278&oldid=134277 * Fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff * (+0) /* commands */
19:10:28 <esolangs> [[Rnadom]] https://esolangs.org/w/index.php?diff=134279&oldid=134278 * Fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff * (+4) /* polyglot */
19:13:20 <int-e> . o O ( I guess the Preview button must be hidden or something. )
19:16:23 -!- salpynx has joined.
19:24:03 <salpynx> huh, the wiki has no article on uxntal or uxn, but there are references to both, and one red link to uxntal. uxntal is the correct name for the language i believe.
19:24:35 <salpynx> the asm looks pretty normal, but the glyphs and hand signs make it more esoteric
19:25:32 <b_jonas> fizzie: sure, but the User-Agent could just be someone else lying
19:27:39 <zzo38> I had suggested before, if maybe someone should add articles about uxn. Note that there is also a nonstandard variant of uxntal (but which can still produce standard binary uxn files).
19:28:57 <zzo38> It is unusual that it has hand signs, which is something I had not seen in other programming languages.
19:33:35 <b_jonas> int-e: if you have a shape that is possible, and you remove its top layer, is the remainder shape always possible or empty?
19:36:09 <salpynx> zzo38: i'll add at least a stub article later today to collect information about the uxn ecosystem, once i get all the terminology straight. I'll focus on the language uxntal. Haven't seen the variant yet, but will add variant details if I find them
19:36:15 <b_jonas> ah no, you already mentioned the parrot, and that's a counterexample
19:36:30 <int-e> Also that 3 layer example from earlier.
19:36:59 <int-e> generally the top layer may contain quadrants that support lower layers while stacking.
19:37:10 <zzo38> If you have questions about uxn then I can try to answer them.
19:38:33 <int-e> b_jonas: Ru------:Rb--Rb--:RcRcRcRc may be the most convincing example
19:39:05 <zzo38> (Note that the code and documentation of uxn38 (except uxnasm.c) is public domain, in case you need a copy of any of that; the official code and documentation is not public domain.)
19:39:40 <int-e> We have to displace the bottom support to make the example work: --Ru----:Rb--Rb--:RcRcRcRc
19:40:50 <zzo38> http://esolangs.org/wiki/Deadfish/Implementations_%28M-Z%29#Uxn has the nonstandard syntax, although the hex dump below it should be compatible with standard uxn (although I have not tried it, but I would expect it should work)
19:43:27 <zzo38> (The hex dump is also valid uxntal code)
19:47:27 -!- Thelie has joined.
19:50:19 -!- tromp has joined.
19:50:28 <fizzie> You can lie about the user agent, but presumably it's harder to lie about the source IP, and I don't think Facebook runs any general-purpose cloud services so that third parties could perform requests that appear to originate from their network.
19:51:53 <fizzie> Though perhaps "Facebook apps" can do something like that, I don't really know what they are.
19:56:35 -!- X-Scale has joined.
20:09:17 -!- X-Scale has quit (Ping timeout: 256 seconds).
20:25:17 -!- sprout has quit (Quit: https://quassel-irc.org - Chat comfortably. Anywhere.).
20:27:48 -!- sprout has joined.
20:31:00 <esolangs> [[Genewrath]] N https://esolangs.org/w/index.php?oldid=134280 * BoundedBeans * (+10420) Created page with "Genewrath is an esolang by [[User:BoundedBeans]] where all data is stored in generic types. ==Reading of code== Carriage returns and line feeds are completely ignored, and they can be used in the code freely. They can be treated as if they are not in the code;
20:31:28 <esolangs> [[Language list]] https://esolangs.org/w/index.php?diff=134281&oldid=134247 * BoundedBeans * (+16)
20:31:44 <esolangs> [[User:BoundedBeans]] https://esolangs.org/w/index.php?diff=134282&oldid=132881 * BoundedBeans * (+15)
21:04:41 <esolangs> [[Burn]] M https://esolangs.org/w/index.php?diff=134283&oldid=129869 * TheCanon2 * (-12) If Burn can simulate Rule 110, and Rule 110 is Turing complete, then we must assume Burn is Turing complete until we can prove otherwise.
21:09:14 -!- Thelie has quit (Quit: Leaving.).
21:16:20 -!- tromp has quit (Quit: My iMac has gone to sleep. ZZZzzz…).
21:37:38 -!- lutherann has quit (Quit: i use arch btw).
21:38:00 <esolangs> [[Divmeq]] M https://esolangs.org/w/index.php?diff=134284&oldid=134171 * TheCanon2 * (+112) Added a note about Divmeq quines
21:38:06 -!- lutherann has joined.
21:42:48 <esolangs> [[Esolang:Sandbox]] https://esolangs.org/w/index.php?diff=134285&oldid=134267 * Tommyaweosme * (+41) cite error: storm
21:46:44 <salpynx> idle observation: there seem to be frequent "X is Turing complete (under specific conditions or modification Y)" statements, where the parenthesised bit gets lost.
21:50:31 <zzo38> If there are some in esolang wiki that had lost one part, hopefully you can fix it. (That includes my user page and subpages; if you found any problems with them you can fix them since it is the wiki that anyone can improve)
21:52:10 <esolangs> [[User talk:Quantum logic circuit]] N https://esolangs.org/w/index.php?oldid=134286 * Tommyaweosme * (+392) Created page with " Welcome to the wiki! -[>+<---]>++.<-[>>++<<-----]>>-.+++++++.---------.++++++++++++.--.--------.>>-[-[-<]>>+<]>-.<<<<+++[<+++++>-]<.-----.>>>>>.<<<<<+++++.------------.---.>+++[<++++++>-]>>>>.<<<<<.>++[<------->-]<.++.--.>>>>>+. ~~~~"
21:55:13 <esolangs> [[Deadfish/Implementations (nonalphabetic and A-L)]] https://esolangs.org/w/index.php?diff=134287&oldid=132736 * BoundedBeans * (+860) C with only for loops
22:03:50 <esolangs> [[Esolang:Sandbox]] https://esolangs.org/w/index.php?diff=134288&oldid=134285 * Salpynx * (+157) /* Tests */ do we use refs here?
22:04:20 -!- tromp has joined.
22:05:28 <esolangs> [[Esolang:Sandbox]] M https://esolangs.org/w/index.php?diff=134289&oldid=134288 * Fizzie * (-36) /* Instructions */ Restore sandbox instructions (yes, the section title is part of them)
22:07:07 <esolangs> [[Esolang:Sandbox]] M https://esolangs.org/w/index.php?diff=134290&oldid=134289 * Salpynx * (+103) more ref on article tests
22:09:49 <esolangs> [[Esolang:Sandbox]] https://esolangs.org/w/index.php?diff=134291&oldid=134290 * Fizzie * (+66) Test the <references /> tag.
22:12:00 <fizzie> Kinda feel like most of the actual tests people use the sandbox for, they could just do in a preview rather than by editing a page.
22:12:08 <fizzie> (Including what I just did.)
22:13:07 <salpynx> fizze: I was literally just doing that with my recent sandbox edit, I considered not saving, but thought maybe other ppl might want to see what refs looked like.
22:14:08 <salpynx> I'm not convinced the appended refs will always work at the end of current articles, it needs a heading to make sense, and it can only be the last thing in the article.
22:17:36 <salpynx> zzo38: I was more making a general comment about some TC claims, prompted by R110 is TC (I think I understand it, but it still feels complicated). Filling in the correct Y clearly is often _hard_
22:19:29 <salpynx> fizzie: ok, I would use the wiki thanks for that edit :) I couldn't figure out the tag to do what you just did
22:20:11 <salpynx> (I was trying wikipedia templates which reasonably don't exist here)
22:24:44 -!- tromp has quit (Quit: My iMac has gone to sleep. ZZZzzz…).
22:26:21 -!- QTpye has joined.
22:27:15 <QTpye> hello, anyone know how to contact an admin from the esolang wiki?
22:27:45 <QTpye> it says on the site this is the offical irc channel. i have been unadle to find any refrance on the wiki itself
22:31:25 <salpynx> this is the right place, admins frequent here (and/or read the logs)
22:40:48 <QTpye> im here wondering if i could change the name associated with my wiki account. i could always just make a new account under the new name, and change all the refrances to my new account. but that would be a lot of work, and be messy in terms of keeping tack of what i have done on the wiki.
22:51:47 <fizzie> It's not self-service, but if you put your desired new name somewhere on the wiki (e.g., your own user talk page, or mine) as a way of proving account ownership, I can try to remember how the rename-user function worked.
22:52:26 <fizzie> (And of course only if the new name doesn't exist already.)
22:54:22 <fizzie> https://esolangs.org/wiki/User_talk:Fizzie has one rename request already, so I must've known how to do it at least back then.
23:05:50 -!- __monty__ has quit (Quit: leaving).
23:15:35 <esolangs> [[User:Aspwil]] https://esolangs.org/w/index.php?diff=134292&oldid=83767 * Aspwil * (+99)
23:16:23 <QTpye> there you go, a request on the user talk page of my account, from the account requesting a name change
23:22:58 <zzo38> I had received many invalid requests on my HTTP server that seems to be TLS as far as I can tell (the first byte of the request is 0x16), although this server is non-TLS (and it is on port 80 only).
23:23:35 <fizzie> QTpye: Do you want redirects left for the old name? (It seems to be the default to do so.)
23:31:56 <fizzie> Well, I'll let it use the defaults, and we can always delete them later.
23:32:02 <esolangs> [[Special:Log/move]] move * Fizzie * moved [[User:Aspwil]] to [[User:QTpye]]: Automatically moved page while renaming the user "[[User:Aspwil|Aspwil]]" to "[[User:QTpye|QTpye]]"
23:32:02 <esolangs> [[Special:Log/renameuser]] renameuser * Fizzie * Fizzie renamed user [[User:Aspwil]] (128 edits) to [[User:QTpye]]: By user request
23:32:47 -!- Sgeo has joined.
23:33:01 <fizzie> The new name should work for logging in going forward.