« first day (3814 days earlier)      last day (1036 days later) » 

5:06 AM
@hyper-neutrino i asked something similar to ais523 a while back, chat.stackexchange.com/transcript/240?m=52007337#52007337
@LeakyNun you think that's turing complete?
 
@lyxal does vyxal support scientific notation?
 
no, not yet at least
 
5:26 AM
@lyxal do you plan on making an IDE for vyxal (with syntax highlighting; code linting; autocompletion etc.)?
 
how much autocompletion can a golfing lang even have lmao
 
@PyGamer0 syntax highlighting is already done, autocompletion is already implicit and code linting doesn't seem that helpful
@RedwolfPrograms the random question userscript isn't showing up
 
also linting golfed code sounds a) like a pain to develop and b) like a pain to deal with
 
I mean half the goal for a golfing language is to reduce the wasted information as much as possible, which is what makes autocompletion possible in the first place
@lyxal Can repro, just a sec
 
Repro'd
Also it would be nice if you could make it a styled hyperlink so you can open in new tab...
 
5:31 AM
@Ausername But you can. Control clicking is implemented.
 
good
 
Oh
I is on mac
 
a) then just cmd click b) i think mac still accepts ctrl
@RedwolfPrograms doesn't seem to work for me (FF)
 
@RedwolfPrograms works on my end
 
5:32 AM
ctrl+click = right click on mac
 
then cmd-click
 
That does nothing
 
lol wut
 
(different)
 
guess you just have to right click + open in new tab
 
5:33 AM
@hyper-neutrino Any errors?
 
@RedwolfPrograms nope
 
@hyper-neutrino (I is firefox)
 
Odd. Did it work before?
 
didn't have it installed until now
 
Hmm. Are you on the homepage?
 
5:34 AM
O
 
where is it supposed to appear? also yes
 
By the active/new/hot/...
 
wait what
okay nvm i'm just stupid then
i could've sworn it was on the left sidebar
 
There's a sidebar one too
 
5:35 AM
 
okay never mind it works fine
 
yeah i had the sidebar one on chrome >:-/
 
CC (Chat challenge {might be hard}): Take one of your esolang and rewrite it in C. It is hard but not impossible. Whoever completes this will get 150 brownie points.
 
Doesn't seem particularly fun :p
 
5:41 AM
@PyGamer0 S10K: void main(){for(int i=0;i<10000;++i)printf("S");}
 
@Bubbler nice
 
Oh, I didn't know you invented the legendary S10k :o
:p
 
wasn't it made just to test TIB?
 
Yeah, I made it up for TIB, but it is a "language" after all :P
 
what computational class /s
 
5:43 AM
DSO has s10k
 
uncomputable /s
 
It's quite clearly TC
3
 
@RedwolfPrograms how?
 
It has loops because you can push the button multiple times
 
@PyGamer0 int main(){} - the like 3 language repos i have on github that are totally empty
@RedwolfPrograms lmao
 
5:44 AM
@hyper-neutrino what about yuno?
 
@hyper-neutrino int main();?
 
@PyGamer0 why would i reimplement yuno a third time
@Ausername ¯\_(ツ)_/¯ i am not good with C
 
Writing any golfing language in C is like writing an essay with your own blood in 72pt font
4
 
@hyper-neutrino why not
 
5:45 AM
just used it for my cs course
 
The cleverest rickroll
 
@PyGamer0 also again, writing raw C itself is for masochists
@Ausername not really, a simple hover still makes it obvious
 
What's the padding supposed to do? Do some browsers hide that it's a rickroll?
 
¯\_(ツ)_/¯ my userscript doesn't detect it
but my userscript is like 3IQ so a 4IQ rickroll is enough to thwart it
 
@hyper-neutrino Go to the context of the discssion, and look at Wezl's PFP
Also any feedback? #2 might be posted, if good...
 
5:49 AM
as far as rickrolls go, i guess it's reasonably clever, which isn't saying much
 
@Ausername Vyxal thing looks good to me
 
Ok
I'll just let one or two people look some more before I post
 
I just 'roll'd myself trying to test an anti-rickroll script :|
 
@Ausername ha u didnt roll me
i saw the link using the ancient hoveer using mouse method
@lyxal WHAT
 
5:56 AM
Nice try
 
@Ausername wdym
 
Wait was that not a rickroll?
 
NO
 
Just how long is the program you
 
i was asking about vyxal
 
5:57 AM
're writing
Wait you don't use Vyxal
it's definitely a rickroll
 
@Ausername i am trying to learn
 
@PyGamer0 no.
 
Then what were you linking?
 
@PyGamer0 I'm not clicking that
 
And why a URL shortener
 
5:58 AM
@Ausername vyxal TIO
 
@lyxal it's definitely a rickroll
 
@Ausername its too long
*why are we arguing over a vyxal tio link
 
Because it's not
@lyxal BTW deltas are backwards
 
look, rickrolls aren't even that funny, please stop overusing them. they're fine in moderation but don't make me bring back the old rickroll rules
 
@hyper-neutrino this room had rickroll rules?
 
6:00 AM
(negated
 
not specifically against rickrolls. but they are by all definitions noise, and users who generate too much noise over meaningful content have been kicked and even suspended network-wide (because we can't suspend someone from just one room even if we wanted to)
this room's current state is pretty satisfactory overall. just would like to keep it that way :p
 
@lyxal Plz disable george
its brokemn
Or nor
 
@Ausername no
@Ausername heh, they probably have been for a while
 
6:29 AM
@user41805 sorry, I must have missed the question back then; the original proof used "the tape head visits one of the cells whose location is in this geometric progression more than once", but I eventually managed to simplify it to "the tape head moves left of its starting location"
in fact, you can actually make it halt properly just by using a half-infinite tape and halting when it falls off the end
@PyGamer0 I think some of my esolangs were written in C, although I usually use Perl, and have been using Rust for some lately
and many, I didn't implement at all, because thinking about the language is more interesting than programming in it
 
@ais523 Hey ais523!!
 
@AviFS it turns out that you can read TNB without JavaScript enabled (via using the transcript), just not post there
and I have JavaScript off by default in general, and don't want to make an exception for here unless I need to post something
so mostly, I just read rather than posting
(and I don't read everything nowadays; I used to, but activity here has gone up by a factor of 6 or so over the last couple of years for reasons I don't understand and it's no longer sensible to keep up with it)
 
That does make sense, but that's still pretty neat! I've never heard of people turning off js though
 
it makes the Internet so much less annoying
 
@ais523 Really? That is surprising
@ais523 And webapps not funcitonal, no?
 
6:36 AM
I also have cookies off by default; that makes it more annoying but that's the sort of person I am
@AviFS indeed, but most pages shouldn't need JavaScript to do what they do
I have a separate browser with JS enabled for when I need to use one (that's what I'm doing right now, using a secondary browser)
 
Makes sense, interesting strategy. I'll have to try it sometime!
 
btw, if everyone had JS disabled (also video and audio tags, like I do), then rickrolls wouldn't exist because they wouldn't work
 
@ais523 Not trying that one!
 
I have never successfully been rickrolled, for this reason; YouTube just doesn't work in my primary browser (and nor do its competitors)
 
@ais523 That's supposed to be a good thing?
:P
I may or may not have a rickroll easter egg in one of my github projects in dev right now, haha
But it's just a hyperlink, why shouldn't it work?
 
6:40 AM
because it goes to a youtube page that doesn't load due to there being no JS or audio or video
 
I forgot about this interview
 
I guess you could link to a text version of the lyrics but that isn't that interesting
 
Ah, good point
 
I also used to have a policy of not following links in chat (originally IRC); I forget the reason, though (I don't think it was related to rickrolls)
I even set my client to delete links from inbound messages altogether
 
By the way, you may be slightly dismay to find out I took over this challenge with one of your less interesting wonderful languages!
 
6:42 AM
this probably isn't a good idea and I don't know what I was thinking at the time; there was a reason, but it probably wasn't a good reason
 
Can you guess which one I used without scrolling down?
 
I guessed which one you used from the URL
although, I'd seen the answer in question before
 
Cheater!
I'm slightly horrified I just got:
> Enthusiast - Visit the site each day for 30 consecutive days. (Days are counted in UTC.)
 
even so, most of my esolangs are deterministic
if I ever got that badge I would delete my account
although I'm a bit extreme in that regard
 
And it's only a silver badge? What's gold-worthy-- a year?!
 
6:45 AM
100 days
 
that badge has caused way more trouble than it's worth
with people trying to automate it, then getting upset that their automated badge progress is getting detected and zeroed
and, there being no way to get rid of it (short of deleting your account), so you're basically advertising the fact that you've put unhealthy amounts of attention into Stack Exchange
 
badges as a whole, IMO, are just a somewhat interesting gimmick that usually ends up causing more trouble than its benefits or at best just not being that interesting ¯\_(ツ)_/¯
still don't get the point in Peer Pressure
 
my question complaining about badges on meta is at -26
fun fact: I set my badge tracker on meta.se to Peer Pressure to advertise the fact that I don't have it
(for whatever reason, your progress on the currently tracked badge is public)
 
I didn't even know it existed, and now I wish it didn't exist.
It's rude.
 
same thing on codegolf.se, with Supporter (I can't upvote because that needs 15 rep)
 
6:49 AM
It's not that bad, though, I think. It becomes like checking your email after a while.
And I'll definitely stop all of a sudden with no warning, and not touch the site for over a year. I go through phases like that...
 
checking your email manually can actually ruin peoples' lives, it triggers the same part of your brain that gambling does ("will I have a new email? no, not this time") so you can spent hours doing it over and over again if you aren't careful
this can be fixed by using an email client that notifies you when you have a new email, and just turning the notifications on and off depending on whether you want to check email or not
that way there's no incentive to check it manually
 
I know, I agree. That sort of behavior is really bad
The worst is things that ding all the time so you think you have something, but most of the time you don't... Discord
Yes, I know you can mute them, but there's always a few that escape
And some group pings are meaningful and others aren't
 
due to an interesting misconfiguration, the bug report email address for a game I work on pinged everyone on the discord server for discussion of the game, even though global pings were turned off
that was interesting the first time it happened
(I don't use Discord, but observed this from the other side of an IRC bridge)
sites which ping on @ signs should probably not ping on them if they're in the middle of an email address :-P
 
wait hyper is playing LoL?
 
CMC: program producing the largest finite output; no byte may be used more than twice
 
7:04 AM
Hypothetical: If you can write a self-interpreter in a language, does that give you some lower bound on its computational complexity?
Do we know it's eg. at least a pushdown automata?
 
if the language is sufficiently low-powered it can sometimes have a self-interpreter more or less by accident
like, the language which is "ignore the program and do nothing" (I forget what it's called, it's probably been created multiple times by now)
I also think there's an HQ9+ variant with a "self-interpreter" command
so no, having a self-interpreter doesn't guarantee any particular level of computational power
(actually, I think that HQ9+ variant also has a command that is documented to make the language Turing-complete, but there's some doubts as to whether it actually does; the documentation for the command is literally "makes the language Turing-complete")
I guess there are three main branches of esolang design nowadays: on CGCC we write esolangs primarily for the purpose of using them; many of my esolangs, I write primarily for the purpose of proving things with them; and there are people who write esolangs as a philosophical experiment into what programming is
(Baby Language is in the third category, incidentally)
 
@PyGamer0 I don't see how this is either surprising or notable.
 
@user41805 oops forgot to link to ais's previous response, starts here chat.stackexchange.com/transcript/message/52024784#52024784
 
oh, I did remember to answer it last time
 
@ais523 you did reply then, in linking now to hyperneutrino i didn't realise that your response didn't immediately follow in the transcript
 
7:17 AM
I guess it's a good thing the two answers are consistent with each other!
 
ninja'd
 
@ais523 I have many other ways to advertise that I spend an unhealthy amount of time here, I don't need to worry about Fanatic :P
@ais523 I always find low voted MSE questions interesting, especially the reactions of a few specific users there who feel like they have to comment on everything
 
I think I've made no secret of just how much I hate Stack Exchange
it's a major contributor towards me not being on CGCC much
 
i ended up hiding vote counts and the rep change notification to stop my obsession with them, which mostly worked
 
It's a shame that there isn't really a golfing site that has the same flexibility and community that you get here. Codidact is trying, but it's really not very active and I dislike their chat system
 
7:27 AM
@user41805 hmm, I wonder if that's a better or worse approach than community-wiki-ing every post?
 
You do still have your reputation shown, and that will increase if you aren't CWing
 
the other reason to lock at 11, incidentally, is that I upvoted some posts before deleting my account, and after recreating it, I could upvote the same posts again
 
I suppose you could disable that as well (or hell, make every post "anonymous") if you really wanted
 
and I have no way of remembering which posts I voted on, so it's safest to not vote at all to avoid double-voting
 
@ais523 Correct me if I'm wrong, but I also remember you saying that you disliked how SE pushed privileges on users?
 
7:30 AM
@ais523 i never tried cw-ing, but i guess the cw post can still get upvotes. ideally though, i would also remove my username from being identified with the post, but the best i came up with was to switch to userXXXXX as my handle
 
@cairdcoinheringaahing pretty close to that, yes, it tries to badger you into doing moderation tasks when you get more rep
and on CGCC it tends to do that to people well before they understand what the tasks are for
 
@cairdcoinheringaahing i think when i tried that, i was already slowing down writing answers here, but anyways it's hard to notice changes to a 5 digit number
 
@user41805 Did you used to be Cow's Quack?
 
without the apostrophe, yes
 
@ais523 Yeah, I've noticed a few odd decisions in the review queues by users who have just gotten access from time to time
@user41805 Ah, I was trying to remember who you were because it's weird to see a user pop up with 14k who you don't recognise :P
 
7:32 AM
my favourite example was a post that needed closing hitting HNQ before it got closed, someone posting an answer, and getting enough rep from that one answer (before the close happened) to vote to reopen
btw, if a mod modhammers a post closed, can it be reopened via voting? or does it need a mod? I can't remember
 
@cairdcoinheringaahing i tried that (hiding the username and repcount from each post), but turned it off after a while, it helps knowing the author's proficiency in say, looking for golfs in their solution
 
@ais523 Both work
 
modhammers is something else I hate about CGCC; I don't mind them existing, but the mod should have an option to not hammer
 
That's been asked for for years
 
as it is, getting a diamond basically means that you can't participate in many of the cleanup tasks because your vote would be binding and sometimes you aren't sure enough to be able to hammer things
err, SE in general, not CGCC in particular
 
7:35 AM
Across the network actually. I think SE [status-declined] it a year or so back
 
@user41805 i guess this has had the unintended side effect of me forgetting to vote posts, even though the up/down-vote buttons are still visible.
 
sometimes I look through the vote counts on my answers, and despair at how bad at voting the voters at CGCC are
the evidence seems to be that more people see posts via HNQ than ever see them after they leave HNQ, so it's impossible for the people who know what they're doing to outvote the HNQ influx
 
@ais523 hyper and WW are both fairly active in the review queues (if they review something, it then leaves the queue and doesn't get additional reviews), and WW is pretty proactive in temp-closing unclear questions then reopening them after being clarified
 
right, that's the sort of situation where a modhammer is helpful
 
@ais523 agreed - i don't like the implication that we always know what's best for the community
 
7:37 AM
@ais523 in an ideal golfing, there should be no votes IMO. i tried to petition this opinion at codidact's code golf site, but it was met with disagreement
with the argument being they needed some way to arrange the answers by quality or something like that, but then again a Q&A site might not be the best place to host a code golf site
 
@user41805 my own thoughts about a dedicated site for golfing had the equivalent of SE/Reddit-style upvotes (but not downvotes), but didn't allow them anywhere the default sort order and de-emphasized them in the UI over objective metrics and Slashdot-style votes (where you vote a post down if you think it's gotten too high, even if you like it)
 
But also, we do have a pretty active user base who have moderation privileges, so we're able to very quickly "overrule" the mods on a lot of things if people believe they made a mistake. Or, if a mod isn't sure about closing/reopening, a comment stating their arguments and allowing others to make the decision is a good idea normally
 
my basic opinion on SE is "a Stack Exchange feature is only ever useful for something other than its intended purpose"
 
I do that if I'm not entirely sure two posts are dupes - I'll wait to close until 3/4 people agree that they are
 
this is probably an exaggeration, but not much of one
 
7:41 AM
@ais523 Idk, I find the answer feature to only really be useful for one thing :P
 
@cairdcoinheringaahing what about the standard loopholes post on meta?
 
What about it?
 
the answers aren't answers in the traditional sense, and we have them separate so that we can vote on them separately
I guess the standard I/O methods post is an even better example
but none of the answers are a complete answer to the question, we split them up so that we could debate and vote on them individually
I guess the quest for tetris thing is another case where examples were used in a nonstandard way
 
I suppose they are answers, in the sense that if I asked "What should be a standard loophole?", I'd consider the responses as answers
 
that said, it's going to look very weird if and when a second answer ever gets posted (there is definitely scope for golfing but I'm not sure if anyone plans to work on it)
 
7:44 AM
@ais523 Have you read this? It's a very interesting take on the motivation caused by rep/badges, written by a former SE employee
 
yes, but I can't remember what it said, I'll read it again
ah right, now I remember
fwiw, my experience with Stack Overflow is of asking questions that are too difficult for anyone to answer, so I feel like even in terrms of "let's get people with questions their answers quickly" it only really works at the low end
 
CMM: should a (assumed idempotent) function expression that takes no arguments be interchangeable with a snippet of its return value? I ask this because In many languages, e.g. Haskell, there is no distinction.
 
we allow doing that in "write a program or function"-style things, as a substitute for the function
pretty much for the reason you say
but it may be that the language has to treat them interchangeably
 
didn't know about slashdot's voting style. but expanding further on my ideal golfing site, here's one idea i had been thinking over. it would be anonymised, so that really the only thing for any challenge in a selected language is its shortest answer written so far and when it was submitted, with maybe a community-written text attached to it. i guess it sacrifices ppcg's flexibility, but interesting to see how it would function
i don't know of any other golfing site that prioritises the golfed code over the submittors
 
I'd like to see a codegolf answers page as a list of answers, with comments attached to them by people (not necessarily the original submitter/submitters) clarifying or explaining or whatever
and ideally with essentially equivalent programs (e.g. differing only in variable names) merged
on anagolf, the answers are kept hidden for a week or so before showing them, so you often get lots of duplicate programs because multiple people independently come up with the same optimal solution
(but the lengths are shown, to give a target to aim for)
 
7:52 AM
@ais523 Well I can replace (Haskell) f x="abcdef"; f () with \x->"abcdef", so I can surely replace x="abcdef";x with "abcdef", but can I replace (Python) lambda:"abcdef" with "abcdef"?
 
yes that could also work, thus not hiding the diversity of approaches
 
@pxeger personally I think doing that's reasonable, I think it might be disallowed on CGCC though
 
@pxeger What about side effects?
 
> assumed idempotent
this would obviously not work for functions with side effects
 
@ais523 Personally, I dislike keeping answers hidden. When I get out golfed in Jelly, I'd much rather see what crazy trick you, Jonathan, hyper, Nick, Lynn, xigoi, etc. have used, rather than wait for a week or potentially never see
 
7:57 AM
but they don't necessarily have to be pure I think, e.g. lambda:open("file.txt") is interchangeable (because we generally assume the environment in which a submission is run doesn't antagonistically change it mid-execution)
 
@cairdcoinheringaahing well, the fun is in trying to find the trick yourself without having it spoiled for you
 
And if I haven't already answered, then I use the grad script leaderboard to see what answers are there, and if I think I can outgolf them, only then will I try. If I see "Jelly, 4 bytes - Jonathan Allan", I don't bother trying, because unless I spot something crazy, I'm not going to beat that
 
and you still get to look back after a week and find out if you couldn't figure it out yourself in that time
 
i think it's a good thing to have duplicates and bad answers
 
@ais523 We have vastly different golfing abilities :P
 
7:58 AM
I had a lot of fun trying to tie the leaders on anagolf in Perl back when I was learning to golf
it gives you the motivation to keep rearranging things, trying to squeeze out one last byte, because you know it's possible
meanwhile, I was looking at an arbitrary sample of Jelly answers to old questions, and in many cases the Jelly was clearly winning but was also improvable, people had stopped trying to squeeze out bytes at that point
 

« first day (3814 days earlier)      last day (1036 days later) »