« first day (3776 days earlier)      last day (1068 days later) » 

ngn
7:00 AM
@Adám why have a grey-on-grey low-contrast scheme at all?
 
Because some find it more pleasant to read. You're the only one I know that strongly prefers maximal contrast, so I thought of you.
 
ngn
pretty sure i'm not the only one
 
Agreed. That would be very unlikely. But I don't know a lot of people.
 
I want something in between those two. Pure white-on-black is a bit too much contrast, and grey-on-grey is ugly as all hell. Give me a little blue tint or something
 
Funny, I thought of adding an option to tint slightly reddish or slightly bluish.
 
ngn
7:04 AM
call me old-fashioned but i prefer pure white on pure black
(at least for the text i'll be staring at all day, like a terminal. i don't care much about individual websites)
 
@ngn What do you get by default when loading the site?
 
ngn
@Adám black on grey
 
@ngn Oh. The page is supposed to follow the visitor's preferred color scheme setting, but it seems that it either failed to pick yours up, or yours is set to prefer a light theme.
 
ngn
and i like how noscript completely blocked your custom fonts :P
@Adám how would it even know?
 
@ngn Ah man. I thought I was being clever by avoiding all traces of JavaScript. I didn't realise noscript would block custom fonts.
 
ngn
7:12 AM
my first thought was "hang on, these aren't sans" :)
 
Does redwolf's custom fonts userscript work?
 
does anything from redwolf ever truly work?
;p
 
@lyxal Well, his fax machine is still trying to understand his JS.
 
ngn
@Adám something wrong with your html near the end
 
Thanks. I'll fix that.
 
ngn
7:24 AM
btw, you don't have to put quotes around html5 attributes that don't contain special chars. it would still be valid html5.
(so few people realize this..)
 
I actually know. But it messes up my syntax colourer and my code reformatter, which makes for an unpleasant editing experience.
 
ngn
then you need a better editor
 
People also think font names with spaces need quotes…
 
ngn
@Adám interesting. is it still valid css without quotes?
 
Yes.
 
ngn
7:27 AM
i just tried it at jigsaw.w3.org/css-validator/validator and it gives a warning:
"Family names containing whitespace should be quoted. If quoting is omitted, any whitespace characters before and after the name are ignored and any sequence of whitespace characters inside the name is converted to a single space."
this should be ok
 
Yes. I've never seen any font name with leading, trailing, or extraneous whitespace. That's be a really bad idea.
 
ngn
right
 
In fact, omitting quotes might well lessen the risk of bugs due to doubled or trailing spaces.
 
ngn
if you ask me, putting any spaces in the font name is a bad idea
 
> The Jigsaw software provides the Consortium's Java-based Web server. With a modular architecture and full HTTP/1.1 compliance, the Jigsaw server is a premier experimental platform for W3C and the Internet community. The server uses an object-oriented approach when it comes to the storage of files and the processing of incoming requests, making it both more efficient and easily extensible.
> Jigsaw.w3.org news

20010116 - Hardware upgrade, see the Status page.
20000911 - Updated jdk
20000828 - Patched the kernel (and rebooted)
wow
 
> premier experimental platform
> 2001
 
ngn
8:11 AM
shameless plug: i'm about to become the biggest fish in my pond (projecteuler per country)
 
8:28 AM
@Ausername Chat hasn't changed since 2018 FYI :P check the list on MM
But also, I read through that post and it was literally someone explaining their long and tedious code to me. I don't mind animated confetti (don't get why, but I don't mind it), but I don't even think they show the "final product" in the blog
 
Ye
(can't be bothered typing an s)
(But can be bothered typing that)
(And that, and this.)
 
8:44 AM
@Adám Site looks good Adám -- though with the initial config of the site (I get: light theme, low contrast, normal weight and normal style) there seem's to be a slight rendering issue. When scrolling the characters blur and that only happens with the mentioned config -- I'm on Firefox on Ubuntu (or possibly it's the screen or OS not capable of properly rendering the scroll with the very grey on top of less grey)
 
@JosephAdams Does it stay blurry if you switch any of the configs and revert back to the initial config?
 
yep, when I switch back the scrolling blur is also back -- I'm wondering if the blur is also happening if, say, only the font weight is switched to bold...I think the effect is lessened then, but maybe I'm just paranoid
 
Can you maybe check this: I suspect the blur only depends on Contrast being "low".
 
yeah, the blur is also quite pronounced if the font is italicized but the contrast is still low -- must be the contrast (being too low)
 
And I suppose, no blur if Contrast is "high", no matter any other settings?
 
8:50 AM
correct
 
Ah, then it is due to the CSS filter being applied. I'll look into replacing that with direct setting of colours instead. Thank you, you've been most helpful!
 
9:04 AM
Also, come to think of it: Adobe Source Sans Pro also has the curled ell, top (and bottom) serif on the 1 and a slashed zero mode -- no way to distinguish capital i and vertical bar though (if you prefer the serif on capital i vs lower case l the 'stylistic alternativ' mode of the font provides that option)
 
I don't have enough reputation in Stack Overflow, so I'll ask here.
 
@StackOverflow Python allows you to use negative indicies for indexing which will wrap to the back of the array being index into -- since h = [1] the -1th item (i.e. last) is 1
not sure why the author chose to do h=[]; h.append(1) rather than the (obviously shorter) h=[1]
 
@JosephAdams Does that apply to the libre Source Sans too, or only Source Sans Pro?
Oh, it appears to be there, though I don't know how to activate it.
 
@Adám uuuuh, I think the libre one too -- if I view the rendering sample from the GitHub page and set the 'zero' font setting I get the slashed zero
also works with the 'salt' setting for the alternate ligatures for capital i and lower case l
 
9:19 AM
oh, works for me though -- I did try without the trailing 1 though
 
Ah, font-feature-settings.
 
ah duh -- sorry my bad for not reading properly
unfortunately though it seems one can't have 'zero' and 'salt' set simultaneously
 
Yeah, just discovered that.
But maybe there's another value. Where did you find these?
 
@JosephAdams "ss01" 1,"zero" 1 works!
OK, I'll add this one.
 
9:28 AM
@Adám I found out about the alternate glyph variants for source sans pro via otf.show -- my initial thought is if there's a standard LaTeX font that has all the sought after features, because that's what I write in mostly (but sadly I didn't find anything appealing)
 
Are there any standard sans-serif LaTeX fonts?
 
yeah, there's the standard Computer Sans -- but generally I find serif-ed fonts more legible (but I also mostly read on paper/printout)
*Computer Modern Sans Serif I should say
 
Right. Serifs are definitely a plus in print (and for good reason).
 
@Ausername omg I remember when that question came out with the (still current) top answer -- all round was in awe of the beauty...even though it's just colours
@Adám also I'm a bit disappointed that one has to resort to the "special feature set" settings in order to have both the slashed zero and alternative glyphs for i and ell. Especially since that doesn't seem to be well documented behavious (at least I didn't find it quickly)
 
@lyxal Lol
 
9:36 AM
@JosephAdams Adobe even refused to make Ɪ default when made aware of this, claiming the tail on ℓ is enough to distinguish them. My site is trying to make the world aware that distinction isn't enough; we need context-free recognition.
 
yeah I agree -- would be nice to have the possibility of disambiguating without the presence of other characters. But I guess fonts, being at least a bit artsy, prefer form over function.
 
Seems very silly, though, when clearly the function of written text is so essential to society.
 
I can see the other side though: so far, everyone's been coping just about fine with the fonts, because usually we have context and also the brain is quite capable of discerning words even if misspelled (where I interpret misspelled to mean that it appears an ell or i was used in place of the other etc)
I think it might only be people that put much weight on absolute clarity regarding which letters are placed where, a subset of which are programmers, that care so much about the recognizability
 
Yeah, maybe.
 
@pxeger Flag?
 
I can't send images yet? I just want to say that I lost my record of 40 unread inbox messages, and I want to send the image of proof.
 
Click upload
Next to send button
Or pop us a link'
 
@Ausername Not available to low-rep users.
@StackOverflow Put your image on imgur, then paste the image url with no other text in the message.
 
Just send it to a draft post then send the link here?
 
10:02 AM
Yeah, oneboxing will take care of the rest.
 
How?
 
Simple, I didn't check the inbox, since I was keeping them since Day 1.
 
@Ausername oh, win printed to stdout with a trailing newline
Oops
 
0
Q: What is the link to your answer?

XwtekGoal Your task is to create a program taking the input that returns the link to your answer. You can find a link to your answer by clicking at the share button on your answer. Rules Standard loopholes apply except for searching for the answer online. In fact, you are expected to find the answer ...

 
10:07 AM
Who here knows some C?
@NewPosts What answer? The answer of this question?
 
@Neil really cool
 
0
A: Sandbox for Proposed Challenges

Alex briesWhat sequence? answer-chainingsequence Stackoverflow distinguishes users by an id that can be found in their their profile url. https://codegolf.meta.stackexchange.com/users/**ID**/user-name You can use this id to look up the number sequence of the last submitter on OEIS: The On-Line Encyclope...

 
10:33 AM
@NewPosts The "taking the input" is confusing me, since @Ausername didn't use input at all.
 
@StackOverflow You know that this is a challenge right? You should *by my understanding) fetch the link to your answer, not hardcode it
 
@cairdcoinheringaahing I fetched it. I don't know how to not hardcode it.
 
For example, sending a GET request to https://codegolf.stackexchange.com/questions/226635, processing the answers there and filtering out all except yours
 
Vyxal's ¨U is broken :(
 
10:59 AM
Not sure if I did fine. Oh well, they did say input.
 
@StackOverflow do you like using ARM V2 Thumb
 
@Razetime What is that?
I'm not interested.
 
@Ausername are you using it offline?
Because if you are trying it online, it's never going to work
 
:(
Why not?
 
Because I disallowed it online
Fully intentional
 
11:12 AM
What's the problem with it?
 
Yeah, I totally suck at doing this.
 
Same reason TIO disables all GET and POST requests: it's too easy to import malicious code
 
^
I asked here once upon a time whether I should allow get requests and the answer was no
I think it was pxeger who explained the reason why I shouldn't allow them
 
@cairdcoinheringaahing although, you could just run code locally, so I'm not 100% sure what the problem would be
 
Now that I think about it, which is bigger, "too" or "so"?
 
11:15 AM
I guess you don't want people using tio for spam and things like that
 
@StackOverflow TBF challenges are not a great starting point, and the challenge is vague around a couple of things (mainly the "input" or not), so its not a nice challenge to try when you're new
 
Mar 25 at 1:35, by Neil
@Lyxal probably to stop someone from turning it into a botnet
Turns out it was Neil
 
:(
Then I guess I'll delete that answer as well.
 
a̲̲̼̙̮͇͔͍̭͈̮̖̲̒͑͐̆̏̌ͬͤ̎͒̆͑̄ͦ͐́̈ͬ̎͟͠ͅa̢͒̃ͯ̒̏̇̏̒͊͐ͧ̆ͤ̏̃̓̐̊̿͏̫̠‌​͎̗͈̳a̛̤̣̩̺̲̼̩͕̍ͭͬ͐͢͢‌​͕͈̪̺͎͓̰̩̙a̸͑̔͋̊̐̿̏̎҉‌​͖̬̬̮̩ȃ̴̛ͪͨͦ͋̎ͭ̌ͩͮͥͫ‌​̱̮͔͔͚̰͙̞̟͚̣͉̬̱̲͖̺͇ả‌​̶̧̛̗̬͈̝͉͖̥͕̰̹̻ͣ̇̃ͬ͘͝‌​͎̥a̸̴̟̗͈̓̎̇̊ͩ̒̌̓̚͢͟͟‌​̮̰a̷̙̘̭͎̺̜ͦ͊̑͗ͥͨ̎̚͜͟‌​̺̟̮̜̜̳̻̲̰á̶ͭͯ̑̔̿͗̀̚‌​̧̨̳͚͍̞̜̺̕a͐̿̏́͌͊̀̏̀̚‌​̷̦͍̺͚̝̙͇̣̖̹̓̌̾̍͛́́́̕‌​͚͇a̡̧̧̓͆̃̽ͣͮ̅͐ͨͬͩ̿̒̀‌​̫̹̺̜͙͖̦͈̗̺̠͕̰̺ͅaͬ̆̽̊‌​̸̴͇͔͈̟̥̘̝̝͚̲̅͞a̛͋̂ͤͣ‌​̨̧҉̗͓̠̙̮̘͔̦͙̪̝̦͚͕̻̩ͅ‌​̣͇a͌ͤ͊̆̏ͬ͌̌͌͏͙̥͕͚͡͝a‌​̸̉̈ͫ̄̓͊̂̔ͭ͛̃̀͒ͧͥ́̉͡҉‌​̟͓̲̰̫͈̭̤ͅa̧ͯ̅̎ͤ̄͒ͫ̕͘‌​͚̤͖̩̣͎̥͈̬͖̩ͅa͛̽̔̽͆͗͛‌​̴̡̳̼͇̺̥̗̫͖́͞ą̇̈́ͬ̾͜͡‌​̨̼̠̭̪ă̶̢͖̺ͤͦͬ̏̓̀ͥ̃̋‌​̭̼̻a̴͎̼̰̩̣͍̥̳̺͌́͂̌͟͝‌​a̴̦̻̻̿̑͒̇͐̏́ͣ͊ͬͦ́͑́͘‌​̯̥̗̝̲͖̜̭̥̠̮̣̩̦̜̰a̓ͣ̑‌​ͫ̌͛̈́͌̈͂͗̈͆͗͌ͨ͒̑́҉̷҉̘‌​̰̥̼̖͚a̒ͮ̔̀ͩ͐ͫ͆ͤ͂̏ͥ̓̅‌​̷̡̹͍͔̪̱̺̺͍̆͋́͜ — Unrelated String 18 mins ago
@UnrelatedString _________ Indeed :P
 
@cairdcoinheringaahing I wonder how that's done.
 
11:18 AM
@StackOverflow zalgo.org
 
I̴̢̋͐̐̓͋̋̌̉̃̒͛̃̀̓͊̄͝ ̵̨̢̡͈̝̬̮͍̤͙̩̥̼̙̰͛̂̎̈̉͛͆͆͂̓́̃̿͘̕̚͠a̸̬̥̝͈̭̫̬̻͌͐̐̓́̉̎m̴̢̺̬̮̞̠̊̿̓͌̊̈́͑̈́̌́͊͜͜͜͠͠͝͝ ̵̡͐͗͐a̴̜̘̖̠̯͙̣̜̮͍̟̰͙̥͍͛̓̐̈́͜͠ ̷̨̗͔͇̱̱͇͉͍͈̺͎͆͗̎̑̿͆̀̀̅̾́͗̋̇̕͜͠͠ŝ̵̨̢̹̬͓̱͈̠͓̙͚̩̙͖̪u̵͎̱̺̗̻͍̹̖̾͋̽͠ͅc̷̢͈͇̪͓̺͙͖̰̰̘̘͔̹͍̐̂̒̀́̓̀̓̈́̈́̐͋́k̷̮̩̆̊͒͋y̶̤̤̞̺̱̲͈̯̔̂̄̒̑̈̃̇̓́̂̿͊̕͜ ̶̛̻̈́͐̀͋̉̿̽͒̆̃̔́g̷͇̋̄̄̆̋̉̏́̕͠ȍ̸̢͉͍͓̠̟͔̭̪̦̻͕͍̣͍͎́̄̓̐͌̌͂̒͘͜͠l̷͎̥̱̼̃͗̋̓̏̐̀̓̇̈́̈̚͝f̸̡̟̟͖̭͓̯̖̖̞̹̲̃̉̔̃̈́̍̎̔͑͐̾͂̕͝͝ͅę̵̞͚͈͕̖̞͎̆̄̄̾̋̅̿̿͗͂̇̅̕r̸̫͎̘̞͈͙͗̈́͘͠.̵̢̳̹̖̩̈́͊͋̔̾͐̓͒̓̏̽͐̑̀͒̀͛
 
Please mess around with that in the Sandbox, not here :)
 
Oh, sorry. But I know I am.
 
Hello! I have this sandbox post, would it be possible to get some feedback from here?
0
A: Sandbox for Proposed Challenges

ophactConic Sections (simplified) code-golf math characteristics Given the equation of a non-parabolic conic section, output its characteristics. Spec This assumes prior knowledge of hyperbolas and ellipses, as well as their characteristics. (This includes circles; they are a special case of the elli...

 
Sure
 
11:33 AM
I'm trying to separate the integers from one another. Like making `18` into `1` and `8`. How can I do that?
I don't feel like writing this on Stack Overflow.
 
Which language?
 
If it's JavaScript, then n=>[...''+n]
I don't know C though
 
You know anyone who knows C?
 
No, sorry
 
11:34 AM
Then I'm asking this on Stack Overflow.
 
27
Q: C: how to break apart a multi digit number into separate variables?

mugetsuSay I have a multi-digit integer in C. I want to break it up into single-digit integers. 123 would turn into 1, 2, and 3. How can I do this, especially if I don't know how many digits the integer has?

 
Hmm, that's kinda old
 
Yeah, I completely forgot about it'
 
How do you go from -1 = 1 to 1 = 1? By squaring I'm assuming?
 
11:37 AM
@ophact Yep
 
Nice idea. I like it. Just saying, you would have to reverse proof it, and since exponentiation is almost always required, you would have to take the square root (which is always positive.)
 
Btw, how's my sandboxed challenge?
@Ausername yes, the first seemed to be a duplicate
 
Yours looks pretty good.
 
So ready to post? I'll leave it there for one more day to comply with rules but then I'll post it. Thanks!
 
11:40 AM
Yep! Do you think mine's ready to post?
 
yes
 
@cairdcoinheringaahing What's funny is that each number is in an array since I'm trying to answer this. I set the credit card's numbers into an array, and still stuck at trying to add the numbers together after doubling the numbers at the odd indices.
 
Ok.
 
0
Q: Make a "proof" generator

A usernameIntroduction How to prove that 1 = 2: 1 = 2 2 = 4 (*2) -1 = 1 (-3) 1 = 1 (^2) You can just multiply both sides by 0, but that's cheating. This is just bending the rules a little bit. Your challenge: Write a program/function that, when given two integers, outputs a "proof" that the two are equal...

 
That thing is instant.
'night everyone
 
11:47 AM
\o
 
@NewPosts I don't get the 3rd proof. (-3)? Does that mean you subtracted or what?
 
@NewPosts Isn't this just equivalent to "Given two integers, find the steps to a pair (-a, a)"?
@NewPosts Added that you can't divide by zero (@Ausername)
 
Sadly, $-3 + 1 = -2$.
 
\$2 - 3 = -1\$. The (-3) means "Subtract 3 from each side"
 
\$-1 = 1\$? Then please proof it.
Because I don't understand.
 
11:52 AM
Essentially, it's a false proof. We know that 1 = 1. Therefore, if we can derive that 1 = 1 from a statement a = b, we can state that a = b. Take 1 = 2. If we can manipulate this into the form 1 = 1, we can "prove" that 1 = 2
 
I get it now.
 
12:49 PM
CMQ: Are there any "criteria" for you that if a post meets, you're almost guaranteed to upvote it? For example, I will always upvote questions I answer, answers whose methods I use in my answer and answers that outgolf one of my answers
 
in a language I approve of that uses a feature specific to that language, such as let/cc in racket
 
@hyper-neutrino i've left you some criticism on this deleted answer codegolf.stackexchange.com/a/55375/2537 but i don't really care if you respond, i'm just letting you know
 
@cairdcoinheringaahing Lately, I do the same. Beyond that, I tend to upvote answers that look like they were hard to write, and wrong-tool-for-the-job answers.
 
@MitchSchwartz Note that we do allow answers that use a different, albeit longer method, so long as they show an effort towards optimising their score. Even if that is trivial golfs, such as removing whitespace, that's fine as it shows some effort towards golfing
 
well i think my stance is pretty clear on that
 
12:58 PM
Still, it'd be a shame if such a cool answer went to waste. Here's a basic golfed version if you want it
 
@cairdcoinheringaahing I almost always upvote answers with clear explanations or that find a new way to approach the problem (even if it's not optimal)
 
@MitchSchwartz Sorry, I forgot to include a link to the policy. This is the meta consensus that I am following right now.
You are correct that my response is canned - it's better for me to have a consistent response each time than go to the effort of writing a new one each time and potentially miss information, make typos, forget about certain things, etc.
 
1:13 PM
Still struggling with the Luhn algorithm.
 
i have no interest in any of that. it was just to inform you
 
Okay. And I am just informing you that nobody's demanding you "appease the moderators" - I'm acting on a community member's flag based on the community's collective consensus.
 
@StackOverflow In what way?
 
C.
The only language I know most.
 
consensus is a funny word. i'm happy to not follow it
 
1:16 PM
@MitchSchwartz the rules require you to follow the policies set out by consensus
 
@StackOverflow btw this month you can get a bounty for learning Red
 
@MitchSchwartz If people can just decide to not follow consensus, whats the point of having any policies, rules or consensus in the first place?
 
@Wezl I'm already learning Python. I'm not really interested in bounties, but I might try.
 
@cairdcoinheringaahing any answer that pioneered a non-obvious method of solving the question, that pioneered a golfing trick, funni jokez, or, like Adám says, wrong tools for the job requiring a silly amount of effort
 
The only reason that this place has significantly higher quality contributions is by strictly enforcing community-decided rules. Additionally, as demonstrated by the program I provided for you, it generally isn't hard to follow these rules
3
 
1:21 PM
@cairdcoinheringaahing If it's challenging enough and pretty good, there's a chance I'll upvote it.
 
@StackOverflow don't worry. Being a good golfer is not a necessary skill on this site ;D
 
I change my mind. I'm not learning Red. I'd stick to Python or C.
 
May I ask why?
 
@user All I'm going to say is that I don't like going link after link after link.
 
lol is it the website
 
1:25 PM
Yes. :(
 
I haven't actually used the website (can't), try rebol.com
 
Also, I'm not interested (at the moment), so I will play around with what I know.
 
if i want to golf it i'll golf it. if i don't, i won't. it's pretty simple really. you have no say in the matter lol
 
@MitchSchwartz just do it so that the rest of us can see the answer :( (btw I disagree with their decision but oh well)
 
If they really don't want to, they're more than free to leave it as it. If you change your mind @MitchSchwartz, you are more than welcome to use my golfed version of your code
 
1:37 PM
yeah i'm not convinced that seeing deleted content should be restricted only to moderators
 
It's not lol
 
a toggle for regular users could conceivably work
 
@MitchSchwartz It's not. Anyone with 10k reputation can see it
 
oh, then i'm misusing the word "moderator," i have to pull up the chart to see the proper name of it i guess
well the chart says "access to moderator tools" so i guess i was just equating them
 
"Moderator" kinda does work, given that the privilege is called "moderator tools". "Diamond moderators" are the most unambiguous way to refer to HN and the other 3
Personally, I can see the merits of lowering the threshold for seeing deleted posts, but it comes with the ability to cast VTD/VTuD on those posts, which I'm not in favour of lowering the rep requirement
There's nothing majorly wrong with users with, say, 500 rep, from seeing deleted posts most of the time. There is an issue with those users being able to VTD posts, and the two abilities are "packaged" together
 
1:43 PM
personally the most useful privilege-restricted features i can think of at the moment are seeing deleted content and seeing the up/down distribution; these don't even have to be thought of as having anything to do with moderation
 
@MitchSchwartz this userscript lets you see the vote distribution without having 1k rep
 
oh, huh, that's cool
 
Damn it, ninja'd by my slow wifi :/
Hey @hyper-neutrino, remember my unclosed flag bracket? I finally closed it :P
 
 
1 hour later…
2:56 PM
If I'm submitting an answer as a function, what needs to be included? I don't remember where the relevant meta post is at.
 
@NewPosts STATUS
 
you beat me to it
i just realized too
 
@AaronMiller If it's recursive, you should include the f= bit if relevant. Otherwise, just include the function body
 
RRREEEEDDDWOOOOLLLFFF!!!!11!!
 
@hyper-neutrino Give it a go, I;m not sure if Redwolf actually whitelisted me
1
Q: Tips for code golfing in Vlang

AphrimWhat tips do you have for code-golfing in the V lang? I have been trying to find ways to shorten codegolfs on there but can't find any resources on it. Do you guys have any? Please suggest tips that are at least somewhat specific to the V programming language (e.g. "remove comments" isn't helpful...

New post for y'all ^
 
2:58 PM
@NewPosts STATUS
 
@RedwolfPrograms fix ur bot
 
We just the meta discussion as well :P
 
ah
 
3:03 PM
@cairdcoinheringaahing i heard this was just a string replacement into C
 
hmm, kinda strange that i chose "enumerate" as the name of my function since it doubles with a built-in, although it's an appropriate name otherwise
 
@cairdcoinheringaahing I remember seeing V on HackerNews a while ago, and it got ripped on pretty hard for not living up to any of its claims. Anyone know if it's any good nowadays?
 
Any feedback? Considering posting today/tomorrow if not
 
have you read my comments?
 
I have, but forgot to respond due to the conversation here about it yesterday. In short, allowing answerers to choose +, - and S could lead to some interesting answers, but is far more likely to result in someone finding a combination which completely trivialises it (even with the updated spec)
Note that my update means that none of addition, multiplication of subtraction are valid submissions anymore - it's now significantly harder
 
3:16 PM
ah ok
> Your programs must each exhibit a maximum of 1 of these 9 behaviours
But each program has to exhibit at least 1, no? So should this be "exactly 1" instead?
or did you mean to change "maximum of 1" to a different number?
 
@pxeger Indeed it should, edited
 
it seems like finding an operation that is neither commutative, anticommutative, nor noncommutative might be extremely difficult
 
@pxeger For your trivial brainfuck NOP challenge, is "any snippet containing [ or ] is undecidable" true for brainfuck? It's true that given an arbitrary brainfuck snippet with [ or ], it's undecidable whether it halts or not, but it is decidable for certain brainfuck snippets (e.g. [-] will always halt)
@pxeger Yes, but not impossible, which is the meat of the challenge
 
have you considered reducing it to only the first 6, maybe?
 
I have no problems with very difficult challenges (in fact, I like to aim for as long a gap between asking and answering when I post challenges), but there is this version designed to be easier in the Sandbox as well
 
3:23 PM
@cairdcoinheringaahing but this is trying to be "trivial". It is possible to get arbitrarily close to all programs by limiting your input set (e.g. if you limit the program to, say, brainfuck programs of length less than 1 trillion, then theoretically you could manually determine the halting behaviour of each and code that into your detection program)
 
@pxeger Yeah, I'm not saying you should modify the challenge, just nitpicking at the wording :)
 
I just meant undecideable for a truly arbitrary program
@cairdcoinheringaahing maybe change all the = signs to ?
 
Seems like might also apply here - the definition of "trivial NOP" is very reminiscent of the balanced program definition often used for brainflak, albeit with brainfuck characters instead of brackets
 
yes I thought about that, but there are just too many tags that might fit (I was also considering ) and I can only choose 5 (well, 4, because code-golf is a given)
 
Okay, so I'm filling out the moderator survey right now, and there's a question I'd like to consult y'all about: "What one non-software thing can the Community Management team do for you or your community to improve your experience as a moderator?"
 
3:34 PM
@cairdcoinheringaahing I feel like for really difficult challenges like this (and also, e.g. complement a POSIX regex), we need a new scoring criterion. The challenge lies largely in doing it in the first place, and [tag;code-golf] seems a bit rubbish. Something closer to , where it's really just the first person to post that deserves to win, and if they spend 10 hours finding 9 operations and proving 81 things, then someone else comes along with their superior golf knowledge but[1/2]
 
Anyone have any outstanding things that we could benefit from? I can think of some things that would benefit me in terms of mod tools but that's software stuff.
 
[2/2]...who's not as good at maths and spends 15 minutes halving the byte count or translating it into a shorter language, they don't really deserve to win. This is a related problem to the "let xnor find the optimal modulo chain and copy him for all other answers" issue, and to the (albeit partially solved) "golfing languages suck all the fun out" problem
@hyper-neutrino ability for the community to change (some) site settings without needing to wait for a CM/staff member through some kind of moderator-guided consensus? oh, "non-software"
 
@hyper-neutrino They've been getting better at this more recently, but listening to us when we want to escalate something to them, even if that's just a "Ok, acknowledged" response
 
@hyper-neutrino stop making drastic (or even seemingly small) changes without prior consultation (e.g. most recently, the font changes) (although they weren't necessarily community-related, more development)
 
@hyper-neutrino One thing that might be a nice "one off" event could be for you mods to "sit down" for a couple of hours with a CM and go through this list of old bugs/FRs with no status tags and sort them into what is and isn't happening
2
 
3:38 PM
oh god that's a lot of stuff to go through...
 
yup
 
It's more of a house-cleaning thing (and tbh I'd be happy to do it and compile a list), but ¯\_(ツ)_/¯
 
I'll look into that, thanks. I think I can probably clear at least half of those on my own; it seems some already have accepted answers anyway and can be completed.
 
Yeah, a lot of them seem to be mostly that a mod forgot/didn't bother to tag them
 
Okay, thank you both for the suggestions :)
 
3:42 PM
@hyper-neutrino Oh, just thought of something: the How to Ask page/Tour
Not sure if that counts as "software" or not, but it'd be nice to be able to have more control over that
 
Uh, which how to ask page, this one? or are you referring to something else
 
yeaahhh.... that's just part of the list of abuses of the SE software
@hyper-neutrino I'm guessing this one: codegolf.stackexchange.com/help/how-to-ask
 
@hyper-neutrino More the entire help center, but ^ specifically
 
Oh I see
Anyway, I feel like that'd be better for a meta post requesting the ability to change that (since I can edit the "what's on topic" but not "how to ask" page). I just realized the question is specifically asking about what'd improve my experience as a moderator, not what'd benefit the community, although they're synonymous for the most part
 
3:46 PM
Responding to escalations even if just acknowledging that they'ev seen it and will get to it eventually would be nice in both regards, though
 
@cairdcoinheringaahing btw i understand that your motive is probably to be helpful, but i don't understand why i would want to open that link, because (1) if it's trivial, then it's devoid of interest, and (2) if it's nontrivial, then why would i want to see your ideas before golfing it myself?
 
It is your program but with space indents as tabs, unnecessary whitespace removed and a sum(f(i) for i in range) as sum(map(f,range)). It might be devoid of interest (tbh I don't really find Python golfing interesting at all), but its golfed which you can use as a "look, it golfed" program, while still keeping your original program and explanation
 
and that is precisely what i don't want to do
 
the idea is even if it's not very hard it still takes a couple minutes, which you've told us you don't want to spend, so caird did it for you because we still want it on the site (it's interesting)
 
Yeah, that's fine. Just thought I'd share it in case you did decide that you were willing to sacrifice a bit of "golfing triviality" in order to keep your answer undeleted
Besides, as much as I dislike golfing Python, it's fun to keep my skills somewhat sharp, so it was no trouble to do, even if you don't want it
Also, is there a ? For example, for this discussion
 
3:53 PM
 
I'd go with status bydesign probably lol
fun fact: there is a
 
> Indicates that the site developers were able to replicate the behavior reported, but cannot address its cause at this time.
aka [meta-tag:status-oh-yeah-wtf?]
 
anyway, you would agree that what you described is the antithesis of competition, right? do some trivial stuff and slap a "hey it's golfed" label on it lol
 
since anything that isn't reproduceable should be tagged no-repro, it seems that anything else should be tagged either planned if they will do it later, declined if they won't, bydesign if it was reproduced but isn't a bug, etc.
I feel like the reproduced tag is unnecessary and everything should be using a more specific tag to indicate progress status so people know what's actually happening with it :p
 
@MitchSchwartz Perhaps. But I think its a necessary "evil" to keep posts on the site somewhat on topic and high quality. Some people will outright refuse to meet this evil (aka, what you're doing), and others will accept it. Ultimately, we can't have a site that accepts any answer that may be "interesting" if it doesn't make some effort to work towards a scoring criteria, as that way lies VLQ
 
3:57 PM
@hyper-neutrino it helps devs/PMs know that they don't need to bother trying to reproduce it, because someone else has already looked at it, but no further decision has been made (i.e. it's basically still but definitely not a fluke/made-up issue)
 
@MitchSchwartz well, the alternative I see is that only the winner gets to keep their answer. I've seen ungolfy approaches; for example, Jonah's J answer to my challenge yesterday was very long compared to the other challenges because rather than doing the straightforward (and rather trivial, for golfing langauges) O(2^N!) brute force, he took an efficient approach, but he still golfed it down.
The point is that we want to allow interesting ideas and approaches, but submissions should still at least try to optimize for the winning criteria
 
Especially given that putting in the bare minimum counts as trying
 
@pxeger that's fair. I personally feel like if it's been reproduced after reviewing it should either be declined or planned if they will eventually do it. But I guess there's no harm in having an extra tag even if just for a few rarer cases
 
well the bare minimum makes it harder to read, which really is counterproductive to why i posted it. i'm fine with it being deleted though, i said that from the beginning. it's just weird to cite "competition" as the reason, because competition is exactly the reason i didn't golf it, because i don't believe it can possibly win, so what's the point
 
@hyper-neutrino Well, it has a grand total of 10 questions on MM and 4 on MSO, so it's very rarely used
 

« first day (3776 days earlier)      last day (1068 days later) »