The Nineteenth Byte - 2014-03-14

« first day (1140 days earlier) ← previous day next day → last day (4000 days later) »

hosch250

00:59

Anybody here?

Doorknob

@hosch250 No, but a doorknob is here

hosch250

OK, @Doorknob

I was just on SO, and I was looking at their tag wikis.

Doorknob

okay

hosch250

They have very detailed wikis, and I was wondering if it would be alright if we 'borrowed' them for our use.

Doorknob

0

Q: Create a simple 2048 game clone

2048 is an incredibly fun and addictive game in which the goal is to create a tile with 2048 on it. Here's a short description of the game: Pressing an arrow key will slide all the blocks in the stage in that direction. For example, if x represents a block, and you pressed the up arrow in thi...

@hosch250 hmmmm, for which tags?

hosch250

01:03

Mainly the languages - C++, Python...

Doorknob

but we already have our pre-baked template for those

and why would people on this site need to know about the languages?

hosch250

I don't know, it just seems a little lame when you review them.

undergroundmonorail

Language tags only really make sense for restricted-language questions, which are pretty rare anyway.

hosch250

Just wondering. I did not make any changes.

@undergroundmonorail Yes, but we have several anyway.

undergroundmonorail

Yeah

luser droog

01:08

There was some discussion on meta about removing the language tags. Particularly the ones whose one Question is the Tips question.

hosch250

That would be reasonable.

5 hours later…

hosch250

05:52

I hope I haven't made an enemy of anyone - I was just the victim of serial downvoting.

luser droog

06:43

No idea. It looks like it was caught and reversed (mostly).

luser droog

07:00

translated the Lukasiewicz interpreter to C.

gist.github.com/luser-dr00g/9542998

5 hours later…

Doorknob

12:08

6

Q: The race has started. Are you running?

PCG.SE is going on a mission. And so the race begins. For Code Review!! -- syb0rg 2 days ago Let's see: Programming Puzzles and Code Golf Voting & Activity PCG.SE Voting and Activity Chart Compare to: Code Review Voting & Activity CR.SE Voting and Activity Chart ...

discussion beta-progress voting

the race is on.

dramatic music

1 hour later…

Gareth

13:15

The wheels and the exhaust pipe are in danger of falling off and you want to drive faster?

Rusher

13:40

Is King-of-the-Hill a good tag for multiplayer challenges involving MORE than one-on-one battles?

For example, a simulation involving lions, tigers, bears, and 50 different wolf classes submitted by code golfers?

2 hours later…

Rusher

15:16

Nobody answered codegolf.stackexchange.com/q/23775/18487 . Do I extend the time limit, delete it and weep, delete it and re-enter to sandbox, or simply let it sit there? Not sure what to do.

Howard

15:48

@Rusher Therefore, we normally do not do any time limits on code-golf. It may take a few days until someone finds the time to tackle a more advanced puzzle and golf a solution.

@Gareth I think we already lost the exhause pipe some miles ago, didn't we?

hosch250

16:07

Is this really low quality?

codegolf.stackexchange.com/review/low-quality-posts/9831

There keep being posts that seem fine to me in the low quality queue, so I have just been skipping a ton.

I figured I might as well ask.

Rusher

16:45

Thanks, @Howard. Someone already convinced me that time limits are not the best restriction to add to a challenge, but I had already posted 4 puzzles prior to that. I hate changing the requirements after posting, but in this case I think I'll make an exception and remove the time limit completely.

hosch250

@Rusher I typically wait a few weeks, then wait until the question has received no action for about a week, then accept my initial answer.

Then, I change my answer if a better one is posted.

Jonathan Van Matre

@hosch250 In addition to user flags there is an auto-flagging mechanism on short posts for low quality

Unfortunately, that creates a lot of false positives when you have code-golf one-liner answers

Jonathan Van Matre

17:15

FWIW, I have arrived at the opinion that any answer submitted in GTB is low quality because the user submitting them is apparently the only person on earth who has a working "compiler" for it. And when I questioned him on that point, there was no response.

5

In contrast, anyone can download RebMu or GolfScript.

Jonathan Van Matre

17:29

Oh, also @hosch250, if you hit skip it will stay in the review queue. If you think something looks good, go ahead and vote it so. It avoids duplication of effort. Someone else can always flag it later if necessary. I reserve Skip only for cases where I genuinely can't make up my mind.

hosch250

18:13

@JonathanVanMatre OK, thanks. I was thinking somebody was maybe flagging randomly.

Peter Taylor

18:24

In case there's anyone who visits chat but not meta, I'm planning to post meta.codegolf.stackexchange.com/a/1268/194 shortly unless anyone raises a comment which needs addressing.

Rusher

I am adding my comments

Gareth

@JonathanVanMatre I've downvoted and will do so for every GTB answer I see until a compiler becomes available. Maybe that will motivate him to make it available to all, though I doubt it.

@Howard Well, it's definitely a lot noisier than it was, so you may be right.

luser droog

18:43

@PeterTaylor I for one have been doing just that. Thanks for the heads-up.

mniip

19:30

http://codegolf.stackexchange.com/questions/17907/#comment50757_24181
Well that was actually kind of harsh...

But I cannot say @Gareth is wrong, either...

Rusher

I found what looks like source code for a GTB compiler... but have no idea what to do with it

I'm sure you guys actually visited Timtech's website before bashing his language

Gareth

@mniip It's not like he hasn't had time to make the compiler available. I posted a meta question which referred to it: meta.codegolf.stackexchange.com/questions/988/… and he was referred to it by victor on this post codegolf.stackexchange.com/questions/17331/…

@Rusher Yes. That meta question I referred to above was about my difficulty in getting any kind of working compiler for the language and having to the translation by hand.

I followed the link that comes with that weird encrypted compiler source and the program that may or may know what to do with it doesn't seem to be available anywhere.

Rusher

I thought that it was possibly a TI-89 or TI-88 program

but I don't know TI-88/89 so I just waved my hands at it

Gareth

In the end, after the translation by hand, the answer to the question referred to above threw an error.

I didn't downvote that answer even though I think he's cheating, I asked that question on meta to get some kind of idea what other people think.

But since he's still posting answers which are essentially unexecutable gibberish unless a compiler is provided, I'm going to downvote them now until he does something about it. When he does I'll retract all the downvotes.

Following the rabbit-hole down from his website leads eventually to this page instant-exe.software.informer.com/download which says that the software is not available. I already downloaded the TI emulator to run the compiled program - the problem is that noone on this earth other than TimTech can compile the programs.

mniip

19:56

@Gareth "software informer" - get out of there quick

But yes. I guess, until there's a compiler/interpreter for the language, things he writes might be considered invalid programs

Rusher

@Gareth The way I see it, your ask-first-downvote-later approach is the correct way to handle the situation, because it means an answer is correct until challenged. The burden falls to the answer to provide proof that obscure code actually works. The alternative would be to assume all answers are incorrect until proven correct. I think that is unnecessary for 95% of posts. Nobody needs to point out how to compile GolfScript or Python every single time.

But if someone challenged me to provide a compiler for my Python code, and I failed to do it, then I probably deserve a downvote.

intepreter*

(don't laugh)

mniip

@Rusher The issue here is that he's been asked to provide the compiler

Not exactly directly, but I suppose he checks meta often enough.

Rusher

@mniip I don't understand your point. I just said that I agreed with Gareth's approach

mniip

Oh, then, well, nevermind.

Rusher

I was only hoping to point out what would happen if the alternative approach were taken

Answers would be pretty bloated in that case

TimWolla

20:11

@Rusher I began writing a solution, but aborted

Rusher

@TimWolla Someone mentioned that negation of the graph can turn a longest distance problem in a shortest distance problem. They were wrong, though

TimWolla

@Gareth Retracting them is hard, unless he edits them.

But I think the same about that GTB thingy

Rusher

I read that shortest path is something like O(EdgesVertices) while longest path with cycles is not even solvable in polynomial time. So I don't understand how you could negate one and get the other

It was @JonathanVanMatre . Care to explain " Negation of the graph enables solving for the longest path by solving for the shortest." or at least provide a link to an explanation?
"

Nevermind, I understand now. Your statement is true IFF a longest path exists. If there is no longest path, then there won't be a shortest path either. Your other comment makes sense now too. Sorry!

Rusher

20:32

For anyone that is working on Spanish Verb Conjugator, the OP modified the scoring significantly by adding a bonus score modification that is potentially worth the amount of unusual verbs in the entire Spanish language.

mniip

Has wolfram upgraded their problem generator, to generate PCG questions?

http://codegolf.stackexchange.com/q/24216/7162
:P

The Q seems quite a bit underspecified by the way.

TimWolla

I just wrote a GTB compiler for myself

6

A: Shrink a number string

GTB, 25 24 26 characters 0→X`_[@_="1":X+1→X@_≠"]~X Explanation 0→X Put 0 in X (remember GTB adds a colon at the front of every program) `_ Input into string _ [ Start loop @_="1" Test if case is finished, if so X+1→X @_≠" If not empty, keep looping ~X After finished, display X

This one definitly does not work as stated in the question

And it even is accepted

hosch250

@TimWolla It was accepted because it is the shortest.

TimWolla

Yeah, but it is not valid

If I skip parts of the question I can get the shortest as well

Jonathan Van Matre

20:50

@Rusher No worries. I could have explained it more precisely but I was assuming I wasn't saying anything you didn't already know.

I liked Peter's suggestion to turn it into an Esperanto challenge

Doorknob

@TimWolla this is a very big problem then. :/

Jonathan Van Matre

no irregular verbs makes for a much more objective target

@mniip I mentioned the inaccessibility of the compiler to him in a comment here: codegolf.stackexchange.com/questions/23362/…

He answered the other part of my comment, so he obviously read it and chose to ignore the query about the compiler

Peter Taylor

Hmm. I thought I would test my theory that Python was a good candidate to win my practical number contest. Even using a good algorithm it's reaaaaaaally slow. My original reference implementation in Java takes a couple of seconds to compute the millionth practical number; the Python port takes just over a minute. I might have to rethink the performance criterion.

Rusher

Did you post the python code?

Python and Java are the only two languages I really enjoy reading

Jan Dvorak

@Rusher not Ruby?

Jonathan Van Matre

21:00

Interesting. What makes a language enjoyable to read?

mniip

@Rusher You should try haskell. All haskell code is just pure elegance and beauty, somehow implementing precise mathematical terms in just a few words.

Peter Taylor

@Rusher Correctly identifying radical changes is an interesting heuristic problem, but until the Diccionario de la Real Academia Española is available in a parseable format it would be a lot of work to produce a full list.

Jonathan Van Matre

@mniip That's precisely what convinced me to take up Haskell. There's an elegant logic to it.

Peter Taylor

I haven't posted the question yet: it's still in meta. And I'm not going to post an answer until other people have had some time to think about it.

Rusher

Well, my opinion of what makes a language enjoyable to read is knowing all the little gotcha's beforehand

Like String pooling in Java

Jonathan Van Matre

21:03

so Fluency == Enjoyment

Rusher

Yea, kind of like reading a book

mniip

You should try brainfuck, there are no gotchas

Jonathan Van Matre

I think some esolangs can contradict that. Shakespeare is fun to read even if you have no clue what is going on.

Also Gertrude, but Gertrude is harder to write unless you roll yourself a toolchain to pluck sentences with the proper lexical ratios from a corpus of text. Which I am working on.

TimWolla

@mniip Just building a solution in Haskell :)

Rusher

@PeterTaylor I like the way you added a performance requirement to a code-golf challenge, but I don't understand how you got away with it being so vague after I have seen other questions with similar requirements be killed because of it. I feel like your rep and your ability to speak gud English has something to do with it

mniip

21:06

@TimWolla It's hard to golf in haskell, mainly since words are huge. All function names are damn giant

TimWolla

@mniip For the fun of it. Even if I don't win

Rusher

It's at least fun if you beat the other Haskell users. I like winning "smallest Python answer"

TimWolla

A bit sad however, that this did not get a single upvote:

0

A: Write a Playfair encryption program

Haskell - 711 Demo: [timwolla@/data/workspace/haskell/PCG]ghc pcg-23276.hs [1 of 1] Compiling Main ( pcg-23276.hs, pcg-23276.o ) Linking pcg-23276 ... [timwolla@/data/workspace/haskell/PCG]./pcg-23276 "Stack Overflow" "The cat crept into the crypt, crapped, and crept out again." SIR...

I mean: It is better than the Java solution

mniip

you should use 1 space for guards and where-s

TimWolla

In terms of length

@mniip It is a single Tab

mniip

21:08

I think you can also chain whered defintions with ;

@TimWolla SE coerced it to 4 spaces

TimWolla

@mniip Thanks about that where and semicolon

I just learned about Haskell in my first semester at university :)

Rusher

Isn't there a tips a tricks page for this kind of stuff

Jonathan Van Matre

Each language has a tips question

Most languages, anyway. Some do not.

Peter Taylor

@Rusher If you look at online judges like SPOJ they enforce a time limit by their server. I don't want to try setting up (and securing!) an equivalent system, but the concept of a reasonable time limit to prevent brute forcing isn't entirely novel. In fact, it's even been used on this site before.

Jonathan Van Matre

"Tips for golfing in X"

Rusher

21:11

5

Q: Tips for golfing in Haskell

What general tips do you have for golfing in Haskell? I am looking for ideas that can be applied to code golf problems in general that are at least somewhat specific to Haskell. Please post only one tip per answer.

code-golf tips haskell

TimWolla

Just found it as well

mniip

@TimWolla Sadface. I tried point-freeing your u function and it turned out to be the same length

u=((ap((,).(`div`5))(`mod`5).fromJust).).(.k).elemIndex

TimWolla

What does ap do?

mniip

ap

TimWolla

I found that as well, but I never needed Monads yet

We did not learn this in university as well

In fact we learned almost nothing from our Prof

mniip

21:14

Basically takes a wrapped function, wrapped argument, applies them, and returns a wrapped result

Haskell is easy once you understand what's happening.

As a cool exercise, one can try reimplementing Prelude

TimWolla

@mniip I did it with some of the functions

But I think this only touches the Surface

mniip

I mean the whole Prelude, it's no big deal to implement head or foldl (intersperse is not prelude)

TimWolla

I'm curious, is there a way to shorten that lambda function here:

p=map(\x->if x=='J' then 'I' else x).filter(isUpper).map toUpper

Peter Taylor

Well, no: implementing head is something which would be mentioned in the first lecture, and foldl probably in about the fourth.

mniip

Well, pointfree gives this:
p=map(flip if' 'I'=<<('J'==)).filter isUpper.map toUpper

Peter Taylor

21:19

(Not that I know Haskell; but the first programming course at my university was in SML)

TimWolla

@mniip That's certainly better

mniip

@TimWolla But since you're in prelude you have to squeeze if' in

and I think that will ruin the whole optimization, as it will be larger than the original lambda

TimWolla

@mniip What package? It is hard to find on Google

Because Google ignores special characters…

mniip

Use hoogle. if'is not defined anywhere. It's just a thing pointfree uses to escape if-then-else

if' x y z=if x then y else z I think

TimWolla

I see

Well, I am using if at more places

Maybe it is worth it

mniip

21:36

Ugh, has anyone ever seen a valid GTB answer, even?

Everything I see is underimplementing the spec or doing things just wrong

TimWolla

If I had a compiler that works correctly

My current one (in CoffeeScript) has some flaws, but it makes the code at least readable

mniip

I'm not good at TI-basic either (at all), but I've seen enough languages to understand which does which

TimWolla

Yeah

mniip

Meanwhile I actually added some description to golflua's page :)

Rusher

That looks pretty scary

mniip

21:47

The language? Or my site?

I just like the vtconsole theme

Rusher

What scared me was realizing that I could write my own compiler that simply compiles my input to the answer of my choice.

mniip

@Rusher There's a rule against that: the interpreter should exist before the question is asked

Rusher

Ah, I see. I guess that invalidates quite a few answers posted on old questions using newer languages

(crosses fingers hoping he doesnt ask me to find an example of one)

If I ask a question, am I seriously committing to downloading that golflua thing and checking for correctness?

Jonathan Van Matre

Still, one could write a GolfMathematica today that gives a Unicode token to every one of Mathematica's 5000+ functions. And use it on every new question from tomorrow onward.

mniip

golflua, though, still that awkwardness of lua... Though it's better than perl at operators

Jonathan Van Matre

21:51

Not that I'll be doing that. It wouldn't be sporting.

mniip

@JonathanVanMatre Contact me when you manage to convince Wolfram to give you the source :)

Also if you're so scarred of unicode you totally should check out Sclipting

Jonathan Van Matre

Don't need the source, just roll a preprocessor.

That's all that several languages seen on this site are (including GTB) - an economized syntax macro for another language.

mniip

Preprocessors aren't fun. See Timtech's stuff

Jonathan Van Matre

Precisely.

I'd rather learn a terse-by-nature language I will actually use (e.g. APL, J) than an unnecessary translation layer for a language I already know.

mniip

Well that makes sense

TimWolla

22:00

244 Rep left for 1000

mniip

495 till 4k :P

« first day (1140 days earlier) ← previous day next day → last day (4000 days later) »

Transcript for

Mar '1414

The Nineteenth Byte

The Nineteenth Byte: General discussion for codegolf.stackexc...

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2025 Stack Exchange Inc; legal

mobile