« first day (8 days earlier)      last day (5131 days later) » 

00:58
@Juan: I fixed the UserScript.
Anyone else around here?
01:10
@George, the latest version is still at 0.16, correct?
Yes.
I fixed the comment bug, though.
Ahh, good :-)
0
A: UserScript to fix the code-size argument once-and-for-all!

George EdisonQuestions to consider: Here are some things to be considered and discussed as this little script grows in functionality. What languages do we want to skip? Someone on the chat mentioned that there are some languages (like HQ9+) that should be skipped and not counted when determining the smalle...

It was tiring having to disable the script in Chrome just to make a comment
@Bass: Sorry :( It was my fault.
It seems like the version of jQuery the script uses conflicts with the one on the site.
...so now the script just uses its own version internally so that it doesn't conflict.
01:12
Interesting, would have never expected a conflict in versions to cause that
Well, the site itself inserts some special functions into the jQuery object itself.
...then when my version gets loaded, it's missing those custom functions.
Ahh
That makes sense actually
Any ideas for improvements?
I want this to be a useful script that does as much as it can to make the site better.
Remove one character off the ending count, as it is always off by one
Even for the HQ9+ solutions with just one, it shows two
I think it's a trailing newline that's doing that.
I can easily correct that.
01:19
Just checked the HQ9+ answer, the website does automatically add a trailing newline.
01:44
@Juan: What advantages does the Perl GS interpreter have over the reference implementation? Curious.
@Juan: I ask because, well, I'm planning to write my own GS interpreter, which will focus on speed.
My initial plan is to write it for Racket, because Racket has an awesome JIT compiler (or so I'm told).
01:58
@chris: apparently the perl implementation makes it easier to debug. Not so useful if it gives different results though
@gnibbler I thought the reference implementation was easy enough to debug with, but *shrug*.
@gnibbler FWIW my alternative implementations are not going to be 100% compatible with the reference implementation either, but should not affect the majority of programs: 1. You can only use symbols as variables, not numbers or strings. 2. You can't use "#{foo}" to execute Ruby code (since the interpreter isn't written in Ruby!).
I want to implement an extension to allow variables bound inside a function to be scoped, but that may break existing legitimate code, so I'll think harder before doing it.
But were it to be implemented, doing native-code compilation of GolfScript code would be much easier.
the not being able to eval ruby code bugs me a little. it's a bit of a messy addon though anyway. I was going to use if for the decision tree, but i never wuite finished the answer
was not going to beat perl anyway :p
Yeah, you just gotta use the right tool for the job. :-)
I'd like gs to support something like generator expressions
What, like Python?
02:07
It was pretty close, I had all the logic done in about 80 chars iirc, but i thought multiplying the floats was going to be about 30 more
yeah, so i can write shortcircuiting loops etc.
i wish "while" wasn't such a big word too. makes while loops almost unusable
maybe {}% could be made lazy
that shouldn't break preexisting any programs, should it?
i guess it could if have sideeffects in the {}%
Hmmm....
You can use "do" instead of "while".
02:27
I dislike the definition of [ and ], from a compiler-implementer's point of view. It means that after seeing a ], you can never really be sure what the stack looks like.
02:50
Do you guys want Code Golf questions to be migrated here from SO, or do you want to wait until you get out of beta?
I don't think they should be migrated wholesale.
They can be migrated on a case-by-case basis.
@RobertHarvey: There is one post you can migrate straight away. I asked rchern to, but she's not around. :-)
1
Q: Code Golf: Robot Finds Kitten

LiraNunaThe challenge The shortest code by character count to help Robot find kitten in the fewest steps possible. Golfers, this is a time of crisis - Kitten gone missing and it's Robot's job to find it! Robot needs to reach Kitten in the shortest path possible. However, in Robot's way, there are a lot...

Much thanks!
No problem! Looks like an interesting one.
:-D
Excellent, thanks for your help!
 
1 hour later…
04:22
:D?
@LiraNuna: o/
 
7 hours later…
11:31
heya liranuna!
i added ascii-art tag to your question
11:43
Would anyone be interested in me explaining how my Befunge solutions work?
Similar to what Chris Jester-Young did for a couple of his golfscripts
 
4 hours later…
15:37
@Nemo157 That would be great!
 
3 hours later…
18:59
Ok, posted an explanation for codegolf.stackexchange.com/questions/369/…
0
A: IPv4 Integer Conversion Function

Nemo157Befunge - 2x11 = 22 characters So close, Befunge will win one day. >&+~1+#v_.@ ^*4*8*8< Explanation The biggest distinguishing feature of Befunge is that instead of being a linear set of instructions like most languages; it is a 2d grid of single character instructions, where contr...

Depending on how busy I am at work today I might get another couple done.
 
2 hours later…
21:08
Okay, everyone, I updated my script to add detection for HQ9+ entries.
I also fixed the off-by-one count problems.
@GeorgeEdison I have to say, your script rocks. It would be great if it became standard
@Juan: Thank you.
I just want to make this site the best it can be.
Any ideas or advice for future improvements are welcome.
I'm currently thinking of expanding the summary at the top of each question...
...maybe to include the language and author of the current best entry.
maybe a quick link to it?
Yeah, I'll add that too.
Ok, this is a bit of a crazier and more complex idea, sort answers by count
21:14
@Juan: That's easy actually, thanks to the way the pages are laid out.
It might take a few days to find its way into the script though.
I'll see what I can do.
Maybe add that as one of the sorting links on the page
By the way, are there any languages you can think of besides HQ9+ that need to be disqualified?
If you need a hand with it, I'll gadly help
Befunge I've been counting differently, basically the smallest bounding rectangle of characters instead of total characters
Should there be a special case for that?
21:20
I'm not sure, I've only seen one other person using Befunge as it's not really the greatest golfing language.
 
1 hour later…
22:49
@Juan: Here we go:
Not there yet, but a start.
It's in version 0.18, which was just released.
Great! As soon as my script updates I'll check it out
How do you figure out the language?
Do you know if there's some way to force Chrome to update the script?
@Nemo: Just click on the script link and it'll replace your current version.
@Juan: It looks through the answer for headings and compares them against a list of known languages.
It's not perfect, but it works pretty well.
If anyone if starting out on golfscript, I just wrote a sort of 'worked example' with factorial (the golfing process, not the solution). It probably isn't going to see much (any?) use, but here it is anyway... codegolf.stackexchange.com/q/629
4
@Nabb: Any reason for the spoiler text in the code blocks in your post?
23:00
Figured that someone might not want to see all the code straightaway
Oh, I get it.
The content of that post could fit better somewhere else, it's a bit huge for that page :S
Maybe in a GolfScript tips page?
It's more of a story than tips though -- although it could always be hacked to small individual tips :P
@George: The script shows the highest post when there is a tie, is it possible for it to show all languages that have the tie?
23:13
@Bass: Yes, that's planned.
Would it be possible to show character counts for all code blocks without re-writing the script?
What do you mean?
At the moment it only shows character counts for just the first block
Can you have it show them for all blocks, but only use the first for determining the shortest language?
@George: Another thing, the note above the question says X bytes while the note below the code block says X characters. Any reason for this difference?
@Bass: Good catch.
As for all of the blocks, yes... that does sound like a good idea.

« first day (8 days earlier)      last day (5131 days later) »