« first day (1451 days earlier)      last day (2233 days later) » 

2:24 AM
I had to make a solution for rectangles since all the other ones were shockingly long ;)
 
 
4 hours later…
6:33 AM
@feersum Heh, I knew my solution was needlessly messy :P
Nice job!
 
Thanks
 
Oh nvm
 
hmm with exec-strings syntax highlighting makes it look kinda gross :P
at least it didn't have a number followed by a letter!
 
Heh well, I just thought it'd make it easier to see what was in the string and what wasn't
You can change it back if you want :)
btw some rule-bending I contemplated was A=x,y=min(s) followed by print A,w,h at the end. It saves a byte and changes the output format, but wasn't sure if that was okay :P
 
how is that supposed to be shorter?
 
6:47 AM
...
I need sleep
 
 
2 hours later…
8:49 AM
hi all
 
9:07 AM
hi
Though I'm sure I was not necessarily thought of when you said 'all' :)
 
you are definitely in "all" :)
 
Cool, I'm included :)
How are you?
 
not bad thanks
always thinking about my next challenge :)
how about you?
 
not too bad either
I'm wondering if SE could do with a Q&A site about open source in general
 
what sort of Qs?
 
9:18 AM
Sounds like a good idea - a place for the self-proclaimed legal experts to debate the merits of the various software licenses!
 
That's the problem, I don't know all that much about open source development.
But I'd love to lurk that site if it existed
I guess stuff about hiearchies in open source project, legal issues, licenses
how to get people interested
how to keep them motivated
How to convince people to use your open source software
Probably about using open source solutions as well
I'm not sure but I feel like questions might run dry rather quickly though
On a more related topic, seems like PPCG is having a rough time generating questions
 
@overactor I could have sworn there is a proposal on Area51 already, but I can't find one.
Just open one yourself?
 
@MartinBüttner I don't quite feel qualified to do so
Though it can of course be improved upon by others
@MartinBüttner It's baffling, isn't it?
 
@overactor We're closer to 4 than 3 questions per day (not right now, but throughout most of the weekdays), which is better than a couple of month ago (but worse than over the summer).
 
Calvin is still easily pulling more than his weight :)
 
9:33 AM
he's pretty quiet at the moment
 
Of the last 5 non-closed questions, he posted 2
 
yeah, those were his first two in a month or so I think
 
I stand corrected.
I've noticed you've definitely done your part as well
I keep finding you over at HNQ
 
ah well, there were two more in early january (including his amazing Wall Paint challenge)... but before that he last posted in November
@overactor haha, yeah, I'm trying...
 
Open Source
For questions regarding motivation, marketing, organisation and coding pertaining specifically to open source projects.
?
 
9:41 AM
sounds good to me
 
Oh hell, why not try it.
 
@PeterTaylor @Doorknob remember that guy from the Free Binary Tree question who just couldn't cope with people asking for clarification? Looks like he changed his username network wide to "Crappy Experience Bye"... I'm guessing something else happened as well, but whatever did happen, that seems like quite an overreaction for someone with 125k rep on SO and 35k on CR.
 
@MartinBüttner wow
Well, f-ed that up quite skillfully
 
I think if you don't post a referral link it'll onebox
 
2
Open Source

Proposed Q&A site for questions regarding motivation, marketing, organisation and coding pertaining specifically to open source projects.

Currently in definition.

@MartinBüttner That certainly was better
 
9:52 AM
you should probably put a couple of example questions in
 
I'm thinking of some
I only get 5 so I have to make the most of them
 
how do you upvote an area 51?
 
@feersum There's no upvoting I thgink, just following
 
The problem will be 70% of questions asking "how do i get (coders|artists|etc.) to do my project?"
 
@feersum you can follow proposals and vote on example questions
 
10:00 AM
Where following means, yes, I'd liek to see this site as a beta
And voting on questions allows you to get your say about the scope of the site if it were to be created
So just upvoting all questions because you like the proposal or downvoting them because you don't is not the way it should work.
 
can I put a deliberately bad question so that everyone can downvote it?
 
@feersum Not sure what the goal would be there, but I guess so
 
@MartinBüttner I'd noticed
@overactor The goal would be to get a consensus among the early committers on what the site scope is not.
 
@PeterTaylor But posting a deliberately bad answer?
I can understand edge-case questions, but I hardly see the point in posting something you know will be down-voted.
 
10:17 AM
@overactor, I can understand why someone, having seen multiple sites pulled way off their original scope, would want to try to pre-empt that.
 
@PeterTaylor I suppose I see your point
Ask an example question, and comment that this is your idea of a bad question and ask to for downvotes
 
Does java (java.util.regex.Pattern) implement PCRE flavor of regex ?
 
(the lookahead and group syntax is similar, just want to confirm)
@PeterTaylor then is it very close to any specific flavor ?
 
I'm sure there used to be a table which compared tonnes of regex implementations / flavours on regular-expressions.info, but I can't find it now.
 
10:30 AM
me neither
from what I can find, Java uses perl like regex
 
@PeterTaylor there are two, one on wikipedia, one on regular-expressions.info
@Optimizer Java defintiely has its own flavour, but like most modern flavours it's perl-inspired
This is a comparison of regular expression engines. == Libraries == == Languages == == Language features == NOTE: An application using a library for regular expression support does not necessarily offer the full set of features of the library, e.g. GNU Grep which uses PCRE does not offer lookahead support, though PCRE does. === Part 1 === === Part 2 === == API features == == See also == List of regular expression software == References == == External links == Regular Expression Flavor Comparison — Detailed comparison of the most popular regular expression flavors Regexp Sy...
 
is PCRE perl inspired too ?
 
Aha: they've got rid of it, but archive.org still has it
@Optimizer The PC is short for "Perl compatible"
 
oh ok
 
@MartinBüttner Yeah, I found that, but it's really limited compared to the one which regular-expressions.info used to have.
 
10:33 AM
@PeterTaylor yeah that's true
I wonder why he dropped it
 
11:19 AM
Geobits has problem with everything . Now people posting sandbox posts for more than 30 minutes makes him sad.
 
I think it might be time for an intervention if that is true
 
@Optimizer just checking, is that JS controller for the rat race still coming?
 
its rat race now o.O
 
it's been a rat race since day 3 or so
 
12:21 PM
@Optimizer is that a "no"? (I don't mind, I'd just like to know.)
 
I vote no on taking the log
 
wrong room
 
oh whoops
 
12:42 PM
@MartinBüttner yes its a no
 
Hmm does Calvin ever come into chat?
 
@Sp3000 rarely, but yes
looks like he hasn't been for a while though so you might not be able to ping him
Doorknob could, though.
 
Hmm k, just reading the new question
 
new question? :D
 
12:48 PM
Calvin's experimental one
I'm concerned that some languages take more than a few chars before you can show anything interesting
 
haha, just saw it... that looks crazy
hm, I'm not sure how well this work out, but I guess we'll have to try
 
I know an interesting 1-char fact for Python
 
Oh?
 
I think starting a factoid with hate as a feature is not a good idea :D
 
Urgh, ><> starts with a > so I need to add a comment so it doesn't become a yellow block of text
Is there a better way?
 
1:03 PM
Marbelous is rather discriminated against in this challenge :(
 
I was going to do FRACTRAN but I'd need 3 to do anything useful :/
 
1:21 PM
I'm so glad I've got Select[Names["*"], StringLength@# == 4&] in Mathematica :D
 
:P
 
@overactor did you find something doable with a single character or are you waiting for two upvotes?
oh, there's the second :D
 
@MartinBüttner I have one character solution that both interpreters allow
 
awesome
on the next upvote I can finally start with mathematica's crazy built-ins :D
 
1:37 PM
I'm getting exclamation mark syndrome, this is pretty exciting (although unfair to people who are asleep right now)
 
1:53 PM
This question is going to eat quite a bit of my time it seems
 
I am gonna update only every 5 upvotes :P
 
Why every 5?
 
to save time
 
@Optimizer Might be rough to get your first 5 with just a factoid
 
it's getting hard to decide...
 
2:00 PM
I just posted a pure algorithms question
I hope it's ok
daringly I posted it straight without a sandbox.. sorry
 
@Lembik you'll want a tie breaker
 
ah yes
my tie breaking rule is always the same :)
of course I don't know the total number of people here interested in pure algorithms questions
I hope many!
I love these brilliantly important questions and answers movies.stackexchange.com/questions/30092/…
@feersum are you there/here?
 
yes
 
@feersum when you say boxes, do you mean the same way I did the time and cost?
or something cleverer?
 
right
 
2:14 PM
how about now?
@feersum also.. does the question look interesting to you?
 
I'm not super clear on what it's asking so far
 
@feersum oh dear.. please let me know anything I can clear up
@feersum I just made a small edit that might help
 
I can't attempt it right now
 
@feersum :(
@feersum if you do get time and something is unclear, please ask
I wonder why it hasn't appeared here yet?
 
it doesn't
0
Q: A fastest algorithm scheduling challenge

LembikThis is my first experiment with an asymptotic complexity challenge although I am happy with answers entirely in code as long as they come with an explanation of their time complexity. I have the following problem. Consider tasks T_1, ... T_n and machines M_1, ... , M_m. Each task takes a cert...

@Lembik the bot only posts meta questions and sandbox answers
 
2:25 PM
@MartinBüttner aha! Thanks
@MartinBüttner it might have made a good code-golf challenge if I had specified a maximum score... next time :)
@MartinBüttner I hope this new scoring scheme works
 
@Lembik If it had been a code-golf I would definitely have voted to close as a duplicate of various existing shortest-path questions. As it is, I'm debating it.
 
@PeterTaylor oh.. you are brutal :)
@PeterTaylor when you say shortest path... do you mean you already know a good score?
 
How many shortest-path questions does the site need?
 
@PeterTaylor I am sorry to be dumb.. but why is it a shortest path problem?
except in the sense that all optimisation problems are
 
How would you do it other than as shortest path?
 
2:31 PM
@PeterTaylor without wanting to give anything away.. dynamic programming is the way I would do it
 
Your vertices are (number of tasks done, current machine, current time).
 
I can see you could write it that way and that might be the best way. The way I thought about it was just as a dynamic programmign problem.. but then my score is only 300,000
no vertices in sight :)
maybe I wouldn't win my own challenge
@PeterTaylor how do you include time AND cost in the graph formulation?
 
Time is part of the vertex, as mentioned above. Cost is the weight of the edges.
Deadlines prune vertices.
 
sorry I meant the switching time
@PeterTaylor there is a time to switch machines
 
2:36 PM
oh ok
 
Each vertex has a directed edge per machine (unless pruned for deadlines)
 
seems you may be making a very nice solution!
what is the total number of vertices you need?
 
I'm not making anything: just pointing out what the question is.
 
@PeterTaylor to be fair, you are seeing the question in a very nice way.
 
In the worst case the number of vertices would be (number of machines)^(number of tasks), but some of them may be pruned and some of them may be unvisited.
 
2:38 PM
ok so this is potentially much larger than my 300,000 right?
or err... let me redo that
ok my 300,000 is wrong :(
 
@MartinBüttner I see why you like Mathematica
 
I actually get 10^9
which is still better than 100^100 :)
ok so if the most obvious formulation as a shortest path problem gives 100^100 at least and there is another method that gives 10^9 then this doesn't seem so bad, right?
 
Wait til Martin has enough upvotes to do GameOfLife[] or whatever it is...
 
CellularAutomaton IIRC
I am amazed that no-one else thinks a question which boils down to "Which is the most popular language?" is off-topic.
 
@PeterTaylor People do think it's off-topic though.
I'd bet.
 
2:46 PM
I'm the only one who's voted to close.
 
I suppose thinking something and acting upon it are two different things, you're the only consequent person on the entire site it would seem.
I'm not even being sracastic.
 
@overactor consequent?
 
Is that not how one expresses that?
 
do you mean conscientious ?
what do you mean by consequent?
 
@Lembik No, though that applies too
I mean that you actually stick by the rules you agreed upon.
A parent who punishes a child and but doesn't enforce teh punishment wouldn't be very [whatever word I am looking for].
 
2:53 PM
Self-consistent is about the closest word I can think of.
But it's not an exact translation. More a loose gloss.
 
Well, you get the idea
 
principled?
 
Seems like I just ported the word from dutch to German.
 
talking of porting.. I see golf has made it to puzzling.se :)
 
@overactor I think it's used with a similar sense in Spanish and French too. English is probably just the odd one out because it's a weird hybrid.
 
3:04 PM
@PeterTaylor are you happy with my question now?
 
In any case, teh question is now very firmly planted on the #1 spot over at the HNQ
 
@PeterTaylor or at least not disgusted by it :)
 
@Lembik Be careful what you ask Peter
 
@overactor oh!
@overactor was there something bad about my question?
 
I haven't had a proper look at it, but you know Peter...
 
3:09 PM
@overactor I don't but I feel you are warning me ...
@overactor my only experience of peter is someone who has written amazing answers to some of my challenges
@overactor and I hope will write more in the future!
 
@Lembik I admit that it has enough structure that it's more interesting that my initial assessment.
 
@PeterTaylor thanks!
@overactor and thank you!
 
But from the impression I've formed of what interests you, I suspect that you're not going to get the answer you're looking for...
 
@PeterTaylor that's interesting. You mean I won't get a smart, well explained and surprisingly fast solution?
 
... in the sense that you seem to care about practical performance, and I suspect that the practically best performing solution will be too hard to analyse to get a tight performance bound.
 
3:15 PM
@PeterTaylor oh I see.. no in this case I am really experimenting with not caring about practical performance
@PeterTaylor I am basically trying out different sorts of questions to see what fun there is to be had on codegolf.se :)
 
Finally back to 3 character score :D
 
@PeterTaylor the only answers I don't like are golfed ones.. (sorry to all the golfers)
 
Funny, I had the impression Lembik didn't like practical performance - he disdains C++ answers :)
 
@feersum I don't disdain them! :)
@feersum I just love things that are easy to read and ALSO fast
everything has to be in nimrod :)
 
3:34 PM
I have just seen how often a fastest-algorithm challenge is posted!
(clue: not once a month)
 
of course, its code-golf, not code-race
;)
PPCG vs PPCR
 
@Optimizer I think these days it was code-* :)
one SE to rule them all
 
the name of the SE is still PPCG
 
yes... but then we also have a department of defense
...
(the don't do a lot of defending in the normal sense of the word)
 
@MartinBüttner Yeah, well aware. All of the drama around that happened about a week ago. :P
 
3:44 PM
codegolf.stackexchange.com/questions/44680/… doesn't python have a common that opens an xkcd comic?
 
hsl
@Lembik Is import antigravity what you're thinking about?
 
@hsl yes! Thank you
you even translated common to command :)
 
4:38 PM
hmm.. no answer at all yet
I wonder if I should put one in with a terrible score
 
5:03 PM
@Sp3000 coming up... (at some point)
 
5:14 PM
13, 15 and 18 missing... then I've got snippets up to 22
 
5:36 PM
finally up-to-date...
 
@MartinBüttner you actually added 1 extra
 
@Optimizer huh?
 
your votes at that time were 18 and you added till 19
 
no they weren't
I got that upvote about 10 or 15 minutes ago
and no it's at 20
 
oh.
maybe my page was not updated.
 
5:48 PM
CH's challenge is close to 500 hotness points o.O
 
how do you know ?
 
hover over the question on the HNQ site
 
ah
second one has 130 :D
 
#1 on HNQ - sure sign of a low-quality question =D
3
 
our recent super popular questions have been very low quality only :D
"cough" "stroupsoup" "cough"
 
5:53 PM
@feersum for popcons that's often true
if a code golf gets there, not necessarily... it usually just means it's not particularly hard
 
@MartinBüttner can you reverse your order ? for every update , we have to scroll down so much.
 
@Optimizer I went with Sp3000's order, because I think it makes more sense this way
 
updates are more important imo
and mostly all timelines have recent on top :)
 
this reminds me - is there a way to cast an anti-close or anti-reopen vote?
 
5:57 PM
nope
 
seems stupid.. if 5 people wanted one way and 800 people wanted the other, then the 5 can win
 
other 800 can reopen , you know
 
well, a situation of ongoing toggling is clearly stupid
 
@Optimizer done
 
@feersum in the review mode, there is an option of "let it be open", but no idea if that does anthing
 
6:01 PM
@feersum in the review queues there's "Leave Closed" and "Leave Open". if enough of these come in, then I think the existing votes get annulled
 
@Doorknob ^ true ?
 
56
A: What exactly happens with the button "Leave Open" (previously "Do Not Close")?

Shog9Normally, a question must hit 100 views before close votes age away, so as to give enough folks time to see and review it. Questions with close votes stay in the review queue until they are closed or the votes expire. But if enough people (currently 3) select "Do Not Close" within the review que...

okay they don't get annulled right away
but they speed up the aging process of the votes and remove the question from the queue
 
:D
Mathematica totally owned this one!
Martin's getting rich. specially considering that a lots of views are gonna flow in for the next day
 
I think if Sp3000 was up to his votes he'd probably outrace me... 2D languages are pretty fun
 
6:20 PM
"`_~"`_~ if I am not wrong, this is the smallest possible quine in CJam, right ?
 
I prefer {"_~"}_~
 
that does not look symmetric though :)
 
it looks a bit like an ASCII Kirby though
 
yours or mine ?
 
6:25 PM
no way
 
maybe I should promise to give a bounty to the first answer :)
as I hate the wait between posting a question and getting an answer
 
dude, your question is tough
 
@Lembik it's not clear whether your switching tables are given for switching from or to the machine in that column.
 
and for setting bounty, you will have to wait 2 days anyways
 
6:31 PM
@Optimizer he could promise it in the post
 
oh . yeah.
 
@MartinBüttner oh thanks... would it help if I added column headers?
 
no
it's clear what the columns represent
 
haha
I was just coming here to ask what the columns are
 
but it's not clear whether it's switching "from row to column", or "from column to row"
 
6:32 PM
for switching time
and cost
I guess it must be machine x machine since that's the only thing there's 2 of
 
@MartinBüttner is it clear now?
 
@feersum yeah I'm pretty sure that's it, but I don't know which way around...
 
@MartinBüttner it costs nothing to switch to the same machine
 
@Lembik yes much better
 
great!
please do ask things like this... I can't always tell what is unclear
I wouldn't mind some upvotes either ;)
 
6:36 PM
you've had mine for a couple of hours
 
thanks :)
 
hm, I wish I could post a prelude answer to CH's challenge... it's a fun language
 
I'll make a prelude answer..if you can give me a good 1-byte program
 
Hmm, I can only get 10 bytes for the smallest negative quine (prints opp of source code) ..
 
@feersum you can print a 0.
it's not that impressive but it serves to show the wonders of starting with an infinite stack.
 
6:46 PM
I suppose
of course, this is our challenge-specific Prelude variant
 
no, the Python interpreter already had NUMERICAL_OUTPUT set
 
oh, right..chars was for input
now, I've gotta dig up interesting facts on the history of Prelude..
 
as the factoid you can mention Fugue
 
 
1 hour later…
7:52 PM
interesting... someone downvoted the leading suggestion for the file system tag, but afaict didn't vote on anything else, either up or down.
 
8:26 PM
1 thing for sure, CH's answer is helping everyone in getting EPIC and beyond :D
 
lol yeah
 
I am still not 200 though :(
but 10K F'Yeah!
as far as I can see there is no 9 byte reverse quine in CJam..
 
@Optimizer oh nice, congrats!
 
 
1 hour later…
9:42 PM
@MartinBüttner Just woke up, got some catching up to do. Coming up with good snippets takes thinking though :P
Although, close votes
 
hmm... I was thinking about a cops and robbers game with the halting problem :)
but I may not have understood the cops and robbers set up
 
can't seem to imagine getting 7 more upvotes in 2.5 hrs .. :(
 
the idea is that one party gives code and the other has to say if it halts or not.
the party that offers code gets points inversely proportional to its length according to some as yet set function
in fact both parties do
so very short code gets lots of points
err... no.. well I suppose the robbers get more points if the code is long
oh damn this isn't going to work I just realised.. someone will just post the equivalent of an open conjecture about primes
oh well :)
@Optimizer I managed one upvote in a day!! :(
(maybe it could be code-golf... the shortest code that no one can prove if it halts or not)
 
@Sp3000 tell me about it...
 
10:44 PM
@Optimizer See meta.stackexchange.com/q/139834/180276 (/cc @MartinBüttner)
 
@Doorknob yep :)
it is done
0
Q: Lab Rat Race: an exercise in genetic algorithms

Martin BüttnerThis is Weekly Challenge #3. Theme: Genetic Algorithms This challenge is a bit of an experiment. We wanted to see what we could do, challenge-wise, with genetic algorithms. Not everything may be optimal, but we tried our best to make it accessible. If this works out, who knows what we might see ...

4
 
Oh, and also...
Congrats on reaching 10k @Optimizer!
 
11:07 PM
+26/-2 hmm
 
@Sp3000 I feel like people are downvoting if you get too far behind your vote count.
 
Not my fault Calvin posted when it was 1am over here :P
 
I had a downvote earlier today when was some 10 snippets behind, and after I added them, the downvote got removed (and maybe turned into an upvote, I can't tell).
 
11:25 PM
@Doorknob if you don't mind I can just post the meta post about our weekly challenges myself
 

« first day (1451 days earlier)      last day (2233 days later) »