« first day (1361 days earlier)      last day (3465 days later) » 

12:11 AM
@MartinBüttner and only now did I remember that there was an answer I wanted to accept :(
 
1:04 AM
Probably not specific to 8, I just notice it more there.
 
@Doorknob I just found this: "Microsoft loves Linux" on WinBeta: winbeta.org/news/…
Or are you a Unix fan?
 
@hosch250 Where do you see that? All I see is stuff "Cloud OS" (whatever that is) and Azure.
 
At 13:14.
Scroll down near the bottom, you may need to load more.
You can also search "Linux".
 
Ah, I see. :P
 
 
2 hours later…
2:59 AM
Wanna ask
Is a 128-bit RSA key secure enough for 3 days
ran out of memory when trying to do a 128bit*128bit multiplication with regex
 
3:19 AM
Hi
@TwiNight I did 768bit*768bit multiplication in the uncracked answer
 
Also 81bit was cracked...
 
@user23013 o_O
ok I give up......
 
I searched in wolframalpha the prime greater than a randomly typed number. Didn't count how long it is in base 2..
I have another idea about random accessed memory, to make the multiplication regex much shorter.
 
3:48 AM
@TwiNight You can have a look at the 17372 byte 81bit*81bit version. It can be improved. And you can also try something more than multiplication.
 
Hi guys
I am looking to get some feedback
0
A: Sandbox for Proposed Challenges

Ilya_GazmanBefore I start, I have few notes. Links will be available once the question will be published. I am keeping it in sand box until the next Monday, October 28. I am going to invest some money in advertising this question, so take under your consideration that it will be visible to many new users ...

 
Hi! I think I took a look at that one, but wasn't sure if everyone would have access to AS3
I might be wrong about Air, but I thought Flash was paid?
 
No, its free
You can use flash develop
Also the source code is fairly simple. You can use a notepad.
 
4:10 AM
@Sp3000 here is an online editor: wonderfl.net/c/hCMB/edit
 
 
4 hours later…
8:13 AM
0
A: Sandbox for Proposed Challenges

Teun PronkI posted this earlier but they suggested I put it in the sandbox which made a lot of sense. Reading sheet music Your job is to take an image of sheet music and make it usefull for someone that doesn't read notes. Output 3/4 (Which is the beat) C F B,A,D E F C (Notes, Space seperated and comma...

 
8:23 AM
hi
 
8:40 AM
With the interpreter idea, it'd be great if you could make Stack Snippets expand out to full screen when you hit "Run Code Snippet"
 
8:59 AM
@Sp3000 The example one by Shog on meta does that
@Ilya_Gazman Are you aware that by posting it on SE, everything you include (like the code) is licensed under CC BY-SA.
 
@MartinBüttner Does it? For me it goes (Run Code Snippet) > [stuff expands] > (Full page) > [expand to full page]
i.e. I have to press two buttons to expand to full page
 
oh yeah, that's true
 
These snippets don't work well on mobile though :( scrolling works funny
Well, time to brush up on some Javascript :)
 
0
Q: I did't got reputation for vote-ups

Sempierecently I got some upvotes but no reputation for that, why is that? I thought this would depend on an delay to refresh the reputation, but this was several days ago.

 
 
1 hour later…
10:16 AM
if you have an array of positive integers, is there a linear time way to tell if one number is three times some other number?
 
@felix lang-pls
 
Hashing?
 
@Unihedron what does the language have to do with the time complexity?
 
@MartinBüttner the implementation does
 
@felix it's unsorted I take it?
 
10:21 AM
unsorted
@MartinBüttner yes
 
First thing I think of is to hash n/3 for every n you see, then when you check the next number you see if it's in the table
 
@Sp3000 that would work... let's try for a non-hashing solution :)
 
Aww :(
 
well I can't even tell you the complexity of the hashing solution so I regard it as a practical hack :)
 
Well are they integers or floats?
 
10:23 AM
integers
n of them and all positive
 
@TwiNight No. 512 bits is enough unless people are willing to spend money on AWS instances; 256 bits is not enough; the exact cutoff is somewhat blurry because it depends on how quickly a cracker is able to get to work. If you're lucky and post just after they go to bed, you can get an extra few hours.
 
Nope, my head keeps coming back to hashing :(
 
@Sp3000 thanks for looking at it :)
 
10:39 AM
@felix Fixed width? Then technically you can sort them in linear time using either radix sort or counting sort. Once they're sorted, it's easy.
 
@MartinBüttner I found a small problem in my uncracked regex submission.
 
@user23013 which is?
 
@MartinBüttner Yeah I know. I been working on it more than a year. It's more for fun, than for money.
 
Unlike other online testers, regexlib.com has two options selected by default. I thought the regex should also work with those options. But it doesn't. And the regex doesn't work anywhere except regexlib.com.
 
Quick question: For BF's tape, does it extend out "infinitely" both left and right?
 
10:41 AM
Can I assume the .NET default instead of the site's?
Oh I didn't try those Silverlight testers.
 
@user23013 Well yes, but if the tester you linked assumed those options that was kinda misleading, I guess?
 
I added "(with all options unchecked)" in an earlier answer, but forgot to do that in this one
 
@Sp3000 There are many implementations, and in line with Murphy's law just about every permutation of design decision is supported by one of them.
The original implementation wrapped after IIRC 30000 cells.
 
Ahaha... well. Here comes 30k, no wrap then
 
So should I just make it safe, or edit and wait another 3 days? I don't think anyone have tried anything similar to the solution, though
 
10:53 AM
@PeterTaylor a very good point
 
@user23013 Technically, if we'd consider this a change to the regex, then yes, you'd have to wait another 3 days.
 
11:12 AM
@MartinBüttner It's a change to the options, or a fix for the inconsistency in the options as I didn't say those should be selected. And if people try it in .NET, they should be using the correct option. It is misleading if they try on the tester.
 
Well, they could have opened the tester first, seen the options there, and taken them to be part of the setup. I think it would be fair to extend the deadline, and with those huge primes I don't think it will really put your submission at risk. ;)
 
OK, I'll edit.
 
11:50 AM
I also started a bounty on an old question for CJam/GolfScript answers. But it seems that nobody bothered to do that.
 
I hope this isn't too off topic, but.. I want to run unison (like rsync but better) between my home computer called A and my work one called C. A can connect to B and B and C can communicate freely. Of course the standard solution is ssh port forwarding but B, which is not administered by me, seems to have disabled that
any clever solutions?
 
You can connect with ssh but cannot -L?
 
@user23013 If you have the time I can show you step by step what goes wrong
@user23013 have you got a few moments?
 
Yes but, in fact, I'm not a native English speaker and didn't chat in English anywhere before today.
 
@user23013 oh well.. welcome :)
but I see you speak regex
so we have something in common :)
 
12:08 PM
I need only one documentation for regex (for each flavor), but sometimes I'll need google for English
 
@user23013 I have to say - good job on coming up with the empty string balancing group thing. Your factorisation problem's insane
 
@felix Glad if I can help. But I only know the basic things about ssh
 
ah ok
well .. I am trying to work it out now
 
@Sp3000 In fact, those strings don't have to be empty, and you can possibly have random accessed memory with non-empty strings. But I didn't try that yet.
 
Oh? I was under the impression that you needed to use empty strings because there's infinitely many of them in any string... unless you're thinking of using lookaheads or something?
 
12:15 PM
Yes
 
0
A: Sandbox for Proposed Challenges

professorfish5 Miscellaneous Flag Challenges code-golf graphical-output geometry We've had questions on the South Korean flag and the flag of Nepal. I thought that some other flag questions could be interesting. I've grouped these into one post to avoid cluttering the sandbox with potentially bad posts. Onc...

 
@Sp3000 For example, ^01,...(?<=^.(?<1>.).*)...(?<=(?:(?<=^)(?<zero>)|(?<=^.)(?<one>))(?<=^[^,]*)\1.*‌​)...(?<-one>)
And use two stacks to get infinite memory
 
Ahaha k :P
Quick poll if that's okay - when you run the code snippet here codegolf.stackexchange.com/a/40091/21487 does everything fit inside the snippet box?
(not a HTML/CSS expert so I know next to naught about browser compatibility)
 
12:40 PM
"timeout" wrapped into the next line in my browser.
 
Hmmm... daamn
Thanks
 
This word and a few px of the button is not visible
 
Hmmm I wonder if the box expands to a different size per browser of what the problem is...
 
I think it's the font
 
Ah... hm...
 
12:44 PM
You can set a fixed line-height and font-size
 
Hm...
 
@Sp3000 And you can have a try with that regex technique if you want to. I didn't plan to do that in a few days. But if there isn't an answer with this after a week or so, I'll try something like much shorter multiplication regex, or brainfuck interpreter.
 
Ahaha I'll be giving regex a bit of a break I think :) there's things I should be doing but PPCG
 
@Sp3000 You meant to say "there's other things I should be doing on PPCG"
 
:P
Already doing
 
12:57 PM
Say, you have a bunch of decimal numbers that add up to a value (classic example: percentages which add up to 100). But you only want to display them to certain precision. How do you round them to ensure that the rounded numbers still add up to the target sum?
 
Keep an offset as you go and as you round make sure the offset never exceeds +-0.5?
 
I was just going to suggest something like that... I just wasn't sure whether a) there's some weird edge case I'm overlooking or b) there's a simpler alternative
 
@MartinBüttner I think something about election already said that algorithm always has strange behaviors.
 
I suppose an alternative would be, to round all of them up, check how much you overshot. And then subtract 1 from that many.
 
Well I guess the question is why do you want the rounded numbers to still add to the target sum
 
:/ specs
@user23013 codegolf.stackexchange.com/a/40091/21487 I tried setting the font and size - is that any better?
 
The font size is reset in textarea.
 
:/
I need to brush up on HTML/CSS...
 
I used to resolve compatibility issues for IE6, where brushing up doesn't help much.
 
Solution: Make your users pay to access the site in IE6
 
1:15 PM
And make the payment page not working in IE6
 
:P I like how you think
 
Or if they really pay, I'll still have to do that
 
1:30 PM
@MartinBüttner I have read something like this: en.wikipedia.org/wiki/Apportionment_paradox , which means your method probably always have edge cases. So I think you can just use something looks simpler, if that's not too important.
 
thanks for the link
I'm going with keeping track of an offset and rounding accordingly now. It seems to work quite well so far.
damn, this challenge is trickier than I thought
my idea was this: generate an image that has the correct amount of pixels of each colour. shuffle all the pixels. and then keep swapping pixels until planar regions form. but that last part is actually quite tough.
 
1:52 PM
Well, I think the rounding part in this question isn't a big problem. You can even round 100.1 up and 100.9 down. The link is irrelevant.
 
I think Stack Snippets is turning into a different sort of golfing for me - how many features can you fit into limited screen real estate
 
2:43 PM
Limited real estate? Can't you use a snippet to open up more windows/tabs? :P
 
More fun this way :P
 
3:22 PM
That mechanized abbreviation trick for abbreviating JS method names reminded me of something we stumbled upon while trying to golf the uses in demoseen.com/crunch.html down
Someone proposed using parseInt on the property name with base 36, but it turns out that rounding issues caused incosistent results between browsers...
 
4:01 PM
hmm... this looks suspiciously like a question from the IEEE eXtreme 24hour programming competition that took place 3 days ago
-3
Q: For example, "abcd" is an extracted word from "sba

MouryaOur goal is to find the length of one of the longest extracted palindrome from a given word w. Of course if w is a palindrome then the answer should be the length of w, and the answer is always greater or equal than 1.

 
4:41 PM
Huzzah I finally figured out Javascript's setTimeout, and it's 3am
 
 
2 hours later…
6:24 PM
@MartinBüttner Ack I stayed up late enough to be awake when 3 days passed
 
 
1 hour later…
7:37 PM
hi @Peter... I'd like to pick your brains about a challenge idea, but it would kinda defeat the purpose of the idea if I revealed it publicly (in chat or sandbox)... is there any way to contact you in private?
 
Stalker alert!
:P
 
 
1 hour later…
9:03 PM
0
Q: Recommend that new users use bytes, not characters

feersumOftentimes new users innocently specify the winning criterion as fewest characters, as opposed to fewest bytes, without being aware of the differences. As a result some of the answers may be encoded into large Unicode values. But who enjoys seeing an answer solely consisting of a big Unicode turd...

 
 
1 hour later…
10:07 PM
Quick, everyone downvote @MartinBüttner! He passed me in the all-time rep rankings! :P
 
@Doorknob ;)
It's not like you couldn't have asked for upvotes on your answers instead :P
Now let's see if I can pass Howard on the last couple of metres to the magic 20k :D
 
I predict one of you will get there within a week. (The race is on! :P)
 
heh
just in time for our graduation on 1 Nov, amirite @Geobits? ;)
2
 
11:01 PM
@MartinBüttner I own the domain cheddarmonk.org, and have a global alias which redirects to my main account.
 
@PeterTaylor Ah, I see. Thanks for letting me know. I'll write something up. :)
 
No prob.
 
@MartinBüttner Did you see the old question I started a bounty for CJam/GolfScript answers? It's OK if all CJamers/GolfScripters see that and decided not to do. But looks like I'm just going to award a bounty to a random old answer after 7+2 days, and nobody knows why.
 
@user23013 I didn't actually re-read the spec yet, but it seems quite tough in CJam... I'm not that comfortable with stack based languages for more complicated tasks yet. Maybe @Peter is interested? :)
(he's done some pretty crazy things with GS)
 
@MartinBüttner When Deadfish hits 256 it actually gets reset to 0...
 
11:15 PM
@Sp3000 hm, it wasn't part of the spec... you might want to comment in the question
 
Just did :) it's on the esolang page, but yeah it wasn't in the spec
If it wasn't for that then I could just adapt what I wrote for the birthday question :/ (I only wrote it to go up to 128)
 
11:29 PM
@PeterTaylor So are you interested?
I think I should have set the bounty for the first answer instead of the shortest.
 

« first day (1361 days earlier)      last day (3465 days later) »