« first day (1591 days earlier)      last day (3253 days later) » 

3:01 AM
@MartinBüttner After writing my solution for it before, I realize I skipped a minor factor, but I think I could include it reasonably well. I'd be happy to share my algorithm so far, if that helps convince you that it's a solvable problem.
for every two bitstrings, calculate the [im]probability that one is a mutation of the other. this calculation is based on the binomial distribution. take the negative logarithm of that improbability, make that the weight of an edge between those bitstrings' nodes on a graph. assemble the minimum spanning tree of the graph with those weights. choose the node closest to the center of that tree as the root, and I think that gives you the most likely family tree.
 
3:45 AM
0
Q: Translate code into English

KyranstarYour task is to write a program in any language that takes source code of another program in a language specified in the answer as input and prints a logical explanation (or translation) of the core in English. For example: public static class Product { int price; int weight; publi...

 
 
3 hours later…
6:39 AM
0
Q: The hourglass challenge

Renae LiderThe Hourglass This hourglass has 60 "sands of time", and it completely fills each chamber. The clock is 19 characters wide and 13 characters in height. Each chamber has 5 rows of sand and there is a row in the centre that can hold 1 sand. The top row can hold 17 sands, the next one 15 and so on ...

 
7:02 AM
Hi everyone !
 
7:36 AM
@Rainbolt Are you aware of our minecraft server?
@MartinBüttner You know you want to join
 
8:37 AM
:/
@Sparr I've been wondering whether the overall maximum probability might correspond to a spanning tree that isn't minimal but where the children are more uniformly distributed across the parents.
(e.g. the minimal spanning tree might be like a star, with all nodes connected directly to one central node... but seems like an unlikely tree because it would require always picking the same parent)
 
8:59 AM
That was the part that I think I missed. The probability of a given shape of the whole graph needs to be taken into account. One long line is less likely than a tree with an average of e children for each non-leaf node.
I am not sure how to describe that algorithmicly, but I bet there's a simple solution somewhere in probability / graph theory.
maybe missing that is why I didn't get the job... :/
or maybe that probability isn't actually a factor in the result? I might try to think of a small test case to prove/disprove that it matters
 
9:28 AM
wow overactor's Open Source proposal on Area 51 is close to launch... after only 4 months. that's pretty impressive.
 
9:39 AM
0
A: Sandbox for Proposed Challenges

sanchisesChinese Checkers I'm considering writing a challenge that will have people write Chinese Checkers players. Image courtesy of Wikipedia The game is quite simple: In clockwise turns, players try to move their 'pegs' across the board. They may only move one peg each turn. Valid moves are eithe...

 
 
2 hours later…
11:54 AM
0
Q: Tips for faster code in PHP

Ismael MiguelCurrently, there's tips for golfing in many languages. There's already tips for golfing in PHP, but nothing about making fastest code. Some questions have a time limit, while others are all about the fastest code. What general tips do you have to speed up PHP code? (Micro-optimizations like u...

 
12:31 PM
hey @Katenkyo
 
@Agawa001 Hi !
 
i dont get why is this for, that would be a big embarrassement for SE ,
 
@Agawa001 It fits better for sites like doctissimo, not SE...
 
yes
 
Moreover, it only can be suggestive as we're all different and behave differently for sexuality...
It's more a "tips only" SE site, than a Q&A/Challenge
(I don't wanna see their chat by the way x))
 
12:47 PM
even the conversations / topics / debates , taken part , can expose the author in unenviable issues
 
But anyway, I think there can be a lot of people that would need a site where they can found answers for all their doubt/question. It's a good idea in a sense. Even if it will overlapse a bit with Health.SE.
SE saw the value of such a content (lot of high quality answers about questions, even if they are about sexuality/a little bit suggestive). I think that's the "why" :)
Big point : will the community grow?
 
@Calvin'sHobbies I am now!
I spent the better part of my afternoon on Saturday trying to get rid of this stupid tree: i.imgur.com/O310UHs.png
And the best part is that I had to do it without looking up. The moment I looked up, my FPS dropped to like 0.5
 
1:07 PM
You're playing with some modpacks?
 
Why get rid of the Yggdrasil? Now were will the tribes go for wisdom?
 
minecraft ...... wonderful game with super-ridiculous design
 
@Agawa001 With some shaders/HD texture pack it can become quite fantastic to see
 
@Katenkyo Yea. Playing with Direwolf20 1.7
 
@Rainbolt Sounds familiar
Feed the beast?
 
1:14 PM
@Katenkyo Yes
 
@Rainbolt Mwahaha, I might come and build some fusion reactor then :D
 
Come where? I'm playing Single Player lol
 
Not on calvin's server?
 
I think Calvin's server is vanilla
 
Hum, it's sad c:'(
 
1:17 PM
Ender IO is my favorite mod now
 
Yeah, really useful
But Computercraft and Applied Energistics are the best :D
 
I just got into Applied Energistics this weekend. I spent more time automating the processors than actually making machines
It's kind of hard to put a specific item in the top slot, middle slot, and bottom slot, and pipe the output somewhere without taking up gross amounts of space
 
A machine room totally hidden and dedicated to applied can be useful, it will always take a lot of space.
But once it is done, it helps a lot !
When I was doing some ultimate hybrid solar panel, I was happy to have it automatised :D
 
1:35 PM
Is ultimate hybrid solar panel the same as high voltage solar array?
Oh, one tier lower
Well I guess I'm comparing apples to oranges.
One of the things I find fun with the modpacks is finding ways to do the same thing in less space and with less lag.
 
2:05 PM
@Agawa001 you thinking that site would be an embarassment is part of the problem that site is trying to solve.
5
Asking questions about sex is not embarassing.
Also, boo on SE for not automatically committing me to my own proposal when it reached commitment stage. Or even sending me a notification!
(also, not sure if you were aware that I was in here when you started discussing it)
 
@Sparr the notification sounds like it's worth a feature request, but I really wouldn't want SE to auto commit me to the proposal just because I posted it.
 
(related to our previous conversation)
 
ah nice
 
2:20 PM
@Rainbolt Lags are maily due to rendering. Big tanks/pipes that show entity are annoying. Every machine with a rotating part, is annoying too.
I once did a tree farm 256x13 (for steam boilers), it was pretty laggy x)
 
@Sparr that reminds me I still don't have an answer on my math.SE question from Friday... time for a bounty.
 
@Sparr boo on SE ?
 
2:38 PM
@Agawa001 stackexchange disappointed me by not notifying me of progress on my proposal
 
did u propose that sex thing ( oh well , its not a profanity -SE logic-)
 
yes
 
i understand now why v u not got responded yet
 
I participate in a lot of online discussions about sexual topics, including a lot of Q&A. Most of it is on sites dedicated to adult entertainment, which are very taboo to use in a lot of contexts. I don't browse fetlife.com at work, despite it having some of the best forums for Q&A on adult topics. Reddit has a few good sections for those questions, too.
But StackExchange is, by far, the best platform for Q&A I've found online, and not having that functionality for those discussions is annoying.
 
The only embarrassing thing would be when Sexuality graduates before PPCG :/
 
2:49 PM
well yes
S-community must be exhaustive , and encompasse all fields
of life-experience
 
@Optimizer It wouldn't be that surprising, sex has quite a broader audience than programming puzzles
4
 
^ not as (a taboo)
 
@Agawa001 are you saying "S-" instead of "Sex"?
 
stackexchage
 
oh, ok
 
2:59 PM
(as we agree that its nt a profanity just like brainfuck)
 
you mean b****fuck ?
9
 
only puzzling-related topic is prohibited here
 
yeah, keep those anatomical terms out of here
 
3:50 PM
i smell like someone would type something that may be starred more than number of all users of PPCG
 
-1
Q: Compact XML Parser

Benjamin WilliamsGoal: The goal of this challenge is to take a simple 6 row by 6 column XML file (without the standard XML header) and parse it out so that the program can return the value of a cell given it's coordinates. The entry with the shortest overall length by language and overall will be declared the w...

 
4:31 PM
hi @MartinBüttner
and all :)
 
hi :)
 
hi @trichoplax
Just posted a followup question.. I know some people don't like them but I can't see another way to ask about two different aspects of the same problem
 
4:51 PM
0
Q: Calculate a probability exactly and quickly

Lembik[This is a partner question to Calculate a probability exactly ] This task is about writing code to compute a probability exactly and quickly. The output should be a precise probability written as a fraction in its most reduced form. That is it should never output 4/8 but rather 1/2. For some p...

 
@Lembik Do you want it to be exactly the same task, just speed instead of length? Would it gather more interest if it was a variation on the previous task?
(Up to you just wondering)
 
1
Q: Downvoting wrong answers in a specific language

Thomas KwaI golf in TI-BASIC (the language that TI-83 and 84 series graphing calculators use), which is an uncommon language, and small even in proportion to the frequency of its use is the number of people in this community who know the details of the language. Therefore, good answers often go unrecognize...

 
@trichoplax it might... the problem is that I find it interesting because it is exactly the same task :)
as the golf'ed version are super slow.. it is interesting to see how much faster you can be non-golfed.. at least to me
but I may be eccentric :)
 
5:06 PM
I also find that comparison interesting - I was just thinking of how much attention the question will get. But if it's that comparison that you want to test, then I suppose it doesn't matter how many other people are interested.
 
5:52 PM
bot is going slow today
poke
 
0
Q: Show all the possibilities of black or white pixels on a 3x3 grid

Filipe TeixeiraI made this python code that generates 512 images with all the possibilities of combining a white or black pixel on a 3x3 grid. (python 2.7) import itertools from PIL import Image combinations = list(itertools.product([255, 0], repeat = 9)) img = Image.new('1', (3, 3), "white") pixels = img.lo...

 
ur poke worked
 
nah
other bot, I'm waiting to hear from New Meta Posts
 
sandbox ?
 
nah, top level meta post
 
6:00 PM
@Sparr Possible duplicate of this?
 
1
Q: How to test solutions in obscure languages/environments

SparrThis question is meant to serve as a reference for setting up environments in which to test answers that require unusual or obscure settings, operating systems, hardware, compilers, interpreters, etc. All answers should be Community Wiki, and there should be one answer for each notably distinct ...

 
Yea, it could probably just be an answer to the other one.
 
that other one wants online environments
mine is more aimed at obscure environments themselves, like special hardware or input devices
 
Hahaha poor TimTech, always being called out every few months or so
 
Oh, wait, I thought he linked this one, sorry: meta.codegolf.stackexchange.com/q/1213/14215
More general question.
 
6:05 PM
Oops, linked the wrong one
 
ok, so we already have questions for online environments, and for installing interpreters/compilers
maybe I should make mine specific to hardware
 
(that was the one I had in mind as well)
 
@Sparr I'm not sure, unless the answer is just "find an emulator and a ROM". That seems to apply to any hardware that has an emu available.
 
@Geobits I was hoping for something a little more detailed than that
 
:D
 
6:08 PM
emulator or simulator? where's a good open source one, or one for various host platforms? how do you find a rom? is there one available legally? how do you load a rom? how do you get code onto the emulated system to test it? etc
 
argh..-1
maybe I should delete it :(
 
I know. I'm just worried about splitting it up between that and the "set up an environment to test" question, because that's really what you're doing. I'd rather them all be in one place I guess, to make them easier to find.
 
I need PPCG advice!
 
the two existing questions are fundamentally different
I think mine is, too
 
6:12 PM
@Sparr Haven't voted to close or anything, because I'm not entirely sure. Will hear others out.
 
@Lembik those two problems are way too complex for the actual golfing difficulty they present.
 
@Lembik I don't normally recommend deletion after a single downvote, but it's up to you.
 
@Sparr hang on.. the second question isn't a golfing question and the first one has nice answers
@Geobits ok thanks.. I'll wait a bit
I'll delete after two downvotes and no upvotes :)
 
Why?
 
@Rainbolt because it means it's a terrible question!
 
6:14 PM
Hell, the fastest-code version probably belongs on math.SE. I wouldn't make a bet that anyone on PPCG has the requisite probability knowledge to come up with the ideal algorithm for that at all.
I won't be surprised if they do, but I wouldn't bet on it.
 
@Lembik soooo tempting to downvote out of trollishness now
 
@Sparr I should perhaps explain that I asked this question because I am confident there is no smart math answer
@Geobits don't! :)
please
 
Why not just let crowd moderation do its job (with the exception that you can short circuit duplicates on your own now)?
 
@Rainbolt you make a good point
 
And who is downvoting because its too hard?
That seems like a ridiculous reason to downvote
 
6:15 PM
@Sparr I think the opportunities for speedups are all about seeing symmetries to get constant factor improvements
 
@Lembik did you see my question about assembling a probabilistic family tree from mutated strings?
 
@Sparr no! URL please?
 
I think I'm also falling into the quagmire of asking too-CS-y challenges on PPCG
 
@Sparr Isn't variety a good thing?
 
3
A: Sandbox for Proposed Challenges

SparrBitstring Family Trees This challenge is reproduced from memory and my own solution, from a challenge that was posted in the job-application section of http://itasoftware.com before they were bought by Google. I reached out to ITA and Google a few years ago, after the acquisition, to ask to re-p...

 
6:16 PM
Some people ask one type of question, other people other types, the union of the sets makes a good website :)
 
Good opportunity for a plug: I posted this question which is hard (to me anyway), and only one guy has answered it. Someone else aught to beat him. codegolf.stackexchange.com/q/23775/18487
Surely someone can do better than three thousand characters
 
variety is a good thing, but questions that only get one or two answers are not necessarily positively contributing to the site
 
@Lembik Depends on your goal. If looking for rep, you should minmax it by seeing what gets voted up the most. Of course, I wouldn't recommend this ;)
 
Also, I miss ProgrammerDan :(
3
 
@Sparr that's a high bar! I think two answers is ok isn't it?
@Geobits I am not looking for rep :)
one answer only means the question has basically failed I think
 
6:18 PM
Why else would you delete after a single DV? :P
 
@Rainbolt same thing I said to Lembik... that question is way too complicated to make it code-golf. That's an algorithm challenge. The only way it could be a reasonable code-golf is if the algorithm is obvious or published or provided in the question.
 
@Geobits one DV and no upvote makes me think I have done something terrible.. call me sensitive :)
 
@Sparr Longest path is a well known problem, right?
 
@Sparr but in my case the actual golf question has lots of lovely answers doesn't it?
 
@Lembik yeah. I haven't looked at all of them to see if they are all/most using the same algorithm, though
 
6:20 PM
well...4
 
@Lembik After a few days, maybe. Judging a post by its first vote only an hour after asking seems a bit strange :)
 
@Rainbolt yes, but the "you get more ammo by visiting different nodes than by visiting the same nodes repeatedly" quirk means the longest path algorithm is barely a start.
 
@Sparr I agree that a question that elicits lots of different sorts of answer is great
also I would like to ask questions that got answers in nim and rust :)
but I am not sure how to achieve that
 
@Rainbolt I started something for your zombie question, but it ended up getting tangled and I dropped it.
 
@sparr Ah, I see. So the graph is infinitely huge
 
6:21 PM
I need to start working on Marbelous again, and come up with a way to make it turing complete, so I can use it for more challenges.
 
maybe I could add something to the score of an answer depending on where the language comes in tiobe.com/index.php/content/paperinfo/tpci/index.html :)
 
to make marbelous fit a much higher computational class, maybe not turing complete, I had the idea to add a long jump instruction to send marbles a distance in a direction, or add a way to cause new devices to be placed on the board, such as a device that copies other devices.
 
@Sparr it is in general quite easy to make a language turing complete
 
simple implementations of either of those run up against the problem that all of the current built in devices take just one input, and all my ideas to implement either of those require two inputs.
 
6:27 PM
I think "turing complete" doesn't quite mean what some people take it to mean, and they are actually talking about a lower computational class.
for a language/machine to be turing complete, it needs the possibility to have unlimited storage, doesn't it?
 
@Sparr well.. most people assume that more tape will be provided as needed when talking about the complexity class of a real machine
@Sparr otherwise your desktop PC also isn't turing complete
which is just silly
 
also some languages require the input to be written into the program at a certain spot
 
they mean an infinitely scalable version with the same design as the physical machine they are talking about
 
not just more tape, but also the ability to reach the new tape
a small turing machine can use an infinite amount of tape
 
I buy a bigger hard drive when I run out of space. Thus my desktop is turing-complete.
 
6:30 PM
you never need to add more board to a marbelous program because there's no way for the program to ever utilize that additional space
 
@Sparr right.. the theoretical ability to reach it.. clearly real machines can't in practice
 
marbelous doesn't even have the theoretical ability to reach it
 
@feersum it is indeed except for the small problem of there being a finite amount of mass in the universe :)
 
I wrote a subleq interpreter recently, and had to make it use relative addressing to make it turing complete (I think).
 
@Sparr ah ok.. I see the problem
 
6:31 PM
with only absolute addressing in an assembler-like language, you can never use more space than the size of your data word
as soon as you add relative addressing, you're golden. a small program can utilize an unlimited amount of space.
 
@Sparr you know my assembly is so poor.. does i686 assembly allow for relative addressing?
 
every real computer you're likely to encounter allows for relative addressing in its machine code
 
phew :)
in any case.. I feel there should be a reward system on PPCG for people who use languages that haven't been used recently :)
oh simply people who post answers in languages unused by the other answers
 
I wish codegolf.com was still up. It was nice.
 
oh and.. I had another idea for a question that would only be downvoted :) If I asked a fastest-code question which said you had to use python
would people hate it?
the motivation is that is a compromise between fastest code and fastest-algorithm
more practical than fastest-algorithm
but requires more algorithmic thought maybe than just coding in low level C
 
6:36 PM
I'm sure some would hate it and some wouldn't, much like any other single-language challenge.
 
ok so the "some wouldn't" is already a data point for me :)
 
It'll probably get at least one downvote, though, so you'll probably just delete it ;)
 
"can I use numpy?" "can I rely on you using pypy?"
 
@Geobits only if it's the only vote!
@Sparr let's say yes to both
pypy +numpy don't really work of course
so you can use one or the other :)
I suppose cython and numba would have to be out though
what I meant was that you can't really use pypy with numpy
 
7:30 PM
@Lembik you could say that the code must run unmodified in cpython (the standard python interpreter), and that provided it does you are happy to run it in pypy.
 
Is there a common loophole posted somewhere on meta about abusing "any convenient input format" to hide code in the input format?
I've been trying to specify "delimiters but not operators" in my challenges to avoid that.
The marbelous room was automatically frozen :(
@MartinBüttner I'm thinking of ways to make Marbelous turing complete. Could I bounce a few ideas off of you?
@NathanMerrill @es1024 ^^
 
7:56 PM
@Sparr I can't find anything specific to that, but I would DV an answer doing that on principle, whether it's explicit on meta or not.
 
in some languages it's a little tricky
the distinction between a delimiter and an operator for various list declaration syntaxes in, say, perl is not immediately obvious
 
I thought you meant blatant, obvious loopholery. Maybe an example would help.
 
8:12 PM
Loopholery. Only on PPCG. books.google.com/ngrams/…
That makes me wonder, is "tomfool" a word?
Trickery : trick :: tomfoolery : tomfool
(gotta show off my third grade verbal analogy skillz)
 
I meant it as a nicer version of assholery, so I wanted it to have the same feel :P
Maybe I don't know enough about statistics, but wtf google? How does one result produce this graph? books.google.com/ngrams/…
 
One result? Where do you see one result?
 
Oh, click in the search box and hit enter again. I guess it doesn't show that from a link.
 
@Sparr go ahead
 
@Geobits I clicked on the link from there and got a full page of results to begin with
 
8:25 PM
That's... weird.
 
@Lembik It might help if you didn't post them on consecutive days. Rehashing the exact same task a day later is usually not all that popular. I think if you want to pull it off, you'll either have to post them at the same time and present them as sister challenges, or you'll have to wait a bit and still clearly state why you think doing the same thing with a different winning criterion is interesting.
 
I got the message saying only one result too
 
@Lembik Otherwise people might think you're just taking the easy we cash in rep with more challenges. (Which I'm sure you don't, but I can see people thinking that.)
Alternatively, just make up your mind upfront which winning criterion would be a better fit for the task. :P
2
 
@feersum That's odd, and not the same link I posted. I wonder why it redirected you away from the ngram viewer.
 
8:27 PM
I clicked on "assholery" at the bottom of the page, about in the middle horizontally
 
If you click on any of the years it redirects you
 
Oh, I thought "clicked from there" was from here. Yea, ngram viewer is just weird, I didn't make that word up for sure.
 
@MartinBüttner I'm thinking of a marbelous device that teleports a marble X cells in a given cardinal direction, where X can be up to 36 specified in the name of the device or it can be the value of the marble.
@MartinBüttner I'm also thinking of a marbelous device that creates new devices somehow, possibly by copying one or more existing devices to a new location, but I can't think of a good way to generalize that functionality for a device with just one input, unless we start getting tricky with something like treating a marble as an 8-bit array.
 
Definitely not the first to use loopholery, either, although I can't say I've ever actually heard that one used :D
 
@Sparr the simplest change is making marbles represent BigInts
you could also take the Fission approach and add stack and queue devices
 
8:45 PM
how might a stack or queue device work filling just one cell?
I'd like to avoid defining core devices that are multiple cells.
that's making all of my other ideas harder, too
also, :/ @ the statefulness that a stack or queue device would need to have, which no other device has
 
@Sparr non-zeroes push, zeroes pop
 
@MartinBüttner hi
I should have posted them as sister challenges
 
@Sparr haha, I knew that would be your main concern
 
my problem is that I think it is interesting to explore the same question from more than one angle
and no I have no interest in rep :)
4
 
^+1
i cant star more
 
8:55 PM
having a sentinel value of zero for pop means you can't ever have a zero on the stack?
that's... uncomfortable
 
9:07 PM
@Sparr why isn't it turing complete?
because of our integer size?
 
@Lembik reputation thing is the most nonesens that been ever created all over SE , whoever thought about it , he s mentally paralysed , otherwise i cant find convenient explanation , for anyone grabbing all these points from asking for help ???
 
@randomra I've modified my approach. It is still suboptimal for your example, but it saves 126 moves over all test cases.
 
Is it "a SQL query" or "an SQL query"?
In other words, how many of you say "see-kwool" and how many of you say "ess-kyoo-ell"?
 
9:25 PM
don't forget "squeal"
 
@Rainbolt both (not simultaneously)
 
i spell it squirrel
 
it's "a" for people who read out all acronyms in full, even in their heads
 
-3
Q: Solve encoded message

VizzerdrixI have this riddle, can you help me solve it? var message = new byte[] { 181, 18, // XXX, XXX, 181, 45, // XXX, XXX, 181, 59, // ...

 
@NathanMerrill Thanks to recursion it might be TC. But barring that, you can only store a finite amount of data on any given board.
 
9:43 PM
@MartinBüttner despite recursion, we don't put a limit on a board size?
 
but any given board has finite size, and you can't move marbles arbitrarily far away from the board
 
just like in real computers you have a finite memory, and you can't store memory in an arbitrarily large spot
 
Is anyone else working on Martin's combinatorics question? I've got as far as correctly counting the number of tilings without taking symmetries into account.
 
@NathanMerrill It's different in that the available amount of memory is hardcoded into the program code. It's not even possible to represent a program that could make use of an unbounded amount of memory.
It's like BF with a finite tape.
@PeterTaylor I think David (Carraher) was working on something, but he said he wasn't sure whether he'd manage to make an actual program from it.
 
can the amount of memory be arbitrarily large?
 
9:47 PM
I was hoping I'd get a submission from Ell, but no luck so far.
 
I wouldn't be against some "repeater" pattern
 
Ah well. I need to go to bed, but hopefully tomorrow I can finish the symmetry calculations and post a few more terms to see whether anyone else agrees or disagrees. (Then I'll need to optimise it).
 
where you define a section of X lines that repeat into infinity to the right
 
@PeterTaylor Looking forward to it :)
 
@trichoplax You said it's "a" for people who spell out "SQL"?
As in, "I wrote a SQL query." -> "I wrote a ess kyoo ell query."
 
9:58 PM
in full == "a structured query language query"
 
Oh, I somehow replaced "read out" with "spell out"
Darn brain helping me read faster
 
10:10 PM
I have a nearly irresistible urge to golf my submission to a code challenge. A 442 byte CJam program just feels wrong... :P
 
nothing can help better than glasses
 
@Rainbolt No I said it wrong - I meant in full as in what it stands for - oh Martin said it already
 
You said it right. I just can't read sometimes
Kinda makes me wonder how I write code all day
 
@Rainbolt write sentence, run test ;)
Mine still looks ambiguous though
That's the advantage with code - less ambiguity
I guess that's why not much poetry is written in code
 
10:30 PM
var myString = 1;
Test passes. Go home.
 
Nice
I was half expecting people to chime in with counterexamples to "not much poetry is written in code"
 
@trichoplax I was tempted
but I think there are one or two challenges for that actually
 
@PhiNotPi @Sparr I'm in the testing phase for code bots 3. However, I'm running into some problems. Firstly, there isn't an easy way to get randomness (you have to copy into a variable at the same time). Secondly, there isn't a way to identify which variable controls the pointer of your thread. Are either of these big issues, or workable (and interesting) issue?
 
@MartinBüttner The ones I remember seemed more like code in poetry, although I can't convince myself the distinction makes sense
 
10:55 PM
i made 4 lines code and all gone with wind
unplugged my computer accidentally
stupid electronic gadgets
 
11:06 PM
bring me back my code idiot computer (i know u hide it somewhere in ur RAM)
 
Have you considered writing in an environment that autosaves as you type?
 
i missed that
since i write in volatile sheets
v u ever wonder, if there s a kind of tool , which extracts some fresh data (in its raw unmodified form) , from random-access-memory ?
 
11:30 PM
@MartinBüttner: I'm thinking about your boggle board question and I was wondering about the last rule. The idea I currently have in mind would require that there are some short strings. Would that count as a forbidden optimization towards the test data?
Well, not require. But work better.
 
0
Q: Seven Slash Display

Calvin's HobbiesWrite a program that takes in a non-empty string of the digits 0 through 9 and prints how they would be shown on a seven-segment display using slashes (/, \). These are the precise digit shapes: /\ \ \ \/ \ \ /\ / \/ /\ /\ / \ \/\ / \/\ / / \/\ \/ /\ \ /\ \/\ \/ /\ \...

 
@NathanMerrill people thought so before that recovering files in intact header form from HD has been revealed feasible
 
@Agawa001 right, but you can measure the decay rate on DRAM
so, if it was feasible through some backdoor
then it would require special equipment
 
lemme gess .... viability of flip-flop information ?
yes , that wouldnt last for long time (constantly refreshed respectively to clock-signal)
 
I read an article, can't remember where
they said the maximum time they were able to extend the life was 10 minutes
 
11:46 PM
@Sparr perhaps have each marble be its own stack, and then have boards such as ++ change only the top value (the only device that has to be two cells wide would be one for pushing)
 
yes , flip-flops and registers , not like platters of hard disk , which have magnetic storage mechanism , canretain the information even durin very long periods
lemme cry my lines now
 

« first day (1591 days earlier)      last day (3253 days later) »