« first day (1300 days earlier)      last day (3543 days later) » 

3:04 PM
The What is a programming language? question has a lot more votes than all the answers put together. That suggests to me that most people are unconvinced. Anyone have an idea for a better answer?
 
I voted it up because I think it's a good question that should have a standard definition. However, my theoretical CS knowledge isn't enough (IMO) to decide the answer, so I haven't voted for one.
 
@PeterTaylor We tried to nail this down in chat several weeks ago and failed.
An important requirement (at least for the purposes of PPCG) that hasn't been mentioned at all yet, but I think would be useful, is that anything that counts as a programming language needs an implementation.
a) For testability, b) Because otherwise by allowing regex we also allow answers in EBNF and stuff like that.
 
Why is this question being asked in the first place?
Does someone need a reason to spit on someone else's answer?
 
Yes the OP himself on his own answer.
And I think it's a good idea, because in the future I'd like to have a reference when people say "you can't answer with a regex" :D
 
Can you execute a regex?
 
3:16 PM
Totally
 
Then it's a language...
If someone, somewhere has built a machine that understand the instructions you give it, then the instructions are instructions.
You can't say "instructions are not instructions"
 
So you're saying that the OP's rot13 is a language? You can clearly execute it, with the string as source.
 
@Geobits in that case you're not really executing the source though, right?
 
I thought a program was a set of instructions that tell the machine what to do.
 
@MartinBüttner I'd say the utility parses the input and executes instructions based on that. How is that different than some interpreted languages?
 
3:18 PM
So if ROT13 tells the machine what to do, it's a language
 
cat is a language too then
or any program that takes string input
or probably even any program that at all
 
(in case it isn't clear, I'm on the rot13 isn't a language side)
I'm pretty sure that all of our KotH controllers are languages too, if all it takes is instructions from elsewhere to qualify.
 
I've gotta go buy some snacks before I can't get back into the office
 
I'm not sure how you can draw an arbitrary line between print "hello world" and a full blown interpreter. print "hello world" is a machine that does one thing. But it's STILL a machine
 
I really think we need to distinguish between the program in question (i.e. the source) and the executing program.
 
3:21 PM
print input() is a machine that takes input and produces a simple output
 
machine != language
 
Sometimes program becomes machine. A compiler is a program on one level but also a machine on another level.
 
My argument against rot13 as a language is that it's incapable of computing even the primitive recursive functions. Can anyone think of something which should reasonably be considered a programming language but which can't at a minimum perform basic arithmetic?
(For the sake of disambiguation, let's take "basic arithmetic" to mean addition of natural numbers and multiplication of natural numbers)
 
@Geobits If you take machine to only mean physical hardware, then a program must obviously be written in machine code (binary). I think that view is wrong, because Java is obviously a language.
 
No, I'm saying that just because something is a "machine" doesn't automatically make it a "language".
I'd assume you could classify any actual language as a machine also, but it's a square/rectangle problem.
 
3:27 PM
I didn't ever say they were the same. I said the program and the machine were the same.
program != language
 
So why did that even get brought up? We were looking for a definition of language, not program :p
 
Because a program is written in a language
 
Ok, and...? I'm sorry, I just don't know where you're going with this, and don't see how it qualifies rot13 as a language.
 
I don't know what ROT13 is. Can we speak in terms of regexes?
 
> I thought a program was a set of instructions that tell the machine what to do.
> So if ROT13 tells the machine what to do, it's a language
 
3:31 PM
Does rot13 contain instructions that tell the machine what to do?
 
Sure
 
Then what is the problem here?
 
the instruction "A" performs the action "print M"
(etc)
(at least, it seems like one could argue that way)
 
@Rainbolt The text I quoted made me think you thought program = language.
 
So if I tell you what to do, would you think that I am claiming "I am you."?
I clearly meant "something written in ROT13".
 
3:34 PM
I clearly just read the words as written.
 
Do you want me to rephrase it or do you think you can handle it?
"If something written in ROT13 tells the machine what to do, ROT13 is a language"
better?
 
That's better, but my original point is that rot13 is just a cipher. The input isn't "instructions", just the encoded output.
It's like saying base64 is a language.
 
So would you disqualify every interpreted language? Only compiled languages should be considered a language?
 
I would argue on a simple level that something which doesn't take input isn't a program, and a ROT13 file doesn't take input so it's not a program, and thus ROT13 isn't a language
 
Why would you argue that? (the first sentence)
 
3:38 PM
That gets complicated though because there are Turing-complete languages which don't do I/O.
E.g. iota and jot.
 
@Rainbolt Where and how did I say that? I said ciphers aren't languages.
 
because it cuts out ROT13 nicly :P
 
@Geobits An interpreted language is essentially a giant cipher.
 
There are no "instructions" in the input. No branching, math, stacks, i/o, logic, nothing. It's not possible to include any of that in the "source".
 
@Geobits Would you be more specific? Like "If it can't branch, it isn't a language."
VisualMelon covered the i/o part already
But couldn't back it up
 
3:41 PM
I'm saying there isn't anything it can do. For every input, there is exactly one output. If you call that a programming language, it's the worst one ever.
 
I saw others try to disqualify the turing incomplete languages, but couldn't give a reason why
 
if stuff that doesn't take input is allowable as a language, then I'm at a loss and will run away (thusly I will run away)
 
So yes, I'd say branching is integral.
 
@Geobits BINGO. Oh but wait. It CAN do something if it answers the challenge, correct?
If a program that does nothing answers the question, it was a poor question IMO
 
No, the rot13 program does it, and has nothing to do with contents of the input.
 
3:43 PM
@Geobits That was also in my answer on meta. Although I think being a bijection is far stronger a condition than required to rule out a tool from defining a programming language.
 
So if he wrote a program that did the rot13 cipher, fine.
 
you could just define a programming language as something turing-complete?
 
@PeterTaylor Can you multiply natural numbers with a regex? ;)
 
that does pretty much what you would like, I think
utilities like Sed and Date are ruled out by that rule, and it seems pretty egalitarian (i.e. no preference for a specific language or anything)
 
@MartinBüttner I think you probably could. Or at least you could test whether the product of two natural numbers is a third.
@EricTressler sed is Turing-complete.
 
3:47 PM
I mean, sure, it includes Life, but who's going to answer a question in Life
fine! then allow it
I didn't know that, but if a language/utility is Turing-complete, why rule it out? That seems like a fair distinction to make
 
But there are serious golfing questions which don't require Turing-completeness to solve. E.g. one which is currently going through the sandbox.
 
@PeterTaylor Ah yes, the test would probably work. Interesting :)
 
It sounds like the general consensus is "I think it should do [insert arbitrary qualification here] because otherwise it would be a useless language." It has to have I/O, it has to be Turing complete, etc. They are all so arbitrary.
 
@PeterTaylor That may be the case, but you have to draw the line somewhere; being Turing-complete seems like a fair line
 
@EricTressler but I like my regexes :(
 
3:49 PM
Heh. I love that answer, but mostly because it's ridiculous
 
@EricTressler, have you seen the current answers to the meta question? One of them has a bit more to say on non-TC languages.
 
no, I have not. I'll read them
 
Turing-completeness shouldn't be necessary in my opinion--plenty of tasks can be solved by sub-TC languages
 
What about, "it needs be able to loop and branch" (not necessarily loop an unbounded amount of times)
 
That pretty much boils down to primitive recursion.
 
3:51 PM
In that case, "What Peter said"
 
@FireFly my objection was that subTC languages are usually specific utilities
so they usually solve a problem trivially, if at all
 
If you can loop, you can branch automatically right? Repeat X times, where X is 0 or 1.
 
@Rainbolt only if you can decide the number of iterations dynamically
 
so they don't really lead to cool answers; they lead to people noticing that Date can solve X and posting it
 
But I guess that's necessary anyway
 
3:53 PM
@PeterTaylor could you point me to the relevant meta page?
 
Otherwise a loop isn't more powerful than concatenation of code
 
I think the problem with this entire discussion is that someone posted a challenge for which an existing utility can solve in one character. It was a poor challenge to begin with.
2
 
8
Q: What are programming languages?

Ingo BürkOften, answers to questions asking for "programs" or talking about "programming languages" utilize things like sed, awk, … in order to get around having to write an actual shell script. Therefore, a question comes to my mind: What qualifies as a language? Sure, ultimately the OP can defines th...

 
@MartinBüttner thanks
 
@Rainbolt Then you're saying that a challenge asking for a quine is a poor challenge.
(regardless of the fact that it's a duplicate these days)
 
3:54 PM
echo is my language for quines :)
 
Not cat?
 
I picked one of many
 
(gotta golf the program name if you can't golf the source :D)
 
well, I basically agree with Peter there
 
@EricTressler requiring TC-ness would rule out answers in e.g. Agda and Coq (not that we have seen many of them, or that they are useful for golfing, but still)
 
3:55 PM
A "language" that isn't turing complete isn't really a language, but a utility, in my stupid opinion
 
:\ I disagree
 
I know the words here are a bit fuzzy
 
I think there's a limit for how powerful something has to be to be a programming language, but it's definitely below TC
 
Well, I'm not religious about it, so I don't hate you for disagreeing
 
@EricTressler You've read GEB, right?
 
3:56 PM
Yea
 
I should read it
 
you should. it won the Pulitzer for a good reason :|
@MartinBüttner what are you getting at?
 
@EricTressler BlooP
I would certainly consider that a programming language if it had an implementation.
But it's not TC.
I can compute primitive recursive functions (like Peter suggested)
 
I think people have implemented it
 
@EricTressler Then it's a programming language ^^ (imo)
 
4:01 PM
That's fine, but I wanted to draw the line between "general purpose" and "utility"
I couldn't think of a better way to do it
it's not fair to bring rot13 to a codefight
 
Yeah that would certainly be ruled out.
As would date and cat. But you could answer in Minecraft.
And in MTG.
 
I read about 60% of GEB. Actually, it's probably more accurate to say that I got about 60% of the way through, reading about 40%. I think I would have enjoyed it as a 15-year-old, but as a CS graduate I found it excessively slow and the storytelling chapters annoyed me.
 
(@Rainbolt still waiting for your first MTG challenge and/or answer)
 
That's cool, IMO. I just wanted to delineate crap answers from good answers
not that it really needs help; but using an existing utility to do all the heavy lifting is usually crap
 
@PeterTaylor I do like the dialogues, but some of the actual chapters do drag on...
 
4:04 PM
@PeterTaylor I also skipped the fiction bits
 
@MartinBüttner The guy who proved that M:tG is Turing-complete is my ex-landlord and the person who taught me to play.
 
I always liked Simon Singh's books, I felt they were a nice balance of relevant history and real stuff
 
@PeterTaylor did you depose him as landlord?
 
I wish I had such a landlord :D
@PeterTaylor Could you point me to a list of problems which all primitive recursive functions can be reduced on by any chance?
 
4:06 PM
It was a short-term arrangement: he bought a flat about a year before he was going to get married; so I lived there for about 9 months.
 
@VisualMelon I don't like Simon Singh; please try John Allen Paulos
 
I don't read much these days atall
 
@MartinBüttner What do you mean? PR-hard functions?
 
basically
 
I raced through Survival of the Fitest because my former PE teacher lent me it, and I'm half way through a book about railways which I've been reading 6months
 
4:08 PM
i.e. a simple way to check if something is PR
"Can your language solve this problem (one of these problems)? It's good to go."
 
@VisualMelon well, what I really meant is that if you like Simon Singh's books, John Paulos is similar and with a little more credibility
 
Interesting question. The Complexity Zoo page doesn't list PR-hard
 
Singh is basically a whore, in academia
 
I don't know if I'd appreciate his books these days, but they are what got me into maths, well, that and Microsoft Excel
(first language: VBA)
 
There is the basic definition: 0, successor, projection of a tuple onto one dimension, composition of functions, primitive recursion (i.e. bounded loops).
 
4:13 PM
Hm yeah, but it might not be so easy to map these to constructs of the language directly.
Which is why I thought being able to check that a certain problem can be solved would be easier.
Especially since (e.g.) regex can't actually compute primitive recursive functions but can only test results for correctness.
 
4:28 PM
@PeterTaylor Would it make sense to post this on cstheory.SE?
 
Regex can do 0, successor, projection. And addition, if we allow replacements: s/+// does addition in unary.
I don't think it would be welcome on cstheory. You could try cs
 
@PeterTaylor I keep forgetting that there are two of them.
 
s/+// would be subtraction in unary, no?
 
@FireFly no
111+1111 --> 1111111
@PeterTaylor There's no way to do multiplication though, even with replacement. But checking a correct multiplication should be possible as you said.
 
oh
 
4:37 PM
Possibly not. /// is TC, though.
 
interesting :)
that's quite different from solving things with a regex though
 
4:52 PM
hey guys, I'm interested in getting this question out of the sandbox and onto the main site
2
A: Proposed Question Sandbox - Mark XIV

chilemagicI got this idea after answering Calvin's Hobbies question which reminded me of a cool programming trick I learned a year ago that can be used to satisfy program number 2. Definitely open to feedback, different formatting/wording, or a different title. Please let me know what you think! This woul...

any advice on improvements that could be made?
I'm not sure of what changes to make it better, or if I should just post it as it is
I'd rather not let it get lost in the sandbox because the main site is slow on questions sometimes and I want to help contribute
 
5:23 PM
@chilemagic Uhh, that looks impossible in... most languages that I know of
No function calls in Python 3, JavaScript, Java, C, C++, C#, PHP, etc. so can't print anything
No way to get any strings in Python 2, so not that either
 
right its not the most inclusive question, but you can still post answers for half of it if the other half doesn't work. I think we might see some interesting/creative solutions though.
it challenges you to think of a language where you can make it work
 
5:38 PM
@chilemagic reminds me of perlmonks.org/?node_id=290607
 
@firefly yeah for sure
 
Some concatenative language might fit the bill for no-symbols
I don't know enough Forth to know if you need any of the words containing symbols in their name
 
I love when testing submits a bug like "Can't install application." followed by "Can't reach page A." followed by "Can't reach any pages."
And that's not sarcasm either. I get to resolve three bugs at once.
#RealLifeRep
 
@PeterTaylor @Rainbolt @Geobits @EricTressler What about providing several problems, and if the language can solve any of the provided problems, it's valid? That is something like "prove that your language can compute any primitive recursive function", or "show that you can accept context-free languages" or ...
 
@MartinBüttner So instead of "It must do A, B, and C.", you are proposing "It must do A, B, or C." Is that correct?
If I understood that much, I think it's an improvement
 
5:49 PM
Yes
 
I'm worried that someone a very long time from now will have this thrown at them and they will think, "Who on earth chose these problems?"
Sure, your whatchamacallit (holy crap, I actually received a spelling recommendation for that word) can solve that problem, but it's not on The List so it doesn't count! sticks out tongue
 
The list could always be amended by community consensus.
(that is by posting a suggestion in the chat and no one objecting :D)
 
Everyone in the community seems to have their own ideas of what capabilities a language must have. Someone wanted to draw the line at Turing Complete. Someone else wanted to put it somewhere below. I didn't want to draw a line at all.
So I think it will be hard to get concensus
 
Am I the first one to say that if your challenge can be solved by cat, it's not a very interesting challenge in the first place?
 
@Doorknob No. Look at the transcript.
 
5:54 PM
No, just the most recent.
 
@Doorknob Since when are quines uninteresting challenges?
 
@Doorknob Well ok, I didn't say the exact same thing (cat is three characters) but the idea was there.
It's so... Texas.
 
Just plain quine is uninteresting, but when you start to add interesting conditions, they become interesting as well
 
@Doorknob Just because it's been done before doesn't make it uninteresting.
 
But anyway, that's not the point
 
5:57 PM
That just makes it a duplicate.
You're saying someone should go back to the original quine challenge and say, "I'm using cat, any program is a quine.", and that that challenge was hence uninteresting to begin with.
That being said, the original quine challenge should have been
 
Would it be more correct to say that "Write a quine." is an interesting problem to solve in some languages, but not in others?
 
@Doorknob Also many interesting variations (especially any restricted-source version) of a quine challenge are still solvable in cat.
@Rainbolt Possibly. But it's an interesting problem in any "language" that does something else than printing it's own "source".
 
Right, which is because cat isn't a damn language :P
 
@Geobits hence the quotes
 
I think regular PCs have basic CPU instructions like add and multiply, correct?
 
6:03 PM
Oh neat, a necromancer badge
 
yes?
@FireFly congrats! :)
 
What if you added "print" to those instructions.
Would cat be a language then?
 
@FireFly I only get these on meta :D
@Rainbolt I don't understand
 
You could call cat an instruction in a shell language, but it's not a language in itself.
The comparison to add/multiply makes me think you see + as a language.
 
I'm trying to figure out how you guys determine these arbitrary qualifications. I thought maybe they were based on what CPU instructions you use
Someone earlier mentioned multiplication
I've also heard i/o and TC
 
6:06 PM
Okay, I think the real problem here is being able to create a language that can anticipate a challenge in advance. We should try focusing on the actual thing we're trying to solve, not a simple side effect. For example, let's say I make a new language called "PythonUpgradedWowSoAmaze," which is exactly the same as Python except that an empty program or a . does {insert task here (quine, etc.)}. Your rules are absolutely worthless now.
 
The disputed "rot13" answer specifically pointed out that it wasn't a language invented for the challenge, so it doesn't help in invalidating that answer (if that is what we want)
 
This is basically the "using language features after the challenge was posted" problem in reverse.
 
Haha I like that. With Doorknob's logic, PythonPlusCat is obviously a language. So is PythonPlusRot13
 
I'm thinking e.g. answering with "cat with any source code" for the quine task would fall into "use a built-in feature that solves the task at hand"
 
@Rainbolt Exactly, and your "problems that language must be able to solve to be a language" is useless now
 
6:09 PM
I wouldn't even care if an answer was submitted using rot13, but it should be counted as a bash/shell/whatever "program" and should count the command(s) and input, not just the input to the command.
 
@FireFly I'm not fan of that being a loophole though and it's very disputed.
I think golfing is a lot about choosing the right tool for the job.
 
That's like answering a quine with 1 and saying System.out.print("") is the language with input going between quotes.
 
Could you link to a compiler?
One that was freely available before the challenge was posted?
 
@Rainbolt Java is the compiler. The syntax for running it is provided by Geobits.
(plus some more boilerplate)
 
@Rainbolt there no rot13 or cat compiler either.
 
6:11 PM
But if you download Java, you have to manually type System.out.print("")
 
@MartinBüttner I don't know, I think it's clear that if a task is about implementing X, then simply using an existing implementation of X is a rather boring answer
 
Just like you have to manually type out rot13
 
@Geobits There IS a machine that will run cat. Not necessarilly a compiler, but maybe an interpreter
 
@Geobits What's the difference between echo asdf | python rot13.py and echo asdf | rot13 blank.r13?
 
But I guess it depends on the language, and a question about implementing X could point that out explicitly instead
 
6:11 PM
@Rainbolt If I download bash/java I have to manually type javac MyClass.java; java MyClass to run a Java program.
@FireFly Boring, but not necessarily a loophole, imo.
 
@MartinBüttner Or you could just download netbeans, copy and paste the code, and run it. I could probably even link to an online Java compiler
 
@MartinBüttner fair.
 
@Rainbolt So a language needs a GUI/online compiler/interpreter to be valid?
 
@MartinBüttner I didn't say that
 
@Rainbolt Then my point remains with any other language where that's not the case.
 
6:13 PM
@MartinBüttner It's not the case anywhere. A GUI is never required for a language to be a language...
 
@Rainbolt But then your netbeans point is ... pointless
If a language isn't required to have a GUI, then potentially I have to manually type something to run the code in that language.
 
@MartinBüttner Geobits couldn't link to a netbeans + System.out.println("") package, could he?
 
So having to manually type something can't be an argument against something being a language.
@Rainbolt You can't link to bash + java + the javac FileName; java FileName package either.
 
What if I wrote a compiler that was simply JavaPlusSystemDotOutDotPrintln?
 
The only difference is where the manually typed stuff goes.
 
6:16 PM
With JavaPlusSystemDotOutDotPrintln, you have two options when writing your program. You can simply type stuff, and if it looks like a Java program it will compile with Java. Otherwise, it will print it to System.out.
This was exactly Doorknob's point a moment ago
If Java is a language, then Java plus anything is also a language
 
maybe we should just Ctrl-F for Programming Language in Wikipedia, and apologize to all the EsoLangs
 
@Rainbolt Is your point that literally anything I type into my computer is now a language?
 
@VisualMelon You can include esolangs.org's pages
 
@Geobits Only if it can be executed.
 
@MartinBüttner I forgot that was a thing, good call
 
6:18 PM
Everything can be "executed" if that's the only distinction. If I type garbled junk into my shell, it gives me output.
 
@Rainbolt Does "cat" execute its input?
 
I don't think the distinction is important
 
@MartinBüttner Yes
 
Well, I guess it does clutter the answer list.. so there's that
A "boring" answer, that is
 
so ffmpeg is a language? apt-get? dir? ls?
 
6:20 PM
You can execute anything that weighs the same as a duck
 
What is your actual question? I don't know any of those terms
 
I threw dir in there as a nod to Windows just in case. Shell commands, installed programs, that sort of thing.
 
@Geobits Is there a compiler or interpreter somewhere that just produces "dir"?
Regardless of what you type?
Let me explain my thought process behind what I consider a language
 
@Rainbolt I think you could probably write one
 
@Rainbolt Okay, so basically your point is "anything is a language" unless you can show me something that can't be executed. In that case what do you propose to do with answer that just use an entire program built to solve the task at hand?
 
6:22 PM
Binary is obviously a language. It runs on a machine. (stick with me here)
 
This argument is still going on?
 
Someone somewhere wrote something in binary to make the rest of our lives easier. It was called an interpreter.
Anything written that runs on that interpreter is now valid code.
Someone somewhere wrote code on that interpreter to make the rest of our live easier.
 
Just be like the supreme court, know it when you see it, done
 
It was called a compiler. Anything written on that compiler is now valid code.
 
@EricTressler I wish it was that simple :(
 
6:24 PM
You can write new compilers on a compiler. This is called bootstrapping. Anything written on the bootstrapped compiler is ALSO valid code
 
@MartinBüttner What are the edge cases
 
Repeat this as many times as you want. You'll end up with a compiler that compiles anything or nothing or somewhere in between. Any language that can be used with that compiler is a language
 
@EricTressler The problem aren't edge cases. The problem are people who want to use rot13 and cat as valid submissions to decent challenges.
 
Add arbitrary qualifications: we get whatever the hell you guys are talking about.
 
@Rainbolt IOW, don't disallow answers in "cat", "rot13", ...
 
6:26 PM
I addressed those earlier
 
@FireFly I tried saying that earlier. People can't follow me apparently.
 
Sorry, I haven't been paying attention all the time
 
@Rainbolt .....
 
@Rainbolt So you're saying anything that can be "executed" by a program written before the challenge is fair game?
 
Basically, every time you write a program, it can be used as the platform for the next program.
It's been done before, and generally accepted.
@MartinBüttner That was not my rule, but I support it.
 
6:28 PM
@Rainbolt I think it follows from what you're saying.
Anyway, gotta eat some dinner. See you later.
 
what about the mechanical turk
 
@EricTressler Not sure how you would post that as an answer, but it's no less valid than a bunch of punch cards.
 
Yes it is
 
If someone changes the question to "What programming languages should we throw stones at for arbitrary reasons?" I will drop my side of the argument.
 
this whole forum would go to crap if people just invented "LangThatHandles4ManStandoffs" and posted their source in that
 
6:32 PM
@EricTressler If it says I am wrong, then it isn't a real language.
Would it disable the Internet for people who are Wrong?
 
what about a language that throws away the input, and just says that you're right
 
@EricTressler Can we invent this? I think we need to enumerate some design goals.
 
My only point was that I think at the bottom, actual people need to decide what a language is
And it's not hard
 
This whole site is based upon disregarding rules, so you'll get tons of it, but a lot of it can just be thrown aside
 
6:38 PM
I can already tell just from our collaborative efforts on my Google docs that a shared Minecraft server would be hugely successful on PPCG.
 
I love minecraft
 
@EricTressler Have you played Direwolf20 (or any of the FTB modpacks)?
 
@MartinBüttner I've finally done some proper thinking and in a couple of days I hope to be posting a new DominoSolver which will solve the boolean operations in 2^m + (m-2)(m-1)/2 operations (74 for m=6, I think)
 
I haven't heard of them
 
6:43 PM
@EricTressler no, sane users would downvote and move on. Hopefully people would get the hint.
 
I've not yet worked out if this is any better than my current boring solution, but I'm going to blindly assume it is
 
@EricTressler Someone decided to start collecting mods that work well together, tweak them so that the game isn't too easy or too hard, and then package them together and distribute. It turned into FeedTheBeast modpack, which was popularized on YouTube by a user named Direwolf20.
They've been through several iterations now, trying to keep up as Minecraft updates and new mods are released
 
I tried a new mod called Captive 3 or Captivate 3, but it was broken
 
@Rainbolt collaborative redstone solving of PPCG tasks?
 
The idea was that you're stuck in a (1+0.5*numachievements) box, but I could move out of it
and the box just moved with me, which I don't think was their intent
 
6:46 PM
@FireFly Someone already created a dual core computer in MC. So we can just use that instead.
If we create the Internet, we can just move PPCG to Minecraft entirely
 
@VisualMelon Nice :)
 
7:46 PM
0
A: Proposed Question Sandbox - Mark XIV

Nathan MerrillCode wars Your goal is to write a bot that modifies other's bots' code. A turn will consist of each bot performing a line of code asynchronously. After 5000 turns, the bot with the most flags wins. There will be 50 copies of each bot. The world will be toroidal, and the bots will be placed r...

 

« first day (1300 days earlier)      last day (3543 days later) »