Discussion for Evolution of "Hello World"

Huzzah, should we edit it then?

@Sp3000 Go for it

Done

nice!

@Sp3000 ok to work towards your secret mission

from here:

#//echo o[.>]tac";cat<<;#&&alert
println("Hello World!");

you can make 3 moves on the first line, what would you pick?

12:04 AM

@chilemagic Looks like Dennis fixed his

chilemagic

@Calvin'sHobbies yeah I saw!

iokanuon

The first line of Denis' code still throws Missing name for redirect..

@chilemagic Ahaha thanks. The hardest part will be program termination, so I'll need ?!;o at one point :P

Think at this point people would ratherthe first line be cleaned up though :P

chilemagic

@Sp3000 good point I just cleaned it up a little. Good luck with getting fish in there!

wow....

used a language already on there XD

12:26 AM

I wonder if this type of challenge could benefit from some sort of external ticketing system - take a ticket to claim rights to do the next answer, but if you don't answer within <some amount of time> then your ticket expires. You can't take a ticket twice within a <some amount of time> period.

Otherwise it'd be annoying to type up a really nice answer only to be beaten to

Is answer 21 Haskell a complete program?

It works if you just run it in the shell

Or is that not allowed?

I didn't write programs in Haskell. So I'm not sure.

@Sp3000 Ideally they should be able to run as standalone files but it's not a huge deal.

k - I think if the shell wasn't allowed a number of them would be invalid actually...

12:40 AM

Yeah, Maybe I should have been more specific on what "program" means. Oh well.

Then we should have been doing something wrong with the deleted batch answers.

May I should add a note that mine needs to be run in the shell too... awkward

On a completely unrelated note, I think there are going to be a lot of people tied for first place. Or maybe the winner will be the person with the right online schedule. That is, I don't think combining "most answers" and "only once every eight hours" works well for scoring, although it's still a fun challenge.

Everybody can be a winner :D

Unless I'm mistaken @user23013 is winning. But Nit and I are close behind. Though I promise i won't be answering a third time.

12:50 AM

Or not wrong, because they have called the language "Batch", which means the batch file.

I already found it not easy looking for another language now.

There's still a lot on Rosetta Code though

@Ypnypn I think it'd be a lot crazier without a time limit. There would be collisions all the time. Though I agree, it would be better if the scoring were more based on effort, not attentiveness.

Aaaaaaah I CAN'T POST A FISH ANSWER

Applescript's "log" makes Fish possible from here but I can't post it

/#//echo o[.>]tac;cat<<;#&&alert (printf
>l?!;o
\"Hello World!"

D:

... I regret posting that Racket answer :/

1:14 AM

Waiting for some "main" to finally arrive...

(){;}is easier

Maybe I should award a bounty to whoever gets C, then everyone would help you guys towards it...maybe.

It'd be like the game Munchkins - whenever someone got close to C, the sabotaging would begin

If only someone would change "pr" to "ma"...

Ahaha that's an idea - slowly use comments to get to Java

@Calvin'sHobbies If we have interesting answers but never manage to fit them in again, can we post it after this is all over as a Community Wiki post?

1:25 AM

Is my Pixie submission OK or too rule-bendy?

(retract it or not?)

@Sp3000 A comment or note in one of your answers would be better since there are already SOO many answers.

A: Evolution of "Hello World!" (testing new type of challenge)

0

Answer 55 - Pixie "//echo o[.>]c;cat<<;#&&alert"(printf "Hello World!") Distance 3 from answer 54 The language is in early alpha and seem to have only REPL. Example session: $ ./pixie-vm Pixie 0.1 - Interactive REPL (linux, gcc) ---------------------------- user => "//echo o[.>]c;cat<<;#&&a...

There is a plenty of unused distance, so I can add "main" there as well.

@Calvin'sHobbies kk, will do if I don't manage to squeeze ><> in :P

@Vi. main() would be interesting :P What's the "12" in the output though?

@Calvin'sHobbies, Use "oldest" answer sorting mode.

I can't tell which part's actual output and actual input

1:28 AM

@Sp3000, The result of "printf" - number of output characters.

Basically it should run like Clojure. You can save the program as hello.clj as run as Clojure.

Ah k

@Vi. It doesn't look terrible but it is a little sketchy.

If it gives Hello World! when run as a file then it's fine.

@Calvin'sHobbies, There's currently no option to run scripts from file. But when it will appear, I expect the script to output just Hello World!.

Why don't you expect the rest of the output (the first string, and 12) to appear?

@Ypnypn, Because of they are printed by "P" in the REPL, not by the program.

Likewise the REPL also prints //echo o[.>]c;cat<<;#&&alert.

So, should I:
0. Retract the submission;
1. Leave it as is;
2. Edit it to include "main" in the "comment" string literal?

1:39 AM

@Vi. Retracting (or using new lang) would be best since technically it's invalid. I won't force you since you've obviously tried to make it valid. Thanks for asking though. :)

Are "pre-alpha" languages allowed in general?

Where should I put retracted solutions, for them to be not part of chain, yet available if somebody wants to explore the "side branches" of the development?

@Vi Languages that were invented after the question are never allowed. I'm not quite sure of our stance on things in development.

You could use something else the leave Pixie as a comment? I haven't tried, but I'd think something like Scheme might work

Like ClojureScript? Or Clojure.NET?

It whould be boring...

As long as it's different enough from Clojure (#36)?

1:48 AM

Is it too soon for another question with incremental-answers like this? I really want to be sure the genre can work.

What's the idea? I'd be pretty interested in another one :D

OK, leaving the code intact, adding Clojure.NET as fallback language.

No idea yet (and I'd wan't it to be a surprise :P ), but I'm considering

Ahaha k :P

I wonder why this isn't on the Hot Questions list.

1:55 AM

@Ypnypn I know I saw it there yesterday

I have found the question from Hot Questions.

A: Evolution of "Hello World!" (testing new type of challenge)

4:54 AM

@Sp3000 Oh I have recognized your avatar.

Feeds

Quincunx has removed Quincunx from the list of this room's owners.

Quincunx

Why did I not remove myself earlier? This should be Calvin's Hobbies' room.

I guess it was a good idea to create a chatroom for this question.

Optimizer

Wow, there was no HTML answer till now !

0

Answer 59 - HTML What? No HTML ?? <echo o[.]c;cat<<;#&&alert" ">Hello World!</vsh Distance from Answer 58 : 6

A: Sandbox for Proposed Challenges

5:27 AM

The person who added @echo o for Windows Batch has failed completely now.

Beta Decay

8:54 AM

Does anyone have any helpful input on this?

0

Evolution of Tetration domino-coding N.B. This is another version of Calvin's Hobbies new challenge idea. I won't go ahead and post it until I get a full thumbs up. If not, I'll convert this to code-golf. Tetration Tetration, represented by na, is used to represent large numbers. Basically, ...

"50% of the characters" - does that mean anagramming? You might want to clarify a little

Beta Decay

Is the example clear enough?

9:10 AM

Well what I meant was something like "int main(){ pe = 991;}" and "ma=1; print 9e9"

Also, comments aren't the only way of having redundant code

Beta Decay

How to control what is used and what isn't?

Especially if the code is written in some unknown language.

A: Sandbox for Proposed Challenges

10:39 AM

discuss

0

Regex Domino Golf domino-coding code-golf regular-expression I just wanted to record this idea here before I forget it. However, I will hold off posting it until I'm convinced that domino coding is actually a good idea for PPCG (which I'm currently not). Feedback under the premise that domino c...

10:49 AM

@MartinBüttner The regex must differs from all previous regexes, or some regexes behave differently in two flavors will make it stuck.

A: Sandbox for Proposed Challenges

11:10 AM

Haha, glad to see others are trying their hand. Here's one I had high hopes for but now I'm not really happy about.

0

So I had written this up as another domino-coding contest but the more I wrote the more I felt it was too complex rule-wise and might not work anyway. I might scrap it but I might as well see what others think. Help Grow the Communal Code Blob [code-challenge] [printable-ascii] Intro Defi...

Here's my understanding of the four points: "Pick four distinct ASCII characters and a previous answer. Put two of the characters in the same line as the answer you chose, and the other two characters somewhere else but not on the same line as the answer you chose. The result must still be a blob and running the line you chose (now updated) with leading/trailing whitespace removed prints N to STDOUT. "

Is this right? (just checking)

@Sp3000 No, not at all lol

You always take the blob from the answer right before yours

and choose your own line to edit

Ah, oops I confused line with answer for some reason

You don't worry about what the other person chose for a line

No prob, I hope it makes sense, I'm not happy with the explanation anyway

I see...

Ahaha no, I just confused myself when trying to type that up. :P

So the example you gave is a tree - can you have something like, say:

01
20

11:25 AM

You need one more character, but yes

Just checking the resulting blob doesn't need to be a tree either :P

And we can use any language that works?

Yes

And it's not a tree it's a blob :P It's fine as long as everything is orthogonally connected.

Ahaha k... so I might have missed something, but is there something stopping me from just doing +1 to print N, and -0 to not affect another line?

And then everybody just does this

You mean like

0

then

0+1

XX

then

0+1+1

XXXX

Yeah that

11:33 AM

You could but the language has to change, and not everyone would do that anyway.

Hmm, well it sounds to me like you could just find 4 languages which accept 0+1+1+... and it would work

I know not everyone would do that, but you said "communal goal" so I was wondering :P

Yeah, possible but unlikely, one or two different chars on that line could mess it up (and 4 was random and could be increased)

k :)

@MartinBüttner I think your challenge could work (despite minor plagiarism :P), but how do you provide incentive for the regexes to be as short as possible? I don't know if the byte count tie breaker is enough.

1:21 PM

@user23013 Thanks for catching that!

@Calvin'sHobbies Sorry about the plagiarism :P ... but I figured why reword your spec if it's really clear as it is. And it seemed to be common practice in the other not-so-common challenge types like rosetta-stone and cops-and-robbers

Hm, good question about the byte count. But with the 8 hour limit I think ties are very likely. And I'm not sure about another scoring which wouldn't incentivise not answering at some point.

Btw, I also considered reducing it to 4 hours... I'll have to think about it.

How about, no time limit, but you can't answer if you've answered in the previous 5? So if you do #3, you can't go again until #9.

2:01 PM

How about changing from AND to OR in "8 hours AND not two answers in a row"?

How about: ((8 hours and not two in a row) or (not five in a row))

This is equvivalent to just (8 hours and not two in a row), because of 5 in a row implies 2 in a row.

I mean before posting your answer after enother your one, you need to wait for 4-8 hours. In order to post 5 in a row, you need to wait 4*8 hours total.

This allows a "lone ranger" to contunie the fight as long as he wants when everybody else lost interest, yet still preserving competition and participation if he is not alone.

@user23013, Or you meant (wait 8 hours || don't post twice in a row) && (don't post 5 times in a row)?

I should mean not to answer if you've answered in the previous 5. Sorry for my English.

@user23013, Maybe in case of "5 answers in a row, with appropriate time intervals" the fifth answer should be accepted and the challenge considered finished?

2:19 PM

5 answers in a row is really not easy...

2:40 PM

BTW whould the current challenge be more funny if deleting characters were prohibited (to make the code ever-growing)?

Changing printf to println wll be impossible in this case.

((4 hours and not two in a row) or (not answered in the previous 10)). I think this is better. But people with the right schedule still have the advantage.

3:26 PM

I'd prefer answers to compete and not users. That would really make any restriction on answering less problematic.

@user23013 I think this would work well.

FireFly

3:56 PM

Argh, I planned to use Burlesque but now that is taken

FireFly

4:08 PM

@Vi. I thought about that a bit yesterday, but I don't think it'd work very well. The area surrounding Hello World would get really crowded to the point where too many languages would be isolating. And developing a second "Hello World" substring over the course of many submissions just seems too unlikely to work

Can there be a "domino"-style challenge that encourages cooperation between users?

4:32 PM

@Vi. It would be hard to design well, but it's certainly possible. For example, give answers points for making the subsequent answer do well.

FireFly

@Calvin'sHobbies mentioned in the comments that they thought about allowing multiple responses to an answer, so that the result becomes a tree

Wtih something like that you could incorporate cooperation by having different branches compete against each other, if that makes sense

...or perhaps you'd simply get points based on how many branches branch off from your answer. Though, that might just mean that everyone decides to branch off of the first answer (because it's easier). Hm... okay, that wasn't very well thought-out

4:48 PM

I also though about a tree. But the current "linked list" format is already a stress for StackExchange Q&A engine (and works only with the help of "sort by date" feature). "tree" whould be even messier.

Maybe CodeGolf actually needs more flexible engine, not just Q&A?

A possible scoring for the regex challenge: for each answer, you score is regex size / 20 (or some other number). your total score is the product of your submissions. consequences: incentivises both answering multiple times and golfing as much as possible; answers longer than 20 (or some other number) are detrimental to your score.

@Vi. that's not going to happen.

that's why I'm still not sure this challenge format is a good fit for PPCG

both a list and a tree might work with a userscript which generates the list/tree automatically

Maybe there can be a separate page using StackExhange API to show multitude of answers in more meaningful (i.e. tree) form?

Are there already any bets about where the "Hello World" challendge will stop?

5:56 PM

@Martin That sounds workable.

@Ypnypn Your challenge also sounds like an interesting take on the genre.

Scimonster

Are there any plans to do more person-above-me challenges like this?

there are 4 proposals in the sandbox now

but not everyone is convinced yet that this is actually a good fit for the site ;)

6:59 PM

@MartinBüttner There are already challenges that require pre-existing "system" to work (i.e. king-of-the-hill challenges). The "tree" challenge can also just require a "system" (but not in form of framework for solutions to be run by, but in form of userscript that shows anwers in a particular way).

@Vi. KotHs still fit the format better... each answer stand on its own... there is no harm in posting answers at the same time, and there is no inherent order to answers. even with a userscript or something, there's no way to prevent the issues that can arise during domino coding

For the tree challenge, the timestamp of an answer is not relevant. Each answers should have a link anyway. Being too late is not a problem.

Now I'm actually thinking about not just a tree, but DAG system of answers (posting root answer gives 0 points, posting answer with 1 parent gives N points where N is height of the tree from the answer being posted; posting a "merge" (multiple parent) answer gives N*M points)

ah yeah, that could work. the only thing would be to require people never to edit submitted answers (also something the system provides which can't be prevented). likewise deletion of answers would be a problem.

I don't like deletion of content anyway (I prefer moving to "graveyard" or graying out instead)

@MartinBüttner The link can point to a particular revision instead of just the answer if needed.

@Vi. that's what it's like to high-rep users ^^

7:10 PM

I wish I had a global internet-wide "viewdelete" access...

And because this is a beta site, high-rep means just 2000 rep.

Example of link to a revision: codegolf.stackexchange.com/revisions/40415/3

My rep is growing slowly (not only on StackExchange). On one other site I have a bookmark for a page that I will able to view when I finally get viewdelete.

7:34 PM

@Ypnypn btw, why did you delete your language-grid answer?

the python was invalid, but that didn't make the submission invalid ... that would just have made it score 3. or you could probably even have found another language where the python version was valid

7:57 PM

@MartinBüttner It wasn't a great solution anyway; it was more to break the ice. I don't know enough languages to do much better.

ah fair enough

I blame you for not making the HNQ :P

9:08 PM

For the record I'm abandoning this but theres a pretty good chance I'll post something similar soon.

@Calvin'sHobbies Would you mind editing out the body of that post?

No problem

@Calvin'sHobbies Do you have any better idea for the scoring of the regex domino idea?

9:34 PM

Not really, maybe cumulative upvotes.

For the regex, couldn't the pattern always be something line ^(A|!(B|C|D|E))$?

Where A is the literal matcher for the previous answer and B, C, D... are the literal matchers for the other answers.

Then this regex works on A but fails on B, C, D...

(using ruby variant rubular.com)

Unless I'm mistaken, without a length limit or incentive for shortness this can generalize to any number of answers

10:12 PM

@Calvin'sHobbies it can always be ^A$

which is why this challenge only makes sense if shorter answers are better

10:36 PM

@MartinBüttner Ah, right. But then whats to stop this going on indefinitely, even if most of the answers are suboptimal? I think domino-coding (or whatever we're calling it) should not be able to go on forever.

(Yes eventually you'd reach the 30k char limit but that's not what I mean :P)

see my new scoring

adding answer that are longer than 20 characters is detrimental to your score

I'm confused about the product scoring thing. Wouldn't someone with only one short answer have the smallest product?

Unless I misssed that that's a tie breaker.

the score is N/20 which is less than 1 for submissions below 20 characters