Does anyone have any feedback for this challenge before I post? It's been up for three days, with three upvotes and no comments.

hi

@undergroundmonorail did you have a TI basic answer you wanted me to test?

@TheDoctor Oh, sorry

codegolf.stackexchange.com/a/26173

It looks like it's been modified since then though

For a "real" example of code trolling I've done, search google groups for "thanks for the fish motherfucker".

groups.google.com/d/topic/comp.lang.c/FQs9lO5Ay_U/discussion

The order is different, but it's exactly what the dude asked for.

That's a part of real trolling we can't really emulate here: the gradual reveal of new complications OP forgot to mention at first. Maybe with spoilers it could be possible.

@undergroundmonorail There's no need to test it. It's easy to observe that it's completely off spec.

On a related note, I'm slightly disappointed with the answer to my comment on the question. I was thinking of implementing the icosahedron by implementing its symmetry group as a presentation of A_5.

@ProgrammerDan Well i have created a proof of concept for the Tweet music competition: drive.google.com/file/d/0B9i-DPAdw_3YOGMzOW44V1ladUE/…

It currently 44 bits per second. The proof of concept it a little contrived as the clustering was trained using the source file but I was really only interested in how the output would sound... i.e. whether it would be noise. So on that basis it seems to be doing ok :P

This output was produced with 11 bits per sample, i.e. 4 samples per second so the training data is just under 2 meg. not an unreasonable download size. The training was taken from 5000 0.25 second samples of the original song. I will try training it over more varied sources to see how it fairs then :)

Do we want by-language challenges?

Do we actually need them? Apart from a score board that lists all winners by language, there is not much difference.

@ProgramFOX Well it does incentivise people to try to solve a problem in their favourite language even if there is already a 0.5-character APL solution. For some interesting problems it can also be really nice to generate this rosetta-stone thing that you can compare how different languages approach the same problem (e.g. I really liked the Mandelbrot one).

I don't understand the code-length concerns expressed in the comments of codegolf.stackexchange.com/questions/26317/miniature-faking

Why is writing simple 2D manipulation methods unreasonable?

I implemented my esolang Manufactoria in Java 8. Is there a good location I can post the code/jar file? It is ~14000 chars long.... For example, is github good? What is the best thing to do.

Cool, this site let me post it on chat :-)

Are you removing it, or is it being auto-removed?

@grovesNL I removed it. I was experimenting.

@Quincunx I see

I don't think people want to have to scan through code that can't even be formatted.

@Quincunx you made an esolang from manufactoria? :O ... how awesome is that!?

@m.buettner Making the esolang was easy. It is basically a direct translation of the game. However, implementing it was much harder.

@Quincunx Both designing and implementing a language are things I still have to do/learn at some point, so you have my full admiration ;). Let me know if you put it on github.

@m.buettner Like I said, I didn't design this language. The game was the design. I simply came up with a translation to text. Implementations are fun. You should try it. I once created a BrainF*** -> Java compiler, in Java.

*Correction: I did design a language. Just not this one.

@Quincunx I'll certainly give it a go at some point, when I've got fewer other things on my schedule.

Wait, I did write a regex interpreter with quite a few features that could have easily been extended to the full ECMAScript flavour (and probably others, too). Not Turing complete but still a language. ;)