I wanted to create a golfing competition that would be split in n different challenges, each posted with k days in between one another.

n and k are open for discussion here.

Each challenge on its own would be a regular challenge, but then the competition would also be scored. I was thinking of having your competition score be the sum of the byte counts of your submissions to the challenges. Much like your score in a golf match is the sum of your scores in each hole.

Furthermore I was thinking of requiring that all submissions be in different languages in order to be eligible to win the competition. The competition prize could be a rep bounty.

I was looking forward for you feedback, as the most experienced users can tell me how to make this work, pitfalls to avoid, etc.



Technically no problem, but you might want to refine some rules

Each individual challenge being a well-formed , I don't see any problem holding such a competition. Just ensure that you get enough feedback in the sandbox as you do for other challenges.


... all submissions be in different languages in order to be eligible to win the competition

This might be a problem because one user can post multiple answers on a challenge. My suggestion is to change the wording to

For the final scoring, each participant should choose their best answers in each challenge. Only the submissions in different languages count (unlike usual CGCC rules, same language with different implementations does not count, but a flag is still zero bytes).

and add this for convenience:

For the final challenge in the series, each participant should include their own list of answers and the final score in the answer to be eligible for the award

You might want to impose time restriction (e.g. 7 days after each challenge is posted) to each challenge too, for the sake of the competition. Of course the challenges should be indefinitely open for out-of-competition answers.

  • \$\begingroup\$ Thanks for your feedback. I thought about getting sandbox feedback, obviously... But if I post the problems openly in the sandbox, then some people might get an unfair advantage over the others... I was thinking of asking a couple of you to do a "closed sandbox" for those challenges, and then the people that helped me would wait some time before also posting their solutions to the challenges... \$\endgroup\$ – RGS Feb 10 at 17:50

I think this your idea is fine as a bounty. It's accepted that you can put whatever conditions you feel like on bounties. This would be reasonable to put a a post in the List of bounties with no deadline and link to that, even though it actually has a deadline. This avoids needing to make a new type of win conditions, since each challenge on its own is just normal code golf.

I like in particular the idea of people trying hard on a set of well-vetted challenges. Sometimes it seems like people don't have a reason to golf off bytes rather than just posting quickly.

One pitfall I see is that winning basically requires using all golfing languages. Maybe this is what you're going for, but it's not the type of thing I'd participate in. Even if you can't repeat a language, there's loads of golfing langs to go around. This is especially if different variants counts as different languages, which seems worth trying to make rules for.

It also seems like solvers will want to pair the most efficient language with the longest challenge. Or, to match languages with challenges that they have built-ins for, which might lead to boring solutions. In particular, lots of people might post the same mostly-built-in solution, or be squeamish about posting yet another copy, or be worried they "cheated" by seeing someone else post it.

Maybe you want to have more non-golfing languages. I'm not sure how to enforce that. You could list the languages one is allowed to use, but people might be unhappy being restricted like that. One thought is to use golfing Elo rankings of languages somehow, but it's still hard to balance, and a language's performance really depends on the challenge.

  • \$\begingroup\$ Thanks for your feedback. Maybe I shouldn't impose restrictions on the languages? Or maybe I could say people are only elligible if they have at least m different languages among their submissions... Like 3 different languages or something like that. Would this make sense? \$\endgroup\$ – RGS Feb 10 at 17:54

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .