6
\$\begingroup\$

A language specific contest is easier to host. A language agnostic contest is more inclusive and perhaps therefore more competitive, but can be more work to set up and host. A language specific KotH has become easier to set up since Calvin's Hobbies introduced a Stack Snippets King of the Hill contest, which does some of the hard work that previously had to be done manually by the question poster.

Would it ever be acceptable to run a Stack Snippets KotH to judge interest and iron out loopholes, and then use that to decide whether to set up a full language agnostic version (which may allow for a wider range of strategies)? Would this always count as a duplicate?

If it is only acceptable to do one or the other, then I may hold off and post a language agnostic KotH at some point in the future. If it's acceptable to do both then I'd like to try a Stack Snippets version first.

Would it make a difference if the initial Stack Snippets version was on a smaller scale? For example, fewer players on a smaller playing arena?


This is a general question to also guide others in future, but if it makes a difference, the specific KotH contest I have in mind is Flit, which is currently in the sandbox.

\$\endgroup\$
  • 1
    \$\begingroup\$ I guess that could work if it's clear that your first post is intended as a test run and will be closed as a duplicate of the full thing once it's posted. Holding off some features for the second post would probably help in keeping both instalments interesting. \$\endgroup\$ – Martin Ender Apr 22 '15 at 15:51
  • \$\begingroup\$ I think it's alright, though since answers could possibly be copied from one contest to another, some may consider it a duplicate. Perhaps instead of two questions, you could just have the Snippet challenge at first and promise to edit it later on to allow more languages if it becomes popular. This may be trickier to manage since the edited contest would have to be backwards compatible with the snippet contest. \$\endgroup\$ – Calvin's Hobbies Apr 22 '15 at 16:22
  • \$\begingroup\$ @Calvin'sHobbies I don't know how much of a speed difference there would be, but I was hoping that the full version with faster languages would allow more strategies. I also wondered about reducing the turn time limit for the full version, to allow a larger arena without increasing total game length. \$\endgroup\$ – trichoplax Apr 22 '15 at 16:29
  • \$\begingroup\$ If the speed difference is small then I guess I would go with @MartinBüttner's suggestion of closing the first question as a duplicate of the second. I'm guessing it would be a duplicate anyway though. \$\endgroup\$ – trichoplax Apr 22 '15 at 16:31
  • \$\begingroup\$ @Calvin'sHobbies Of course, if you edit one post to include all languages later on, it might not get the necessary attention for a KotH. \$\endgroup\$ – Martin Ender Apr 22 '15 at 16:35
  • \$\begingroup\$ Even disregarding speed, JS is definitely not ideal for koths (easy to mess with globals/DOM/language, hard to have time limits). @MartinBüttner Funny, I was just thinking the opposite, that the first contest might spoil the experience of the second because by that time everyone will be familiar and perhaps bored with the mechanics. \$\endgroup\$ – Calvin's Hobbies Apr 22 '15 at 16:46
5
\$\begingroup\$

Having written many KoTHs, I've found that getting people to answer to your KoTH is hard. You have to get people interested AND make it easy to answer. That's the fundamental reason why we like cross-language controllers.

However, I'm becoming more and more convinced that the way to do it is actually how Create your own Wolf did it. Basically, you initially make it limited to a single language. If it does get popular, then you can write a wrapper that will enable other languages to participate as well.

This has several advantages:

  1. You don't have to create multiple questions (and have people answer with the same answer)
  2. Your KoTH will actually run faster as the submissions in the original language won't have to go through STDIO to communicate.
  3. It provides an easy way for people unfamiliar with STDIO communication to participate (as there are always some quirks with piping, such as needing to flush in Python)
\$\endgroup\$

You must log in to answer this question.

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