I would like to discuss the idea to include Code-Golf-esque Magic the Gathering style questions on the B&CG StackExchange.

Background Info

What is now the Programming Puzzles & Code Golf StackExchange, started off as the tag on StackOverflow. The Question Asker would post challenge questions with potentially many answers, and users would compete to post the best answer according to fairly strict requirements. I believe it might be fun to post similar style questions focused around MTG.


Would a similar tag be appropriate for B&CG StackExchange?
If yes:

  • Whether this should live on Live or the Meta (or similar to Code-golf where questions are originally asked on the meta, refined/tweaked and then posted to be answered on the actual StackExchange)
  • What sort of restrictions should be placed on when the tag can be used (anytime or once a week type thing)?
  • What type of questions would be On-topic (Build a deck around a specific theme/interaction/card, try to build an instant win combo with X card, do X with only this colour, or some other style)
  • For each On-topic question a set of requirements that would be used to evaluate answers (Win in least amount of turns, least amount of mana spent/CMC of cards, most upvotes wins, etc.)
  • Any other ideas you may have on how this could work

If no, that is ok too :)
Though it would be helpful if you leave an answer with a reason (Too off-topic, not something you are interested in, etc.)

  • Puzzling.SE does have some questions that deal with Magic that would probably fit this style of question
    – diego
    Aug 30 '17 at 17:37
  • @diego Interesting, I had not seen those questions on Puzzling.SE previous, just flipping through the ones you linked it seems like people not knowing the gamerules that well might be a common problem there. Seems like something that would be less of an issue here on B&CG
    – Malco
    Aug 30 '17 at 18:54
  • 1
    I would like to point out that asking for a deck to be built is just generally too broad, even with restrictions.
    – murgatroid99 Mod
    Aug 30 '17 at 21:13
  • @murgatroid99 I agree it is usually too broad since there are so many possible answers. In order for deck building questions to be answerable there would have to be some sort of defined criteria for judging them. Code-golf does a good job of reducing the near infinite amount of answers (in nearly any programming language) by judging based on byte count. I'm not sure what would be a good analog for "Lowest byte count wins" in a magic deck though, so deck building questions might just need to remain off-topic...
    – Malco
    Aug 31 '17 at 13:50
  • Asking someone to build a complete deck is more like asking for a complete piece of software on Stack Overflow than asking a code golf question. You can't just call it a puzzle and make it OK.
    – murgatroid99 Mod
    Aug 31 '17 at 17:00

(Note: these are my personal opinions, and should not be construed as describing policy)

I do not believe that this type of question is well suited to this site. These questions come in two primary varieties, and there are issues with each

Puzzle Questions

These generally provide a board state (plus cards in hand, etc.), and ask how to win during the current turn, or some similar task. If designed correctly, there is usually one correct answer that any reader can verify.

The problem with these questions is that they screw with the intended purpose of this site. In theory, the asker "should only ask practical, answerable questions based on actual problems that you face", and the answers answer that question in order to help solve that problem. With puzzle questions, the asker is creating a game for the answerers to play.

Keeping this in mind, some questions in this category are perfectly fine if framed in a certain way. We have had questions that ask "Given this board state, can I win this turn?" But questions like this should focus more on how cards and/or rules interact than on enumerating all possible lines of play and determining whether any of them resulted in dealing exactly enough damage to the opponent.

Optimization questions

These are questions that provide some restrictions that may narrow the card pool, then ask for the best outcome that can be constructed from those cards (e.g. deal the most damage, draw the most cards, win in the fewest turns).

These kinds of questions can have more of a place here, but many varieties are too broad or have other problems. Even if there is a restriction like "only use blue cards" or "only use sorceries", there are still thousands of cards to consider. And if you narrow it too far, you risk turning the question into a game of "figure out which card I was thinking of", or running into the same problems that puzzle questions have.

Code Golf

Code golf is different in some ways that allows it to avoid these problems. For one thing, in most programming languages, the number of building blocks is relatively small, and much smaller than the number of Magic cards. And the relevant languages are Turing complete, which means that you can do (almost) anything with those building blocks. This means that there is the potential of an endless variety of questions, and that anyone with algorithm design skills and even cursory knowledge of any programming language can provide an answer with relatively little additional research.

  • 1
    Thanks for taking the time to write an answer. My intention with the question was to gauge interest and pick brains for potential pitfalls. As much as you specify you aren't describing policy, your position as a high-Rep user (and ♦) does add a bit of weight to your answer.
    – Malco
    Aug 31 '17 at 17:07
  • 2
    For what it's worth, Magic: The Gathering is Turing Complete, but it would be pretty lousy for golfing questions. Sep 4 '17 at 20:36

You must log in to answer this question.

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