Keeyai.com

I’ve finally rated all 71 other LudumDare entries. Most were good, some were crap, and some were great. There were some really clever ones, and some funny ones, and some that were just plain good. Most though were good thoughts that didn’t have enough time (imagine that).

I had fun with the LD trophy system - I gave away 10 I think. Unfortunately there were only 5 other trophies given out amongst the other 71 entrants. That isn’t even correct as one was from someone who didn’t even enter and it was totally lame: they gave ‘The Python Award 2008′ to a game that was crap. It was one of maybe 10-15 written in python, so it’s not like it was unique in that form, and many of the python games were leaps and bounds above that one in all categories. I would approve of the Scheme award going out to the scheme game, and similarly for the haskell game, but honestly, let’s not tread on python’s good name. I guess I should be happy that a trophy was given at all though…

Having been through both pyweek and LD now has given me a lot of ideas if MUA ever hosts a coding competition. I love the trophies, but the functionality should be expanded there. MUA will probably require all python, just for ease of use — although forcing an executable could help alleviate those problems (and cause a whole new genre of other problems). I like the idea of the pyweek uploader and packager, even though they don’t work, and I’m going to work on a pyweek game loader before next time, which should minimize the pain in the ass of loading and viewing all the games one by one. Also, the pyweek torrent was awesome - I tried to do that for LD but nobody cared.

I also like the Does Not Work and Disqualify options in pyweek — there were a couple in LD that deserved that (like the mac only) but are going to probably get high scores and all N/As, instead of failing for barely working.

One idea I had before the compo and briefly discussed in IRC was a compo built from something already made. For example, maybe some of the ‘regulars’ put together a set of images, and those are the only images you can use in your game. It would be fun to see how people manipulate the images in all different ways (macro, tile, mini, collage, masks, etc) to get variety out of it, and it would level the playing field a little between artists and non-artists.

A major issue with LD was hosting. Everyone hosts their own file, which essentially means cheating can run rampant. The way pyweek handles it isn’t much better though, they just host it all themselves since richard apparently gets free bandwidth. I have two ideas about this.

1. upload to the main server, pyweek style, but don’t allow downloads. instead set up a torrent and let the masses serve the bandwidth
   • PROS:
     • far less bandwidth use for server
     • more ’secure’ since people can’t change what is in the torrent after the deadline
   • CONS:
     • forces people to use torrents, which people are scared of bc of the RIAA and such
     • forces people to download torrent software to get the game
2. write a game loader that MUST be used that checks MD5 hashes. users upload their hash instead of their game, then host their own game
   • PROS:
     • minimal server bandwidth
     • the MD5 hash should keep people from changing their entry
     • can build judging into the game loader
   • CONS:
     • have to write a good game loader that doesnt impede game development
     • people have to download the loader to judge the games

The second option really only works with games in one language and cross-platform, though (python).

The last thing I want to talk about in this post is theme. Pyweek had it down to 5 themes for the last few days, making it really easy to plan out or even start games beforehand. Ludumdare did a better job having lots of themes, but we ended up with minimalist, which the IRC people didn’t seem to want. There was a method for the regulars to control and police the themes, but it wasn’t used much (just like trophies). Maybe that is a failure of the community, but that doesn’t change the fact that it didn’t work right. I think a good way to handle theme would be to either create and vote on themes in the last hour or two, or just let the theme be decided by an outside party (sponsor?) when it is voted down to 15 options or something. I know I whine a lot about these rule infractions, so I need to make this clear — I’m very forgiving about these things when I’m judging — I almost always let people use their bug fix versions instead of marking them all 1s because of one major oversight — but I would really rather the opportunity for these other things just not exist. That way nobody has to even wonder if this game, which fits all 5 themes, was actually done during the time period.

Ok, this post got REALLY long. I’ll pick this up again on the MUA site if we ever come to that, unless people want to have a discussion in the comments (lol).