On Fri, 20 Feb 2009, David Artman wrote:
Anyone interested in a Winter Icehouse Game Design Competition, while we've still got a month of winter? :)
Heh. Sorry for that. I had kind of disappeared by Halloween and missed winter. Also, I'm glad to see someone put the final tally for last contest up; I went into hibernation right around the time the wiki got fixed, and never got around to it.
Restriction - Product/component, mechanical, terminological?
I've always been attracted to the terminological restriction (as its the kind of "restriction" that can focus creativity). If we go with this, I've got some ideas (hint: 200th anniversaries.)
Judging Categories - Yes or no; and if so what and how weighted?
I think it's hard enough to get people to even vote, let alone get them to vote across multiple categories.
Tangentially related, since a year and a half ago (when I aggresively argued about how we should count the votes for these things) I've done a lot of reading on voting systems (William Poundstone's "Gaming the Vote" in particular), that has changed my opinion about Condorcet and ranked pairs.
I'd now like to advocate for score voting (every voter gives a score on some scale (0-5, 0-9, 0-99; doesn't really matter*), highest average score wins**.)
I remember someone argued for score voting way-back-when, and I'd like to now say "you were right, I was wrong".
It might be a bit late to go changing the rules for this contest, but maybe we can talk about it for summer.
Once and Future IGDC Coordinator
I never did change the password to the coordinator gmail account, if you're saying you'd like to take it back :)
*Strategically, you should always give your favorite candidate the highest score and your least-favorite the lowest possible score (usually 0), even if you don't consider them to be the best (or worst) game _ever_; you're only comparing the games within the contest, not all games across all time.
**Voters can also vote "no opinion"; a vote of no opinion doesn't count into the average (i.e., it's not the same as a zero vote). This leads to some concern of "stealth winners", where most voters have no opinion on a game (didn't play it?), allowing a very small strategic group to get the stealth candidate a disingenous win. To offset this fear, a number of "soft zeroes" can be added to every calculation (for instance, if the number of soft zeroes is 3, and 8 people have an opinion on game1, sum their votes and divide by 8+3=11; and if 3 people have an opinion on game2, sum their votes and divide by 3+3=6). What an appropriate number for soft-zeroes is, is up for some debate, but 1/10th-1/5th of the number of voters is probably alright. Which, based on our last contest, would be 1 or 2 soft zeroes.
-- Dale Sheldon dales@xxxxxxxxxxxxxxxxx