Cogito, ergo ludo.
There exist all sorts of recommendation systems that suggest new whatever based on whatever the user has liked in past. For example: movielens, last.fm, thisonenext.com and StumbleUpon.
If one exists for gamers, I would be happy to hear about it.
To my knowledge, they use whatever algorithms (some academic material here) to calculate the proximity of different tags and users and then make suggestions based on those. Or something.
I think such a service would be a great addition to NearbyGamers. Problem is that the algorithms tend to be fairly consumptive. I think it is possible to get past this by updating them rarely (like, say, once a week).
The site already has a large database. Consider every person and possible every tag as having some other tags. Count which are similar. Suggest, say, three of the closest matches to every person. List them in the profile. Update weekly or whenever there is an opportunity or whatever.
I think this would be beneficial to nearbygamers in that it would be an extra incentive to be here, it would help in discovering new games, and provide a reason to visit once a while (as opposed to coming and making an account and never returning).
(Also, I am forced to enter a location for the discussion to post this at all.)
Lazy Gaming at its Best
I don't really think this would work.
The program would recommend which games you should try based one what the people around you like and based on your existing tags. I think it might feel more like spam then a real suggestion (if you like x try y).
Because I'm Snarky...
I think this may be a good idea. I think you misunderstood, BrooklynKnight, this wouldn't be based on location (three closest tags physically), but common tags.
Lets say I only play Chez Geek, Great Dalmuti, and Munchkin (all great games). You'll notice that Chez Geek and Munchkin are both Steven Jackson games. Some large fans of SJGames may have both of those, as well as Ninja Burger. Due to 5 other people having Chez Geek, Munchkin and Ninja Burger, my 'interest' in Ninja Burger could be 5 (or some other raw score, really statistics ain't my thing). After you've ranked the different tags, just show the top three 'interesting' games for a given profile.
I somewhat like the idea, as I pop into a games store every month or two to look for something new. I find myself staying in the SJGames side of things because I'm worried about spending $40+ on a game I may not like. If I could get a small list of games to think about buying, I'd be able to look them up online and figure out which sounds best.
Just my sleep-deprived $0.02
Lazy Gaming at its Best
Isint that kinda what meta-tags do?
The gamer that runs this site
It a pretty similar idea to meta-tags: they allow you to browse around to find similar stuff. This would be a computer program analyzing what tags people use and share automatically. It's not a bad idea, but I think we need a lot more gamers before there's enough data in the system for the program to provide meaningful results.
I know I am not the only person here who peruses BoardGameGeek.com. Not only does it have it's own internal rating system to sort games rated by users, but there are also user created queries and lists that allow you to search for games by number of players, type of game, level of difficulty (light/heavy), playing time, etc. For example:
The Ultimate Boardgame Selection Tool
Discussions are closed and stop accepting new posts if a moderator closes
them or 60 days of inactivity passes.