Kode Wrote:I'm planning on adding some kind or voting/favourite/rating system, not sure what yet though, any suggestions are welcome, then in the api you will be able to order by favourite. I can also add a limit option so it only returns a single results for each.
http://fanart.tv/help/ has some initial details on the API but bits may change.
Let users(/visitors?) flag/favourite/star/upvote a logo when there are multiples, the results of which dictate which logo is returned first or when the limit is set to 1. A more complicated (less gameable) implementation would be to enforce that only one logo per show was flagged (radio button style), but that would require users were logged in (and the backend would have to load this info every time a page was loaded so a user could change their vote if new logos were submitted or they realize their choice doesn't work as well in practice, etc).
This is probably not super important, though, and the moderators could arbitrarily pick the favorites if that's deemed to be less work overall....
If you only went one way, though, I think a more useful feature would be downvoting or flagging for removal. This could be opened up to the userbase at large (with a few set reasons for why it was flagged as such). This could move a logo down the ranks when returned as multiples, and after reaching x number of downvotes/flags, could put it in a queue for moderated removal.
I think that could be a really simple modification, and might even be able to take a load off the moderators. If uploads don't need to be moderated and the combined userbase is allowed to pass judgement on quality, then the moderators only need to step in when something is obviously wrong. Or at the very least, they would only need to check for glaring problems.
This could also lead to much greater quality overall, because a secondary queue of flagged images (when the only logo/all logos of a show are flagged) would bring them to people's attention and more submissions of [hopefully] higher quality would come in. Rather than take on a really hard logo that no one has been able to do yet, someone wanting to contribute could take a stab at a good-but-not-great extraction that otherwise wouldn't have been brought to their attention. (This is not a dig at the current moderators--I'm sure you're only letting the very best through--but the first initial batch import was much less stringently moderated and there are some shows that appear filled but could benefit from a more perfect recreation, etc. More eyes and all that....)