(2012-04-09 15:06)dzan Wrote: I do think it will be useful for a meaningful amount of the users, I don't know if English is your mother tongueAlso a Flemish speaking Belgian computer science student (but from the city with cheaper beer ). And indeed it might be nice to have a Dutch plot (mainly for my parents) and everything else from imdb. The way you explain it I can see it being usable, as long as people can also download all-in-one scrapers (which of course can still work with modules behind the curtains).
About the speed: I think the current scraping is all done by a single thread (correct me if i'm wrong, but I believe I read it somewhere), maybe there are opportunities for parallelisation.
And actually it's not the scraping that really annoys me, it's the looking for file changes. Some watched folders feature could improve this, but I don't know how feasible that is with all the different file protocols xbmc supports.
(2012-04-09 15:06)dzan Wrote: I too have knowledge of recommender systems ( did a project on it for university and had to read a lot of research papers on it ) and you are only partially right about the cold-start problem. There are solutions and workarounds. A recommender system is not the focus of this project, it's an example of what 2-way communication would allow us to do. I don't think xbmc will ever build one but existing once could use xbmc to acquire more data. One user's library will indeed probably not suffice but that's the reason a feedback channel would be implemented.Okay, nevermind, I got this all wrong, I thought the feedback was from the user to the scraper (eg when the scraper presents multiple options of found movies and the user selected the best fit) instead of from the scraper to a service.