a) they provide a unique id for each movie
b) movie metadata does not exist digitally otherwise, thus imdb managed to become _the_ place for metadata about movies.
c) movies have a well defined set of metadata values they all share
now what does that mean for software?
a) gitreview currently uses the name of a software package as id but that is not unique. even less unique than title/year for movies. I recommend to switch to numbers right away. I.e. if http://gitreview.com/packages/nginx/ is not nginx(http://nginx.org/) nothing should be.
b) there are many online repositories of software with various levels of metadata. and the software itself is also available online from the authors most of the time. what does gitreview add to make it _the_ place for metadata about software?
c) there is no well defined or commonly agreed metadata set for software, what information is gitreview collecting exactly? currently i see name, description, link and some tags. Why are you not using something like DOAP(https://github.com/edumbill/doap/wiki)?
Regarding the names and urls - you have a valid point, they should more clearly indicate the library they are referencing.
