Slightly facetiously, I'm mildly tempted to vote to close as a duplicate.  However, I shan't and shall merely point out that most of the answers to the question: <http://mathoverflow.net/questions/13619/is-a-free-alternative-to-mathscinet-possible> are relevant to this question, particularly mine.  The question <http://mathoverflow.net/questions/3038/errata-database> which covers similar ground.

(The [rForum](http://www.math.ntnu.no/~stacey/Mathforge/rForum) mentioned in the latter has stalled a little.  It could be revived if sufficient interest is shown.)