I suggest to introduce a limit for the number of tags any project can be tagged with.
In the last time I found that Ohloh often displayed "related" projects which were not really related to the project I was browsing. This seems to be caused by the fact that those projects were tagged with more than a hundred tags, therefore increasing the chance to show up on any other projects page.
For me that's just spamming other projects pages, so please think about a hard limit for the number of tags per project.
I totally agree with you, Bananeweizen, but not everyone here at Ohloh agrees with us.
I'm of the opinion that tags should be short and to the point, and that dozens of silly tags like "software" and five different ways to spell "content-management-system" do not help.
I won't try to speak for others, but the pushback I hear is that more tags makes search work better, because we have more content to search against.
Tags, search, and project browsing are currently undergoing a lot of interface and feature changes right now, so there may be some changes to tagging in the near future.
I'll argue to reduce tag spamming :-).
Thanks for your reply. I wonder whether those arguing about improving the search really tried searching something. Having more results displayed on the page doesn't mean that the search got any better than before.
If search results shall be improved, one could think about how to include useful parts of the projects code (which is already scanned by Ohloh) in the search index. E.g. if the Ohloh scanner can differentiate all those file types it might also be possible to extract useful keywords, tags or meta data automatically.
Another thing one might think about is to grab the (top level only) homepage of the project and to use that content for keywords. Because the project description on Ohloh is typically very short compared to the description on the project homepage. The same content grabbing idea might work for the links provided on the project pages.