For example, I'm always looking for little snippets of XSL, JS or Cocoon code to do certain little tasks. Once I use them, I dont formally store them anywhere. So i'm always looking at my previous xsl to work out how to do something. If its something big, I may put it on my blog. But the small pices are usually lost till I need them again. With TagFacts, I can now store them and retrive them quickly. This is pretty what Dima is doing himself.
Although the appeal is not as great as reader2, it could have some long term appeal and become a place where people look before hitting the newsgroups and forums?
It could become a resource like wikipedia, but in actual fact its the total opposite of wikipedia. See rather then collabative working on the same fact, the collaboration comes from the choice of tags. Tagfacts will work best when you have trusted friends on it, so I could check out how Mr M Kay reverses a XML tree.
There is however something which I cant quite get my headaround. Each fact is so different and could actually be wrong. How is it possible to work whats poplar? There is also no commonality between facts except the tags (in reader2 the isbn is used while in del.icio.us the link is used), this strikes me as quite a difficult thing to keep a track of in the backend. Anyhow, I'm sure Dima has it covered.