I'm Ken Krugler, founder of Krugle - a search engine for programmers.
Below is an updated version of my introductory post on the priorart-discuss mailing list.
Krugle is a search engine for code and technical content. Millions
technically oriented web pages, plus source code from repositories,
and 50K+ project descriptions linked to code.
Right now tags (notes) are free-form. They have a name, a note, and then
either a line number range in a file, or an entire file/page.
In 1999 the USPTO issued this RFC: http://www.uspto.gov/web/offices/com/sol/notices/priorart.htm
Which meta-information for files getting timestamped do we store
All infos are optional, but in my prevalent-implementation i required that any three options had to be specified, but no specific one was required.
These are of course additional to the hash-values, which are required :)