I forgot where I posted about Subversion before, but it's a stated goal
of this project to replace the CVS version control system. They intend
to release version 1.0 by the end of the summer.
A quick summary includes diff-level granularity with versioning
(Subversion's primary aim) and security (provided by apache) written to
a virtual file system stored into a BerkeleyDB database unlike CVS.
An arbitrary number of name-value pairs can be used to "categorize"
The code for the API for their storage method is available via the web
Acrobat page number 41 of their design document gives a relatively current
schema for the "content" storage system. Perhaps with XPath and
XPointer extending Subversion an OHS storage system could be partially
implemented. Given a system that works at a diff-level, a logical step
(needed for an OHS) is an editor aware of the changes. I suspect this is
relatively simple from a technical back end point of view but some would
say an overly extreme imposition on the user.
Working forward, a good architectural diagram was shown at SV-LUG last
night outlining their front end. It consists of either a local library
connection or a WebDAV (over Apache) connection. A set of libraries
handles these cases. Front end libraries are the hook into both a
command line CVS-like client and a (planned) GUI client.
-- -- Grant Bowman <email@example.com>
This archive was generated by hypermail 2.0.0 : Tue Aug 21 2001 - 17:58:03 PDT