> At the risk of broken-record-ness, the SCCS weave format is
> consistently smaller than RCS format for the same set of versions, and
> takes the same time to extract any revision.
In the original RCS paper, they explain why they think RCS is superior
to SCCS. Essentially, the time to extract any revision from SCCS is
proportional to the total number of pieces in the SCCS weave, whereas
the time to extract a revision from RCS depends only on how distant
that version is from the trunk --- not on the total number of
revisions present. Since most accesses are to recent versions, this
is the right tradeoff.
Received on Sat Oct 21 14:36:07 2006