While this is a twisted idea (I mean that in a good way ;) there may be 
some other good that could come from such an effort:

Consider: if you seperate the interface from the actual storage, you 
could in theory enable:

1) storing the data as a compressed image. This would help reduce the 
growth rate of the repository.
2) Storing the data in some form of database. This could allow the 
addition of atomic commits and other A.C.I.D. operations.
3) Making the backend of CVS into a kernel module (ducks and dons 
flamesuit laughing maniacally ^).




_______________________________________________
Info-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/info-cvs

Reply via email to