On Mon, Aug 18, 2008 at 11:12 AM, Andrew Lentvorski <[EMAIL PROTECTED]> wrote: > Bob La Quey wrote: >> >> On Mon, Aug 18, 2008 at 10:28 AM, SJS <[EMAIL PROTECTED]> wrote: >>> >>> -- >>> HTTP is a lousy way to synchronize filesystems - use rsync for that. >>> Stewart Stremler >> >> As best I can tell rsync is not governed by any RFC. Please correct me >> if I am wrong. Whatever the shot comings of http it is at least backed >> by a standard. http://www.w3.org/Protocols/rfc2616/rfc2616.html > > And Microsoft's office format is an ISO standard ... ;) > > It is true that rsync doesn't have a "RFC". However, it's documentation > tries very hard to explain exactly what it is doing. And the fact that the > people associated with it (the Samba folks) grok Windows *and* Linux > filesystems very deeply gives me a much better feeling than some random RFC.
I understand that. I just think it is interesting that we have these two approaches. One, an RFC that supposedly documents existing best practice, and two, a project with code and docs. It is not obvious to me that one is better than the other though one could argue the RFC provides a more abstract definition of the problem and its solution, for better or worse. BTW, i do not think I would call the HTTP 1.1 RFC 2616 "a random RFC." It has a pretty decent track record. > If, however, you don't trust rsync, there are also things like "scp" and > "sftp" which are part of ssh and are documented. Been using scp, which is fine for what it does, but differential backups would require more support wrapped around it. BobLQ -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
