> Yes, it could be done, but if you specify everything for all that
> functionality, you pretty much invent "rsync over HTTP" :-)

"Rsync over HTTP" has already been (partially?) done:
http://zsync.moria.org.uk/

The only wrinkle is that there'd have to be an index file or similar,
because zsync is aimed at single files.... The advantages of zsync over
rsync include that the diffs are pre-calculated once per file change as
opposed to every transfer, and the actual server is simply a HTTP 1.1
compliant web server. The diffs can calculated off-line and uploaded to
the server or servers.

Debian has implemented a similar scheme for it's package files.

On a related note, one of the requirements that Tridge had in mind when
designing rsync was maximising throughput even in the presence of high
end-to-end latencies and lots of small files. He did this by minimising
the "chattiness" of the protocol and thus the effect of round-trip
times. I think that is another property of rsync that would be useful to
retain.

Rob

-- 
Robert Loomans                                 Email:  [EMAIL PROTECTED]
Programmer/Analyst, APNIC                      Phone:    +61 7 3858 3100
http://www.apnic.net                             Fax:    +61 7 3858 3199

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

_______________________________________________
Sidr mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/sidr

Reply via email to