On Thu, Sep 01, 2005 at 04:14:50PM +0300, Constantine Dokolas wrote: > Michael Rogers wrote: > >Does anyone know how browsers react if they request an entire file and > >get a 206 (partial content) response? Do they send additional requests > >for the remaining data? If so, then perhaps partial content responses > >could be used to deliver pieces of the file out-of-order? > > > >Perhaps it would be worth talking to the Mozilla people about > >out-of-order downloads, since BitTorrent faces this problem too. From a > >quick look at RFC 2616 I get the impression that servers are allowed to > >send pieces of the file out of order, provided each piece is sent in a > >separate response (you can't send a multipart/byteranges response to a > >request for a single range). The question is how gracefully the client > >will handle the situation... > > You (and others) assume there is such a thing as a "part of the file". With > FEC encoding there is no such thing. If you have enough blocks (2/3), you > have the whole file; otherwise you have nothing. Matthew can correct me, if > I'm mistaken.
No. You may have some of the plaintext blocks. And on any codec other than onion, you may have enough check blocks to reconstruct some (rather than all) of the plaintext blocks. Even with onion you can do a partial decode - but we probably won't since a partial download of a block isn't verified at all. -- Matthew J Toseland - toad at amphibian.dyndns.org Freenet Project Official Codemonkey - http://freenetproject.org/ ICTHUS - Nothing is impossible. Our Boss says so. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: <https://emu.freenetproject.org/pipermail/devl/attachments/20050901/1d685d90/attachment.pgp>
