On Thu, Nov 10, 2011 at 7:12 PM, Andrew Deason <[email protected]> wrote: > On Fri, 11 Nov 2011 00:08:27 +0000 > Simon Wilkinson <[email protected]> wrote: > >> On 10 Nov 2011, at 23:12, Andrew Deason wrote: > >> >>> 4. afsUUID >> >> Not relevant to this draft, but come RPC refresh, we really should >> replace afsUUID with a UUID encoding that just represents it as an >> opaque 128bit value on the wire. >
yes! I wholeheartedly concur. > Or encoded as a bunch of explicit int32s packed in a specified way or > something. I think Tom had something for this? Or maybe he just > expressed the same "hey we should do this" sentiment. > Alas, my involvement is merely the latter. My recollection is we settled on something along the lines of "typedef afsUUIDng opaque[16];" at the Pittsburgh hackathon back in April. Matt, please correct me if I'm wrong, but I believe you tentatively volunteered to write a brief i-d describing this encoding; do you still foresee having time to write that? I know you have several drafts on your plate at the moment... Cheers, -Tom _______________________________________________ AFS3-standardization mailing list [email protected] http://lists.openafs.org/mailman/listinfo/afs3-standardization
