|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: [iSCSI]: Key negotiation procedure proposalPaul Koning wrote: > > I think it's unnecessarily complex. The "simple implementation" alone > is plenty. Certainly! But if we want more than that, then mathematics tells me that those are the rules. (I.e. those are the minimum consistent set of rules which achieve the task at hand. It also tells me that there is no ``in the middle'' implementation.) Please also note that ``simple implementations'' can co-exist with ``complete impementations''. It's just that complete ones will get the TCP FIN, when rejecting a simple implementation's key=value pair, or just TCP FIN if the simple implementation rejected theirs. Julian, most of the negotiation the way we have it now, already observes most of those rules. And as to your request ``if they can be proven useful'' -- a few examples on your own will convince even the most doubtful heart. :-) -- Luben P.S. I just don't want to end up writing a paper about this (co-authors welcome) as it happened for CRC32C.
Home Last updated: Thu May 23 17:18:33 2002 10268 messages in chronological order |