Darren Reed wrote:
> In some email I received from Calabrese, Christopher, sie wrote:
> [...]
> > Yeah, I'm willing to concede the point. I'm beginning to think
> > YYMMDDmmhhss.fraction really is better since it's easier to deal with in a
> > tcpdump, etc.
>
> Y2K bug alert!
Oops, of course I meant YYYYMMDDmmhhss.fraction.
> How about Year-month-monthday-hour(24)-minute-second.fraction ?
I think this is good, though you might have to compromise a bit with those folks
who think that anything more than a 32-bit (binary) representation is taking too
much bandwidth (argh!).
> I said it like that because I'm not sure this would be easily understood:
>
> YYYY[Y]*-MM-DD-HH-MM-SS.fraction
>
> Solves the Y10K problem :-)
Hmm, somehow I doubt this protocol will still be around in Y10k, though you
never know. Also, experience tells me that just defining the protocol this way
doesn't mean the implementations will be able to handle 5 digit years.
--
Chris Calabrese
Internet Infrastructure and Security
Merck-Medco Managed Care, L.L.C.
[EMAIL PROTECTED]
.
