[JDEV] Routes, etc..
tcharron at ductape.net
Mon Oct 11 15:24:24 CDT 1999
Quoting Colin Davis <cdavis at thepentagon.com>:
Thomas Charron wrote:
> > The only way to get the CLIENTS timezone, and not that of the 'upstream
> > transport' will be if the client adds a node entry to the route tag.
> Agreed. Either way is acceptable, but I favor the first, for the reason
> stated below.
I tend to agree as well. Basically, the first tag would have no recieved
date, only a sent date, and would contain the client data..
> Some people did not like this, as timezone may change between
> messages,and the recieving client would hae no way of knowing when to
> That is why I am in favor what I previously wrote, where the sending
> client would modify it's timestamp tag (or whatever) including it's
> timezone _when it had changed_
> This would be placed on only the first timestamp, which would be sent by
> the client.
Yea, I/Q is kinda hazy about this. I'd say we'd still need the data on ever
single stamp, becouse the sending client also has no way of knowing that the
client recieving the message already knows the timezone data.
> These are different, but similiar. The timestamp could be modified by
> the client, as I had said, to includ the timezone. This might be more
> appropiate somewhere else, however.
I like that idea. Basically, the first entry in route would be the sending
client. But I'd have to say we really need to include the TZ in every message
for it to be reliable. Neither end can really tell what's going to happen.
<< Wanted: One decent sig >>
<< Preferably litle used >>
<< and stored in garage. ?>>
More information about the JDev