Re: Heirarchical error codes

["Frank Dawson" <fdawson@xxxxxxxxxxxxx>]

On Thursday, March 18, 1999 8:26 PM, Doug Royer wrote, in part:

>2) I think we had a discussion about being able to SET the
>   CAP version that the client knew about. No more conflict?


Hhmm. I don't remember reading about this discussion in any minutes. Maybe
you can refer me to that discussion?

In any case, it would seem appropriate for the CS to disclose the CAP
version that it understands in a resposne to the CAPABILITY command.
However, the requirement that a CS must adjust and talk in a version that
the CUA specifies, seems odd and not consistent with other protocols that I
have read. Did you have some IETF protocol in mind?

I actually think that it would be much more appropriate for ease/complexity
of CS implementation and CS scalability reasons to make the CUA adjust to
the version of CAP that the CS supports.

Hence, the CUA would read the level of CAP support in the CS with the
CAPABILITY command and react accordiningly in the command syntax and
semantics that it sends to the CS. Certainly is the CS responds to a
CAPABILITY command with "CAP v1.0, CAP v1.x", then the CUA could choose to
use either v1.0 or the later v1.x command set.

- - Frank