On Sat, 2003-08-30 at 00:43, Erik Nordmark wrote:
> Thus you seem to silently assume that these devices have a rich user interface
> that would allow humans to fix this on the devices.
Not exactly. I'm saying that if a device allows a user to name it, the
user is ultimately responsible for resolving any naming conflicts. If a
device does not provide a UI for this, then it should (preferably) have
a globally unique name in the first place.
> An alternative would be to be able to qualify the names on the multihomed
> node e.g. "[EMAIL PROTECTED]" where the part after the @ sign is
> the interface on which to perform the name lookup. Thus it would be a local
> convention, and the camera itself would know its name as "canon-camera"
> only.
To my mind this is not really an alternative but one of the ways in
which the user might try to resolve the conflict once it has been
detected. I.e., if "canon-camera" comes back ambiguous, the device might
tell the user "Choose one of: 1) [EMAIL PROTECTED], 2)
[EMAIL PROTECTED]". I think of this as device assisted conflict
resolution. The human user still has to make the decision.
In many cases this approach would be preferable to unconfiguring or
renaming one of the ambiguous names, which would potentially impact
multiple human users.
MikaL
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page: http://playground.sun.com/ipng
FTP archive: ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------
