In response to review on the uri-review list, I have changed the URI
schema name from "ab" to "addrbook".
In further discussion on the uri-review list, Graham, who will be the
expert reviewer we'll have to convince when we get to that point, has
pushed back on our defining an "addrbook" URI scheme for this. At the
moment, he's not speaker as the expert reviewer, but this tells us
where we'll stand when he does. It's still possible for us to
convince him, but for now, he's not convinced.
He suggests that we we use the "urn" scheme, either coming in as part
of the existing IETF parameter namespace:
urn:ietf:param:sieve:addrbook:default
...or defining and registering our own new namespace:
urn:addrbook:default
We're concerned that the former generates too-long and too-ugly URIs
for our purposes. The second is possible, but I actually think it's a
*worse* idea to define a URN namespace for "addrbook" than it is to
make a URI scheme for it.
For those who want to see the discussions, here are the three most
significant posts from Graham:
http://www.ietf.org/mail-archive/web/uri-review/current/msg01408.html
http://www.ietf.org/mail-archive/web/uri-review/current/msg01416.html
http://www.ietf.org/mail-archive/web/uri-review/current/msg01424.html
...and you can look at the other, related messages.
I spoke with Pete Resnick, who suggested using something related to
CardDAV to handle this. I've asked Cyrus whether there's anything we
can do in that regard, and I'm waiting for his response. Pete is not
inclined to go against Graham's recommendation, and also is concerned
about the scope here.
At this point, I don't know what to do, so I'm taking it back to the
working group. Alexey thinks grabbing a URN namespace is acceptable,
even if we have to hold our noses a bit. Ned, what do you think?
Others?
Barry
_______________________________________________
sieve mailing list
sieve(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/sieve