Re: I-D Action:draft-ietf-tsvwg-iana-ports-07.txt

[Magnus Westerlund <[e-mail]>]

Hi,

The diff between this version and previous one is here:

http://tools.ietf.org/rfcdiff?url2=draft-ietf-tsvwg-iana-ports-07

This version contains a lot of polishing, alignment and a bit of
restructuring. We would very much appreciate any feedback on the changes.

I do note that we have not included indication of which are the primary
service name for other services that has aliasing than for
http/www/www-http. If more cases should be included in the IANA
considerations please let us know.

Personally I have reviewed the critique in
http://www.ietf.org/id/draft-gudmundsson-dnsext-srv-clarify-01.txt and I
don't see any real issues.

1. We have inverted the registry name to be: Service Name and Transport
Protocol Port Number Registry

2. For me personally I don't see an issue with giving different service
names to an application that uses different substrates. I think the
following paragraph from section 5 of this document is the one applicable:

   For future assignments, applications will not be permitted that
   merely request a new name exactly duplicating an existing service.
   Having multiple names for the same service serves no purpose.
   Implementers are requested to inform IANA if they discover other
   cases where a single service has multiple names, so that one name may
   be recorded as the primary name for service discovery purposes.

But as different substrates aren't exactly the same service I don't see
an issue.

3. Unfortunately it seems like we still have a broken ABNF. I have
suggested a fix:

   SRVNAME = (ALPHA / (*(*DIGIT [HYPHEN]) ALPHA)) *([HYPHEN] ALNUM)
   ALNUM   = ALPHA / DIGIT     ; A-Z, a-z, 0-9
   HYPHEN  = %x2d              ; "-"
   ALPHA   = %x41-5A / %x61-7A ; A-Z / a-z [RFC5234]
   DIGIT   = %x30-39           ; 0-9       [RFC5234]


4. When it comes to the length of service names, I fully agree that from
a DNS perspective longer service names wouldn't be an issue, but for
using them in transport protocol near usages it would be an issue. As we
are going for unification of the name space to avoid collisions the 15
character limit is needed.

We likely submit an update in the next few days to take care of the ABNF
issue. If you have any more nit please tell us immediately.

Cheers

Magnus Westerlund

----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM
----------------------------------------------------------------------
Ericsson AB                | Phone  [phone]
Färögatan 6                | Mobile [phone]
SE-164 80 Stockholm, Sweden| mailto: [e-mail]
----------------------------------------------------------------------