Re: Regestry Identifiers

["Rick H. Wesson" <wessorh@ar.com>]

On Nov 18, 11:26am, Christopher Ambler wrote:
> Subject: Re: Regestry Identifiers
> Rick Wesson (>>) and Dave Crocker (>) said...
>
> >>Would it be better to work out a common set of templates or is everybody
> >>going to use the ones the InterNic uses. Would it not be better that an
> >>IETF WG define these templates and write a public domain parser for them?
>
> > This sounds like an entirely reasonable approach, to me.  It has
> > the benefit of not requiring the IAHC to get into the middle of it but,
> > instead, let the usual standards process develop and mandate it.  An
> > effort like the IAHC can then get by with "implement and conform to all
> > relevant Internet standards" languge and be done with it.

Most ISPs will want to regester domains via a automated system. Using
a web interface might be a great augment to a regestration service but
the vast majority will staii want e-mail templates.

I envision one day that the ISP parses the template on their systems and
directly insert the appropiate info into the regestries database system,
this is what I'm working twards with backwards compatibility with templates.

> I don't know if I agree. Our web interface for .web is, IMHO, much easier
> to use that InterNIC's templates. While we have the ability to parse an
> InterNIC template, we've made some design decisions that make them
> somewhat redundant. For example, we've collapsed the "billing" and
> "administrative" contact into one. We found that most registrations had
> them the same, and it's one less data item. If required to have both, as
> the InterNIC does, that's mandating we abandon what we think is a good
> design decision. Fair? I don't think so.

IMHO an e-mail template is the minimum, I'm tring to draw up some docs
on the topic so I can get some feed back. I'm working on public domain parsers
for a basic set of templates at the moment.

> On the other hand, a MINIMUM set of requirements might be a good thing.
> Perhaps after registries have been selected, we can get together as a
> group and work out a standard that has consensus. But after, not before,
> so that registries can be in on the process.
>
> > 	My very strong personal opinion (strongly personal, strong opinion)
> > is that all DNS-related technical matters -- defining technologies such as
> > protocols and formats -- MUST be pursued ONLY through the IETF.
>
> I agree.
>
> >>The situation I would like to avoid is the one where the ISP is confused
> >>as to the procedures used to regester and manage data related to
> >>    o requesting new domains
> >>    o protecting others from using their name server w/o their approval.
> >>    o notification of deleteing a domain.
> >>    o domain modification.
> >>    o where to get/send domain templates.
>
> If you want to register a .WEB, you go to <http://nic.web>. What could be
> more intutive?

We need a much more defined protocol like we use reg.int and every regestry
gets
a handle from IANA and that handle is mapped into .reg.int. so you could
look up web.reg.int allong with any other TLDs you serve.

> As for security, we're wrestling with that issue now. I'm *VERY* open to
> suggestion. Verisign?

I would like to see a BEFORE-USE gardian implemented. I will be happy to work
twards that goal too.

-Rick


-- 
Rick H. Wesson