Re: [ga] a quote from Lynn

[Sotiris Sotiropoulos <sotiris@hermesnetwork.com>]

Sandy Harris wrote:

> Eric Dierker wrote:<snip>
>
> > Yes I take this a little out of context but I do not like the
> > fundamental position that any system would require only one right answer
> > to the same question.
>
> That's been one of the basic ideas all along. Way back in 1987, RFC 1024
> "DOMAIN NAMES - CONCEPTS AND FACILITIES" states the design goal for the
> DNS system, and the very first one was:
>
> " The primary goal is a consistent name space ...
>
> This is fundamental.

I think we should go a bit further and ask the following question about
consistency: `Consistent' in whose favour?  To be honest Sandy, I think you
and I can have an interesting discussion...

> > It sounds like he is saying that users are too
> > stupid to handle a choice therefor we must not give it to them.  No
> > rereading his policy paper; it does not sound like it, it is it.
> >
> > There may be reasons for this policy but to lay it off on the users
> > being to stupid to make alternating choices between roots is wrong both
> > technically and morally.
>
> It's not that the users are stupid or incapable of making choices, the
> issue is that the system should not force them to make such choices.
> Names should map to addresses and resources in a predictable and
> consistent way.

But ain't this the rub?  Predictability.  It's an interesting notion to be
sure.  But what exactly is the predictability factor in the case of a global
DNS, within generic second-level names esp. within the COM top level domain?

> When I type 'www.whatever.biz' into my browser, I don't want the
> result to depend on how some administrator at my ISP has set up
> their DNS server. Yes, I could figure out what was going on and
> complain to the ISP, switch ISPs or set up my own name server if
> I didn't like it. However, I shouldn't have to!

I agree.  The consistency of namespace is important for everyone.  However,
the standard(s) of consistency are what we must focus on and correct.

> So the question is not whether we need a single consistent namespace,
> but how to build it.
>
> I'd say we do that by accepting the fact that ICANN has been created
> to do this job and given the responsibility for that namespace, then
> trying to make ICANN work as it should. Among the things I'd like to
> push for:

But ICANN needs the balance of public input to the tune of 9 AtLarge Directors
to bring about a more equitable perspective from which to debate and decide on
matters which relate to the management of the consistent namespace we all
desire and require.

> The complex political compromise process during ICANN formation led
> to a board that was to have nine elected at large directors and nine
> from various interest groups. Step one is to achieve that balance.
> (It isn't the balance I'd have chosen -- I'd like to see public
> interest groups like EFF with as many votes as all the business
> constituencies combined -- but it's in the bylaws and better than
> what we have to date.)

Agreed.

> Think about remedies for the over-use and over-selling of .com, the
> biggest problem in the current namespace.
> (My first thought is just stop .com registration, but is there a
> better solution?)

Why stop .com registration?  It's still possible to get decent, memorable
labels.  The generic or popular labels in *any* TLD will be much coveted, it's
simply that COM names are currently the most popular.  I think the
introduction of new TLDs will not have much of an effect on COM overall, but
to think that any really choice names in *any* TLD will be easy to come by is
rather naive to say the least...  So, I think market forces that relate to DNS
should be allowed to flourish and develop as the market and demand dictates,
and not as ICANN directs.

> Fix the UDRP; see my comments in other threads.
> (I'd like to say scrap it, but doubt that's practical.)

If it can be "fixed" and is retained, it must not apply to ccTLDs.  Countries
have their own laws for TM protection and infringement.  However, I am for
scrapping it altogether, I think it's more than served its biased purpose and
it's time to drop the whole fiasco.  I completely concur with the proposal for
a three year original registration protection for SLDs of generic TLD
domains.  How about you Sandy?

> Technically, a TLD is no harder to set up than an SLD. The $50,000
> application fee is absurd. Scrap it. Aim at 100 new TLDs by 2002,
> with cost well under $1000. If it's possible to give the "alternate
> root" people some sort of olive branch in the process, so much the
> better.

I admire your pragmatism.

Sincerely,

Sotiris Sotiropoulos

P.S.  For those who've been trying to reach me:  I have been in and out of
town over the past week, and I expect to be periodically unavailable for the
rest of this week.  I will be in touch when convenient.

--
This message was passed to you via the ga@dnso.org list.
Send mail to majordomo@dnso.org to unsubscribe
("unsubscribe ga" in the body of the message).
Archives at http://www.dnso.org/archives.html