Re: PAB some policy advising: Arts

Kent Crispin (kent@songbird.com)
Wed, 25 Feb 1998 09:02:46 -0800


On Wed, Feb 25, 1998 at 03:15:35AM +0000, William Allen Simpson wrote:
> > From: Dan Busarow <dan@dpcsys.com>
> > Maybe someone is multiple registering famous marks but I haven't
> > seen it.
>
> (sigh) We've only been discussing this for 4 or 5 years, so perhaps you
> missed it. It was Proctor and Gamble, among others. That diarrhea was
> one of dozens of names multiple registered in .com, .org and .net, and
> they attempted .edu. My reference should have set off alarming
> memories.

I know from personal knowledge that both of you have been involved in
this debate for a long time...longer than I.

> > For example, many ISP's have both .com and .net.
> > There may even be a BCP on that practice.
> >
> Cough, cough. Most ISPs do not have both .com and .net. And I do not
> remember a BCP even remotely suggesting such a thing. I'm pretty sure
> that I would have noticed.

While I wouldn't claim to be representative of anything....
Songbird(r) has both .com and .net, and a trademark, to boot.
Originally Songbird was "Songbird Software", started for the purpose
of writing music software. The .com was appropriate. Later Songbird
became a web page hosting/consulting service, so the .net was
appropriate. But I keep the .com one as well, because both uses fit.

> > Because they want it and are willing to pay for it. It's really
> > that simple. This is not a technological issue, it's business and ego.
>
> Actually, these _are_ technological issues. This is all about keeping
> the Internet useful!

Pretty broad statements on both your parts, I would say :-). There is
absolutely no doubt that ego, business, and the law are very important
in domain name issues, and, while there are also clear technical
issues, they play second fiddle, because in fact the technical
constraints are rather loose. Eg: there are close to a million names
in .com; technically there should be more hierarchy, but there isn't.
Ergo: The technological concerns are not as weighty as the ego and
business concerns.

> How do similarly named entities in different fields both register? We
> don't want folks registering in every TLD, so that similarly named
> entities can use the Internet.

This is not a technical issue, is it? Technically dns doesn't know
who registers what.

> How long will it take to do the gTLD key signing? How often will the
> gTLDs be signed? We don't want bloated TLDs.

This is indeed a technical issue, but the performance limits are broad
and unknown, and until they are clear and measurable the technical
constraints don't carry much weight.

[...]

> > > How about:
> > >
> > > 2. The same or similar name shall not be registered by the same
> > > organization in any other zone of the DNS (such as under a
> > > country TLD), unless an individual exception is granted under
> > > the registration appeal process (described below).
> >
> > Mark, from NetNames, already questioned how you would police the original,
> > this only makes the situation worse.
> >
>
> Police? Are we going to need police?
>
> The same way as .edu and .gov and .mil....

You don't need much policing when not much of value is at stake.
Names in .edu, .gov, and .mil are not high-value names, so generally
disputes can be resolved without too much friction. And I suspect
that names in .arts will not be high-value names, usually (though
consider mgm.arts?, time-warner.arts? chicago-symphony.arts? I
retract my statement...).

> This is the same mechanism for contest of names. It _is_ a subset of
> contest of names.

It is, but the ACPs are designed to work with a subset, as well --
those concerning intellectual property issues, and so they follow
closely the general principles of IP law. The rules you are
suggesting have no legal background to rely on -- they are rules
purely concocted for DNS. As soon as a high-value name is at stake
the rules *will* be challenged in court. Indeed, we can have in the
contract that you agree to abide by the rules of the TLD charter, etc.
But then the charter becomes a serious legal document, and the rules
it defines must withstand legal scrutiny.

In fact, I find it hard to concoct a realistic example where IP concerns
don't cover what your rule is supposed to cover. Here's a feeble attempt:

ABC Inc, a multimedia company, registers abc.arts and abc.com. Along
comes ABC Ltd, a music production company, comes along and says they
want abc.arts, and invokes the .arts charter. The registrar removes
ABC Inc's registration, and gives it to ABC Ltd. ABC Inc sues,
challenging the charter, claiming that the *charter* unfairly
discriminates against them, and that it is unenforcably vague... Or
the registrar does nothing, and tells the parties to work it out in
court. Suit and counter suit. The judge looks at the respective
contracts, and casts his eyes over the charter...

The point is that the charter becomes part of the registration
agreement. It *is* a legal document, and the registrars will be
involved in enforcing it, so they are a target for legal action.

It's true that the registrars can bear that burden -- it's a cost that
would be passed on to domain name customers, in the end. We have to
balance whether the cost of enforcing the restriction is worth that
expense to the end users. It might be, it might not be -- it is
certainly a debatable point.

But if so, the charter better be a legally meaningful document, and
therefore we need expert opinion on it.

> You either want exceptions, or you don't.
>
> But the .arts has to be for arts, and not for diarrhea. Diarrhea by any
> other name smells bad.

Yes. Perhaps we could eliminate it from the discussion (hyuk hyuk hyuk).

> > You're not going to stop speculators with rules. (The ACP's giving
> > added weight to a challenge against a speculator excepted)
>
> If we "are not going to stop speculators with rules", then we might as
> well pack up and go home. We _are_ a "policy" advisory body. We are
> supposed to write rules.

Indeed. And we have to debate which rules are reasonable and which
ones are not.

-- 
Kent Crispin, PAB Chair			"No reason to get excited",
kent@songbird.com			the thief he kindly spoke...
PGP fingerprint:   B1 8B 72 ED 55 21 5E 44  61 F4 58 0F 72 10 65 55
http://songbird.com/kent/pgp_key.html