[wg-c-1] Re: Next question: which gTLDs? how many?

[Onno Hovers <onno@surfer.xs4all.nl>]

> [....]
>While other discussions continue in parallel, I would like to continue with
>the next questions: Which? How Many?

Why even ask the question "How many"? Is there any good reason to
limit the number of new gTLDs before we even start talking about what new
gTLDs to add?

>The results will make sense if we either decide on a model in which the
>Internet Community is the one to choose the new gTLDs or if more than one
>model is proposed.

>I personally think that it should be the community to choose. Otherwise all
>new gTLDs would only be oriented to "be sold" and not to serve the
>community. gTLDs such as ".arts" which may add much more richness to the
>name space will not be added, because others may bring in more money.

I agree that we should not give in to the greed of a few and the previous
claim nonsense. It should be a choice of the community what
to do with the gTLD space.

>There is a proposal on the table for 7 new gTLDs. It was a list that was
>reached after long public consultation.

>.firm .info .web .arts .rec .nom .shop

>The number seven was a middle point reached between the
>technical/business/IP communities. A list short enough to be easy to handle
>and long enough to add a new dimension to the name space.

What is so good about that list? It looks to me that the names in that
list are very arbitrary and that they are chosen because they 'sound
good' or 'fill a need'.  There is no logic behind that list.

The logic behind the original gTLDs like COM/NET/ORG/EDU/.. was that
they divided between the type of registrant: commercial (COM),
non-profit organizations (ORG), network service providers (NET), etc.

Within that logic there is a clear need for a gTLD for individuals.
If I understand correctly, the NOM TLD was supposed to fill that hole.
Another point (more for wg-c-3) that I would like to make about a
gTLD for individuals is that a first-come first-served kind delegation
of SLDs may not be the best idea. Individuals have a (in most cases)
unique non-contested name, which always has a first-name last-name
kind of structure. Why not use that structure and require
first-name.last-name.NOM ?

The FIRM TLD sounds to me like the same concept of a COM TLD with
another name. The COM TLD has certainly become overcrowded with
more than five million domains in it. But this indicates to me
that a generic commercial TLD is too broad. The problem of COM
overcrowding would be better solved with more subcategories instead
of another COM domain with a different name.

The WEB TLD is one of the worst proposals IMHO. The word WEB is
used for everything in the world wide web. So it will certainly
become an overcrowded TLD. Anyone operating WEB will become a
very rich man. But adding WEB would set a very bad precedent.
Nowadays a site is usually accessed using <protocol-name>.<domain>.<tld>
A website is usually called www.<domain>.<tld>. A file transfer protocol
site is called ftp.<domain>.<tld>. An irc server is usually called
irc.<domain>.<tld>. Sometimes, a site can also be accessed by
<domain>.<tld> using an A record. But by bringing protocol names
like WEB into the root, we would be encouraging a shift to
<domain>.<protocol-name-tld>, like <domain>.web instead of
www.<domain>.<tld>. Next we will have <domain>.list, <domain>.ftp,
<domain>.irc, <domain>.mail, <domain>.dns, etc. This would mean that
the same domains get registered in different protocol-name TLDs.
This would be a waste, IMO.

The other proposed gTLDs (INFO, ARTS, REC, SHOP) are more like the kind
of gTLDs which need to be added: categories. INFO and SHOP are still very
broad, though. Adding new categories instead of just random
good sounding names avoids a trademark hell both at the gTLD level
(proprietory trademarked gTLDs) and at the SLD level. And it creates
a more sensible division of the namespace, one which can last longer than
a decade.

If the new set of gTLDs form a comprehensive set of categories like
the categories in Yahoo, the Yellow Pages or whatever, it will effectively
kill any chance that a proprietary registry with a trademark on that
word can legally challenge that gTLD. How big do you think the chances
are that someone with a trademark on Arts could challenge Yahoo because
they use an Arts&Humanities category? 

If the set of new gTLDs form a comprehensive set of categories, instead
of a set of overly broad popular gTLDs which could apply to anyone,
the chance of trademark disputes over second-level domains in these new
gTLDs is greatly reduced, since most trademarks only apply to a limited
set of categories.

>The merits of having .sex or .xxx were discussed. The reasons for having
>this gTLD seem to be related to classifying the contents of website, and
>enforcing that other gTLDs do not have any sexual contents. As this WG does
>not deal with other existing gTLDs, we could only propose that it be
>created, not that action be taken to move sexual contents out of other
>gTLDs. I don't see the merit of this independent action, and therefore of
>having the .sex gTLD.

There is a vivid pornography industry on the internet, so it makes
sense to add SEX as another category. However, we should not fool ourselves
that when we add a SEX TLD all porno sites will automatically start
using SEX, and no porno sites will be found in other TLDs. 
But the fact that SEX does not solve the classifying/censoring problem
is no reason not to include SEX.

Regards,
-- 
Onno Hovers (onno@surfer.xs4all.nl)