[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: New Internet Draft on registering IDNs

tedd <tedd@xxxxxxxxxxxx> wrote:

> > Similarly, fu<umlaut>hrer.de might be unavailable to everyone except
> > the owner of fuehrer.de.
> So then, under this policy the owner of fuehrer.de won't have to do
> anything?  He doesn't have to take any steps protect his original
> name.  He doesn't even have to register fu<umlaut>hrer.de?  He can
> just sit there, do nothing and let your policy protect him -- nice for
> him.

Yes, nice for him, and nice for everyone in the internet community who
doesn't have get surprised by fuehrer.de and fu<umlaut>hrer.de belonging
to unrelated organizations.

> What about other TLD's?  Does he not have to worry about them as well?

The whole purpose of a hierarchical domain name system is so that the
exact same label foo can exist independently under multiple parent
domains.  It has always been understood that foo.org and foo.com and
foo.de are three independent names that don't necessarily have any
connection.  No one is proposing to change that.

> This proposed policy doesn't sound right to me and I am sure that
> registrars won't like missing the revenue generated by customers
> protecting their name.

If registry revenue is the only metric, then the entire topic of this
mailing list is completely pointless.  No registration should block any
other registration, there should be no bundles, and registrants should
have to register every individual name that they don't want someone else
to have.

However, many people are concerned that the whole naming system will
lose its value if human beings cannot recognize when two names are the
same and when they are different.

So registries will try to find a balance, designing tables that bundle
enough names together so that names continue to be recognizable and
useful to the general public, without bundling so many names together
that registrants would get away with paying a single fee for what are
obviously (even to humans) multiple names.

> If the end result of this policy making makes it such that the
> original owner of a DN doesn't have to worry about like-registrations,
> then registrars will lose money.

This mailing list is designing an architecture, not the actual tables.
The tables will be designed by the registries themselves, who will
presumably look out for their own interests.

I have no position on whether u<umlaut> and ue ought to be considered
neighbors in .de; my point was that *if* the registration of
fu<umlaut>hrer.de blocks the registration of fuehrer.de, then the
registration of fuehrer.de ought to block the registration of
fu<umlaut>hrer.de, and therefore if fuehrer.de is already registered
when the bundling system is introduced, it should be grandfathered in
and expanded to whatever bundle it would have been if the bundling
system had existed when it was originally registered.

> It appears to me (perhaps I'm wrong) that this group is trying to
> predict and solve all possible problems that may arise from IDN
> registrations because of look-alike possibilities within the Unicode
> database.

No, this group is not trying to predict them all.  It is trying to
devise an architecture that can be extended to deal with the issues
incrementally as they are discovered.  The first tables to be developed
(not by this group, but by experts in the relevant language) will
prohibit all Unicode characters except those essential to a particular
language, and will contain carefully crafted mappings for those
characters.  Then some tables will be developed that combine two
languages, with special care taken for the characters used by both.
Gradually, as more experience and expertise is developed, more inclusive
tables will be created.  Each registry will decide what table to use for
each of its zones.

> I don't know the actual number of additional characters added thus
> far, but the upward limit is 65,535.

It's actually about a million.

> So, as I see it, you will have some 65,000 different possibilities
> of character confusion at a single character domain level (i.e.,
> a.com). Now, move to two characters (aa.com) and figure becomes much
> larger -- something in the order of 65000 x 65000 range.
> Now, what's the upper limit to the number of characters allowed in a
> domain name and what's it's factorial?

That's irrelevant.  If I'm told that ue and u<umlaut> are confusable, I
don't need to be told separately that xue and xu<umlaut> are confusable,
and that foouebar and foou<umlaut>bar are confusable.

> you're making decisions about glyphs in other languages that are not
> you're own.

This group is certainly are not doing that.  It's not designing the

> I think this group has made some significant progress in that
> some characters have been already mapped to others -- such as all
> occurrences of glyphs looking like "A" and have been mapped to "a" and
> so on.

What are you refering to?  I don't know of anyone who has done that.

> considering the shear number of combinations and thoughtful
> considerations required for each one -- I don't think this group has
> enough time nor resources to accomplish the task.

Right, which is why this group is not even going to try.  At most, it's
going to develop an architecture into which tables can be plugged, and
other groups with the needed expertise will devise the tables.