Re: homograph attacks

"Martin v. Löwis" <[email protected]> Thu, 17 Feb 2005 02:08:27 +0100
Newsgroups gmane.ietf.idn
Message-ID <[email protected]>
Michel Suignard wrote:
>>> I have seen European standard bodies spending forests of paper to
>>> try to establish these language tables, but there have never been
>>> an authoritative version because simply you can't.

> This is not a language table, it is basically all IDN allowable Latin
> characters from Unicode block 0000-007F and 0100-017F. This covers
> many European languages. As long as you don't call it a language
> table we may be in agreement. Note btw that by doing this they are
> excluding Welsh and Irish Gaelic which needs Latin characters in the
> 1Exx range. And obviously all Latin letters from the Extended-B are
> not included, which is excluding de facto many European minorities.

All true - but "so what?". This is an official policy what characters
can be used in a "German" domain name, in the .de zone. Whether that
makes it a "language table", I don't care. This kind of table is what
we have been talking about all along.

> I am not advocating for script mix as
> you may hint. I am not saying either that the the DENIC list is bad.
> It is in fact reasonable (with the minor caveat expressed above), but
> it is by no mean a German language table or even an European language
> table.

This is not the point. The point was your claim: "there cannot be an
official table", and I'm merely trying to say "but there is". Now,
I see that you make a difference between a character table for a
language, and a "language table"; I was unaware that "language table"
is a reserved term.

> I never said that. I am just saying that subsetting Unicode script
> per language won't work.

But we are not talking about typesetting here: this is about DNS.

Regards,
Martin