Re: [idn] Prohibit CDN code points

[David Hopwood <david.hopwood@zetnet.co.uk>]

-----BEGIN PGP SIGNED MESSAGE-----

Edmon wrote:
> From: <DougEwell2@cs.com>
> > If everything is 1-to-1 between TC and SC, the algorithm is trivial.
> > I was talking about the algorithm that would be necessary to do TC/SC
> > conversion in the case of 1-to-many or many-to-1 conversions.  But,
> > since you say there are none, I guess we can proceed with implementing
> > the tables.
>
> It is possible to implement a 1-to-many or many-to-1 conversion.
> All it is required is that all "equivalent" characters be matched as
> "equivalent" in the authoritative DNS.

This doesn't work for the same reason that folding look-alike characters
doesn't work: you would need to put characters into equivalence classes
according to the transitive closure of the SC->TC and TC->SC mappings, and
that ends up putting completely unrelated characters in the same class.

I.e. start with a single character. Add all its TC/SC variants. For each
of those variants, add all their TC/SC variants, and so on, recursively.
For all I know you may end up with all Han characters in the same
equivalence class; perhaps not, but you certainly end up with very large
classes.

If TC/SC folding is done, it can only be done for the 1-1 subset of
mappings, and any variations in the algorithm must be based only on the
name itself, without making use of any other context (such as language,
etc.) I'm not qualified to judge whether it is useful to do TC/SC
folding under these contraints, but I suggest that those who advocate
it concentrate on that question, rather than making proposals that aren't
implementable.

- -- 
David Hopwood <david.hopwood@zetnet.co.uk>

Home page & PGP public key: http://www.users.zetnet.co.uk/hopwood/
RSA 2048-bit; fingerprint 71 8E A6 23 0E D3 4C E5  0F 69 8C D4 FA 66 15 01
Nothing in this message is intended to be legally binding. If I revoke a
public key but refuse to specify why, it is because the private key has been
seized under the Regulation of Investigatory Powers Act; see www.fipr.org/rip


-----BEGIN PGP SIGNATURE-----
Version: 2.6.3i
Charset: noconv

iQEVAwUBPE42MDkCAxeYt5gVAQG8iAgAqZtwQXJD1L80QdOi0t9dOhX2/cF90CKe
pF4CmdBsA9C2mOcRmzQPTXDt7aIApFkE1A2N9TP+avo38J4fqZF/qPVKb/ZSU6uk
JgZodaeyiyyOhc7p2MZ5+MR/3/EwWbrAXgpWkpG8JbbnLNKQ8asR81bPdCq0L0gK
a3Ifie1IHnHJcefWOzuBTHhfPcUkFLDhaJ/8YnAUjzDTaQDOohg7HgSNw6JBUoaO
/bgrPGRQBl9svrvQvVSSw4L4a0LJlQwikWUG7QVyH9W0mYbqO5E3xwJgKIh3qJn+
W2jN0LSjVYXmHamrTQXw2lcHKMxcSoyFLpLS0BgJOtkkTAUyjBG5CA==
=MiLW
-----END PGP SIGNATURE-----