Re: [idn] Re: CDNC Final Comments on Last call of IDN drafts

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

 



Simon Josefsson <simon plus idn at josefsson dot org> wrote:

>> It has never been a goal of IDN to create equivalence between
>> different character sets.
>
> This means IDN is not guaranteed to be secure on non-Unicode systems.
> There are alot of non-Unicode systems out there today...

This problem, which IMHO is overstated anyway, is not confined to
Unicode.  AFAIK there are no *formal* equivalence tables between all the
possible overlapping character sets in existence, whether mapping
through Unicode or not.

That means if you want your domain name searches to match *only* for the
character set you are using, things are fine, but if you want the search
to span across different character sets, you are no better off without
Unicode than with it, because any mapping tables are informal or
proprietary.  And without Unicode, the user (or browser or mail agent)
is required to declare the encoding for EVERY domain name and derived
identifier it sends, because there is no other "obvious" default.  ISO
2022 doesn't solve this problem because you still need mapping tables.

It's easy to blame the big, monolithic, high-profile Unicode Consortium
for "dropping the ball," but their responsibility is to create and
maintain Unicode and provide "best available information" mappings for
everything else.  That is what they have done.  They don't create or
maintain other standards; national bodies and vendors do.  And those
national bodies and vendors have extended their standards in proprietary
ways, while browsers and e-mail clients have been loose in character
encoding terminology (e.g. Outlook Express allows me to format this
message in "JIS," which could mean many things).  Neither Unicode nor
IDN, nor any other body, can be responsible for providing official,
authoritative mappings for every character set and corporate extension
out there.

If national bodies and corporations have *official* mappings between
their standards and Unicode, then perhaps those organizations can
contribute them to Unicode (and promise to maintain them regularly and
promptly!) so they are available in an easily accessible, central
location.

-Doug Ewell
 Fullerton, California


[Index of Archives]     [IETF Annoucements]     [IETF]     [IP Storage]     [Yosemite News]     [Linux SCTP]     [Linux Newbies]     [Fedora Users]