Re: kerberos auth against w2k server with 8 bit chars in password

[those who know me have no need of my name]

<9m31ge$ej9$[EMAIL PROTECTED]> divulged:

>Actually, nobody adheres to the standard.  Not MIT, not Heimdal, not
>Microsoft.  The RFC says that the principal name is a ASN.1 GeneralString
>which means that all strings other than plain ASCII must be encoded
>using ISO-2022 escape sequences.  

that's what i meant by `the deployed world,' i.e., microsoft chose to make
win2k 8bit handling neither what the standard requires (iso-2022) nor what
virtually all other kerberos distributions do (raw octet-string, therefore
usually an iso-8859 alphabet).

>There is a serious discussion taking place about ways to handle this
>problem.  

any feel for when we might see a solution approved, or near enough to begin
serious work?

if an improvement on utf-8 were developed as a result i wouldn't be
unhappy.  utf-8 sucks for those regions of the world that don't typically
use ascii codes, e.g., israel, which forces a increase in the total
data-stream size, sometimes a huge one.  something like a locking shift is
needed so you can shift a different basic character set into the low order
code points.

-- 
okay, have a sig then