Re: Subaddressing

Marc Mutz <[email protected]>
Newsgroups gmane.ietf.imaa
Organization KDE
Message-ID <[email protected]>
On Saturday 15 February 2003 19:31, John C Klensin wrote:
<snip>
>         Whether one uses an IDNA-like approach, or something of
>         the general character of 8BITADDRESSES (I suspect I'm
>         going fairly quickly come to hate that term), we don't
>         want to disrupt local-address opacity.  Consequently,
>         anything that is to be done with subaddresses --or any
>         other special-purpose parsing of the local-part into
>         subdivisions for differential processing-- must be
>         done before encoding on the sending side and processed
>         post-decoding on the receiving side.
<snip>

One very noteable feature of IDNA is that it requires no modification 
whatsoever on the server side.

OK, let this sink in.

Now, why should IMAA?

I'd like to lay out why it _shouldn't_.

I haven't followed the IDNA discussions, so I don't know how much of an 
issue server security and scalability had on the design of IDNA, but, 
looking back, IDNA does a very good job of keeping complexity out of 
the server. The complexity is completely contained in the clients, 
which results in better scalability, since the clients run on 
over-(cpu)-powered PCs, and in better server security, since they don't 
change at all.

So the requirements to balance are
1. keeping complexity out of the server for the reasons stated above
2. keeping the concept of opaque local-parts in all it's strictness.

Why are those requirements contrary and that need to be balanced?

Because strictly opaque local-parts require the interpretation to happen 
after the ToUnicode operation, which increases the complexity of the 
server (US-ASCII/char*->UTF-8/wchar_t transition).

The reverse also holds: Keeping complexity out of the server requires 
(partial) interpretation of the local-part (splitting at a set of 
characters) to be performed before the ToUnicode operation.

I assert - and the IDNA design agrees with me here - that it is a great 
deal more important to keep complexity out of the server than to keep 
the strict opaque local-part concept.

Marc

-- 
"Similia similibus curentur"
           -- Bush's new motto in fighting terrorism.
signature.asc (application/pgp-signature, 189 B)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQA+Tq/d3oWD+L2/6DgRAtWtAKCFolNmJwZmUATJUaqV/Ft4WtUqBACdEEwD
iU4Yrh9mgoOgyrWyWbDR6Yo=
=Abvx
-----END PGP SIGNATURE-----
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.