[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [idn] Why we can go directly to UTF-8
- To: Martin Duerst <duerst@w3.org>
- Subject: Re: [idn] Why we can go directly to UTF-8
- From: "Eric A. Hall" <ehall@ehsco.com>
- Date: Thu, 24 May 2001 20:38:07 -0700
- CC: Keith Moore <moore@cs.utk.edu>, idn@ops.ietf.org
- Delivery-date: Thu, 24 May 2001 20:38:43 -0700
- Envelope-to: idn-data@psg.com
- Organization: EHS Company
Martin Duerst wrote:
> 'every' is too general. It's just those in the relevant paths.
I agree that "every" is too general, but so is "just those in the relevant
paths".
The number of systems which will break with UTF-8 is greater than the
number of systems which break when they encounter an underscore in the
domain name. Moreover, the "path" includes protocols such as BOOTP/DHCP
clients and servers, X clients and servers, and a bunch of other stuff.
Moreover, some of these systems are firmare-based boxes that are no longer
maintained. Think of all the terminal servers and printer servers that
were acquired, orphaned or died.
If we were to go to UTF-8, we would pretty much have to cleanly break with
history in order to ensure interoperability.
--
Eric A. Hall http://www.ehsco.com/
Internet Core Protocols http://www.oreilly.com/catalog/coreprot/