Re: comments: UTF-16, an encoding of ISO 10646 to Informational

From: A. Vine (
Date: Wed Aug 18 1999 - 15:07:09 EDT

Frank da Cruz wrote:
> > The current state of the draft is what it is precisely because of a
> > perceived consensus that the IETF is NOT in a position to legislate byte
> > order.
> >
> But it is. If the IETF allows UTF-16 to be sent in different byte orders,
> it is legislating that every application has to be coded for every byte order,
> instead of just one. How is that better than legislating that every
> application should be coded to put only one byte order on the wire and expect
> only one from the wire? If we allowed this sort of waffling with (e.g.)
> IP addresses, the Internet wouldn't even work.

The IETF tends to lean toward the "already implemented" rather than the
"dictatorial setting of" standards. Usually Internet-drafts are written on the
basis of something which has been implemented and is working to some degree.
Prescribing what should be done doesn't seem to work very well in the experience
of the IETF, and so they advise thinking through as many scenarios as possible
(and if possible, implementing one or more) to verify that the proposal will

It tends to be more of a cooperative system than a ruling body.


Andrea Vine
Sun-Netscape Alliance messaging i18n architect
I always wanted to be an architect. }sigh{  Of course, I _am_ an architect.

This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:51 EDT