Re: TR6 charset

Larry Masinter (masinter@parc.xerox.com)
Thu, 20 Aug 1998 13:01:46 -0700 (PDT)


Date: Thu, 20 Aug 1998 13:01:46 -0700 (PDT)
From: Larry Masinter <masinter@parc.xerox.com>
Subject: RE: TR6 charset
In-reply-to: <Pine.SOL.3.95.980819132548.25073V-100000@elwood.innosoft.com>
To: Chris Newman <Chris.Newman@INNOSOFT.COM>, Roman Czyborra <czyborra@dds.nl>
Cc: unicode@unicode.org, ietf-charsets@iana.org
Message-id: <001601bdcc75$5c1baf00$15d0000d@copper-208.parc.xerox.com>

> Personally, I think registering this charset will be counter-productive to
> the acceptance of Unicode and is a bad idea.

Chris, it may be counter-productive to USE a varient encoding, but keeping
something from being registered doesn't seem to keep it from being used,
or even discourage its use.

It's probably BETTER to register it, and include in the registration
a note about why (or at least when) you shouldn't use it.

> The more variant encodings of Unicode that end up used in practice, the
> more people will hate Unicode and want to use their good old localized
> character set which "just worked."  While Unicode is a relatively easy
> sell to implementors it is a hard sell to end-users and more encoding
> schemes just makes it that much harder.
> 
> This encoding scheme is particularly bad because it can't be used in a
> MIME text/* media type and it permits NUL octets, so it will have to be
> encoded in most transport environments.

For example.



--Boundary (ID uEbHHWxWEwCKT9wM3evJ5w)