W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > May to August 1996

Re: proposed HTTP changes for charset

From: Larry Masinter <masinter@parc.xerox.com>
Date: Mon, 8 Jul 1996 11:44:41 PDT
To: yergeau@alis.ca
Cc: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
Message-Id: <96Jul8.114441pdt."2757"@golden.parc.xerox.com>
The IETF process does not prevent the promotion of standards with
'bias'. From RFC 1602:

>      IETF Working Groups are generally able to reach consensus, which
>      sometimes requires difficult compromises between differing
>      technical solutions.  However, there are times when even
>      reasonable and knowledgeable people are unable to agree.  To
>      achieve the goals of openness and fairness, such conflicts must be
>      resolved with a process of open review and discussion.
>      Participants in a Working Group may disagree with Working Group
>      decisions, based either upon the belief that their own views are
>      not being adequately considered or the belief that the Working
>      Group made a technical choice which essentially will not work.
>      The first issue is a difficulty with Working Group process, and
>      the latter is an assertion of technical error.  These two kinds of
>      disagreements may have different kinds of final outcome, but the
>      resolution process is the same for both cases.

I believe that we have considered the alternatives on these issues
adequately. Although I understand the basis for your disagreement, I
don't think there is grounds for you to claim that your own views "are
not being adequately considered".  And I believe that the choice that
we made will 'work', with some careful attention to backward
compatibility on the part of implementors -- but this kind of
attention is always necessary and part of implementation advice, and
not necessarily the protocol itself. So I don't think there are
grounds to claim that "the Working Group made a technical choice which
essentially will not work." So, that's why I think we're done. If you
disagree on either of these points, you must convince IESG and/or IAB.

There is a third alternative listed in RFC 1602:

>      *    If a concern involves a reasonable difference in technical
>           approach, but does not substantiate a claim that the Working
>           Group decision will fail to perform adequately, the Working
>           Group participant may wish to pursue formation of a separate
>           Working Group.  The IESG and IAB encourage alternative points
>           of view and the development of technical options, allowing
>           the general Internet community to show preference by making
>           its own choices, rather than by having legislated decisions.

I don't think this is actually a serious possibility, but it may be
the only alternative left if you follow out the process.

This is all a very touchy issue. Hopefully, those of you who want to
pursue this issue can reach those implementors whose products are in
"beta" (which is what I presume the 'b1' is all about) and convince
them to fix their browsers if they break when handed explicit charset,
at the same time convincing content providers to label their
content. Doing so is important to the community, but not something
that can easily be accomplished by sprinkling more 'MUST' and
'SHOULD's in the HTTP/1.1 specification.

Regards,

Larry
Received on Monday, 8 July 1996 11:53:51 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:32:04 EDT