Re: misconceptions about MIME [long]

Tim Berners-Lee (timbl@www3.cern.ch)
Thu, 29 Oct 92 16:38:18 +0100


Date: Thu, 29 Oct 92 16:38:18 +0100
From: Tim Berners-Lee <timbl@www3.cern.ch>
Message-Id: <9210291538.AA06151@www3.cern.ch>
To: Dan Connolly <connolly@pixel.convex.com>
Subject: Re: misconceptions about MIME [long] 
Cc: www-talk@nxoc01.cern.ch



>Would it be reasonable of us to take Dan's cue and change HTTP2 so that
>  >the returned message is a MIME format?
>  

>  You mean use MIME content types for HTTP2 format negociation: exactly!


	Sounds reasonable to me in the interest of reasonableness.

>  > and also
>  >use an IETF standard URL a the external reference format?
>  

>  I'm not sure I understand this part. Sure you can have a format
>  called text/html, and sure you can put URLs in it, but that
>  doesn't mean URLs have to become part of the MIME standard.
>  


I mean that there is a clash between the URL/URI work and the MIME
external context format.  For example there are two ways of writing
a reference to an FTP archive. It would be neat to amalgamate them.
I  personally think the MIME format is too longwinded (look at the
ref on the end of your earlier message -- that could have been 1 line).

Of couse a generic URL could be made another specific MIME external
reference format, as FTP is now, which would provide a transistion.

---- by the way...

Data formats:
It is true that a generic system ought to be able to represent the
complexities of reality. But it is also true that if one makes
arbitrary choices and only accomodates those, then reality may in the
end toe the line.  MIME registration will give a lot of weight to
a particular set of paramaters for a data format, and so the
use of arbitrary deviants from the registered formats would tend to
die out.

To a certain extent only, of course. There will always be people
who use weird formats, but then noone will be able to
hear anything of them anyway... ;-)

	Tim