Re: The use of MIME on the web: issues, toward a revised "finding" or joint W3C/IETF document

Adding to the confusion, the BNF in BCP13 is:

       type-name = reg-name
       subtype-name = reg-name

At any rate, it's important to remember that HTTP is a MIME-like protocol; i.e., informed but not necessarily restricted by MIME.

HTTPbis lists the differences in <http://tools.ietf.org/html/draft-ietf-httpbis-p3-payload-09#appendix-A>. One way to resolve some of these issues is to add text that outlines how HTTP selectively leverages the media type registry. 

Cheers,




On 05/04/2010, at 4:09 AM, Larry Masinter wrote:

> The main problem I've seen is when people are not careful to 
> distinguish the "internet media type" from the "content-type
> string", because the latter can contain parameters
> 
> text/plain;charset=iso-8859-1  vs. text/plain;charset=windows-1252,
> for example, which share a "Internet Media Type" but are
> different content-type strings.
> 
> So
> 
> draft-abarth-mime-sniff talks about MIME type sniffing, but the
> charset sniffing is in a different document. I think the general
> topic of "sniffing" (i.e., examining content using heuristics
> to determine the likely actual type, whether or not there is
> a content-type label supplied) might be approached more holistically
> if the terminology were clearer.
> 
> I'm not sure if there are other terminology confusions
> that are as serious.
> 
> MIME was the protocol design which defined Internet Media Type
> registry, subsequently adopted by HTTP (between HTTP 0.9
> and HTTP 1.0), so it's understandable how "MIME type" and
> "Internet Media Type" are sometimes used interchangeably.
> 
> 
> Larry
> 
> 
> 
> -----Original Message-----
> From: Dan Brickley [mailto:danbri@danbri.org] 
> Sent: Sunday, April 04, 2010 9:15 AM
> To: Sebastien Lambla
> Cc: Larry Masinter; www-tag@w3.org
> Subject: Re: The use of MIME on the web: issues, toward a revised
> "finding" or joint W3C/IETF document
> 
> On Sun, Apr 4, 2010 at 3:06 PM, Sebastien Lambla <seb@serialseb.com>
> wrote:
>> As both a web framework author and trainer, I find that there is
> lots of confusion introduced when various names are introduced for
> various areas, such as the Internet Media Type being referred to as
> MIME, as in "MIME sniffing" or "The use of MIME", as opposed to the
> historical MIME type.
>> 
>> For me, tomatoes and tomatoes, but for the people that I introduce
> to those standards, I feel that those things are a barrier to wider
> communication.
> 
> Yes, this is a good point. I've been involved in standards plenty but
> have only a pretty vague instinct for how I'm supposed to refer to
> this stuff. Generally I tend to *say* "MIME-type", but have this loose
> guilty sense that the real label is "Media type", but that as a phrase
> seems somehow less specific whereas "MIME-type" has the feel of a
> precise technical term. In either case I mean "those things like
> 'text/html' and 'application/rdf+xml'". It would be great to have a
> terminology summary / bluffer's guide.
> 
> cheers,
> 
> Dan
> 
> 


--
Mark Nottingham     http://www.mnot.net/

Received on Tuesday, 6 April 2010 01:17:55 UTC