W3C home > Mailing lists > Public > public-html@w3.org > August 2007

Re: review of content type rules by IETF/HTTP community

From: Robert Burns <rob@robburns.com>
Date: Sun, 19 Aug 2007 18:45:27 -0500
Message-Id: <D8D2BAC5-C834-4ADB-BD86-2D20EB87CDBE@robburns.com>
Cc: Dan Connolly <connolly@w3.org>
To: "public-html@w3.org WG" <public-html@w3.org>

On Aug 19, 2007, at 6:23 PM, Robert Burns wrote:

> On Aug 17, 2007, at 3:45 PM, Dan Connolly wrote:
>> According to a straightforward architecture for content types in the
>> Web[META], the HTTP specification should suffice and the HTML 5
>> specification need not specify another algorithm. But that  
>> architecture
>> assumes that Web publishers (server adminstrators and content
>> developers) reliably label content. Observing that labelling by Web
>> publishers is widely unreliable, and software that works around these
>> problems is widespread
> BTW, do we have data on this? Do these content-type headers suffer  
> from mislabeling on a widespread basis? It seems to me the more we  
> give up on this the more we'll lose this feature. We'd have to  
> institute a new header, "really-the-content-type", and make that  
> authoritative.

I meant to add here, that I'm asking about this, because my own  
experience is that I rarely encounter mis-configured HTTP servers  
anymore with respect to content-type. I do encounter them often with  
respect to charset. However, content-type is easily handled by  
mapping filename extensions  (visible metadata) to MIME types. While  
charset encoding requires some other knowledge about the file. Apple  
I believe has started storing charrset encoding in a file's extended  
attributes. The UTF encodings offer the BOM technique but that's not  
widely deployed (or not wide enough for my taste).

In any event, I"m surprised to hear that there's any widespread  
problem to fix with the content-type MIME types. The charset issue is  
almost always  handled through a UA exposed menu for user selection.  
Could MIME type be similarly handled for incorrect authoritative MIME  
types? That would might also be a way to address the issue raised by  
Sam of document subtypes, where the document is XML, but what kinds  
of XML does it contain and should those kinds be treated in a  
specialized way or as a generic tree presentation.

Take care,
Received on Sunday, 19 August 2007 23:45:39 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 29 October 2015 10:15:25 UTC