Re: ISSUE-261 (Scoping bogus 200 responses): Scoping 200 responses that should be regarded as 406 responses [Content Transformation Guidelines]

To help scope the support for displaying the response body of a 406 
response, I prepared a test:
http://www.w3.org/2008/06/rejected-mwbp-test/

or as a shortened URI:
http://tinyurl.com/4qgumz

Results are available here:
http://www.w3.org/2008/06/rejected-mwbp-test/results
(we need more tests, obviously!)

May I ask that you all take a short minute to enter the above URI and 
give it a try with your favorite mobile browser?

Note this links to ACTION-771 on Heiko who proposed that, on top of 
mentioning that the server should reply with a 406 instead of a 200, we 
could say that if it's not possible because the resulting user 
experience would be extremely poor (giving it a try with a browser that 
doesn't pass the test, I thought the test was wrong because I could not 
understand the error message...), then the server could reply with a 200 
and alert with a flag using the <meta> element.

I don't particularly like this idea, but then, it's probably better than 
sending a 200 without any clue that a 406 was intended...

Anyway, on your browsers, get ready, go! Thanks!

Francois.



Jo Rabin wrote:
> 
> fwiw, afaik, MSIE does not display the contents of 406 responses 
> (possibly because some people say impolite things about the choice of 
> User Agent in such responses) so I guess that it is common to put such 
> impolite messages in 200 responses so they do get displayed.
> 
> Of course it may be much more common for this to happen the other way 
> round - i.e. if you're not using IE you get a message back from the site 
> saying that they don't support you.
> 
> Either way, it would be interesting to know.
> 
> I wonder if, in advocating the proper use of 406 to say that the User 
> Agent can't be handled, we ought to reintroduce a note about non-display 
> of 406 pages.
> 
> Jo
> 
> On 13/06/2008 10:21, Mobile Web Best Practices Working Group Issue 
> Tracker wrote:
>> ISSUE-261 (Scoping bogus 200 responses): Scoping 200 responses that 
>> should be regarded as 406 responses [Content Transformation Guidelines]
>>
>> http://www.w3.org/2005/MWI/BPWG/Group/track/issues/
>>
>> Raised by: François Daoust
>> On product: Content Transformation Guidelines
>>
>> Raised as an issue so that we do not forget to address this during the 
>> F2F in Sophia.
>>
>> Being able to scope the number of web pages that actually return a 200 
>> OK status with a human message that says: "Sorry, I can't process your 
>> response" would be an invaluable measurement of the state of the Web 
>> today re.. the need to alter the HTTP headers of a request.
>>
>> Anyone with some stats to share?
>> (sharing may be done on a member-only basis)
>>
>>
>>
>>
> 
> 

Received on Friday, 13 June 2008 14:51:14 UTC