Re: [wmlprogramming] Verizon, guidelines

Thanks Francois.

What I am seeing at the moment is that CTG's current draft has the 
following underlying message for the mobile ecosystem:

"transcoders should not reformat mobile content, but if they do, the 
onus of protecting their own content is on content owners with 
cache-control:no-transform".

I would be very interested to know if this changes, because it's the 
core of the disagreement between CTG and the Manifesto (which places the 
responsibility of recognizing and preserving mobile content totally on 
transcoders).

While I'm here, please allow me to observe that, even assuming that 
no-transform is respected by all transcoders, it is not a very viable 
(nor desirable) way to protect content from transcoding for the 
following reasons: even complex application will still refer to a large 
amount of static "resources" (CSS, Scripting, (X)HTML to be injected by 
javascript into an existing page, pictures and so on) to work properly.
Requesting that each and every of those resources is served by a program 
that adds no-transform (as opposed to simply being served by the web 
server directly off a file-system) is impractical and, I am sure, 
impossible in many circumstances. Among other things, it clashes with 
the idea that anyone can serve mobile content off any kind of hosting, 
no matter how cheap (which includes only being able to publish static 
content).

Thank you

Luca


Francois Daoust wrote:
> Hi Luca,
>
> I apologize for the shortness of my first message and the lack of 
> responses afterwards.
> I was off and actually managed not to have a look at my emails up 
> until today...
>
>
> Luca Passani wrote:
>>>>>> Very well. Can someone from W3C comment? can I safely state that 
>>>>>> W3C thinks that developers should had no-transform to their 
>>>>>> mobile applications if they do not want their content transcoded 
>>>>>> under any circumstances?
>>>
>>> Anyone? I think that Rigo's advice covers this myself, but can 
>>> anyone else clarify?
>> Yes. Come on, W3C. You have been discussing around this for one year 
>> now. Does the Stetement above reflect CTG or not?
>>
>> OK, I will fix it for you. 24 hours from now, if no evidence for the 
>> contrary has been provided, I will assume that the above statement is 
>> true and publicly state that, according to W3C, if mobile developers 
>> want to protect their application from transcoders, they should 
>> modify their applications and start using no-transform.
>
> Technically speaking, there is no "W3C position" yet since the 
> document is still a draft. There are personal opinions and an 
> intermediary published position that represents a consensus in the 
> task force at a given time. This intermediary position is not final.
>
> The only official position I can think of is in the HTTP RFC 2616:
>  http://www.ietf.org/rfc/rfc2616.txt
> ... section "14.9.5 No-Transform directive".
>
> It does state that "Cache-Control: no-transform" is the standard way 
> to prevent content modifications under any circumstances.
> As of today, the directive is not respected by all the existing 
> content transformation proxies, and is unfortunately (incorrectly, 
> some would say) respected by some gateways, preventing WML to WMLC 
> conversion and/or other optimizations that are useful on mobile 
> networks. It is still the existing standard mechanism to prevent 
> content transformation in any case.
>
> We cannot be more prescriptive than the HTTP RFC, but we may precise 
> good practices for some specific circumstances. In particular, I do 
> not like the idea that this directive becomes the de facto required 
> directive for mobile content. Mandating the respect of some heuristics 
> (typically the use of e.g. XHTML MP, XHTML Basic, WML doctypes) is 
> still under discussion. I personally think it should be part of the 
> guidelines, in the absence of any better way for content providers to 
> express their position regarding content transformation.
>
> Francois.
>
>

Received on Monday, 5 January 2009 14:53:02 UTC