Re: [wmlprogramming] Verizon, guidelines

Tom Hume wrote:
> Folks
>
> At the risk of reigniting old debates... I realised that we've not  
> heard anything from Verizon for a little while (since they launched  
> some grotty documentation for how to inhibit transcoding of services  
> via their network).
>
> A quick check this evening showed that they've updated their  
> documentation wrt opting out of transcoding. The page refers to "W3C  
> addressing models" which is still a bit iffy IMHO, but the PDF they  
> link to now refers to using domain names, MIME types, DOCTYPEs, LINK  
> tags or no-transform to inhibit transcoding. New version can be  
> downloaded from
>
> 	http://www.vzwdevelopers.com/aims/public/OptimizedViewOptout.jsp
>
> They're not fully compliant with the W3C doc, but given that it's not  
> finished I don't think we can hold that against them (yet). Definitely  
> closer to both Manifesto and CTG tho, so a step in the right direction.
>   
well, yes, you are reigniting. The Verizon document still reads:

"If, however, the mobile site requested does not have a URL that matches 
any of the mobile site naming conventions listed above, then the User 
Agent string passed to the web server by the Optimized View for Mobile 
Web server will be that of the server’s desktop PC browser so that 
Verizon Wireless will receive full HTML content for optimization."

which is not compliant to the Manifesto, and not even to the CTG. Of 
course, the "unless" clause in CTG makes it easier for Novarra and 
Verizon to justify the abuse, as compared to the Manifesto.
I find it particularly enlightining that Novarra is sitting at the CTG 
W3C table, while at the same time one very large customer of theirs is 
publishing for developers which break those very rules while referring 
at CTG. What this enlightens me about is that CTG is being used by 
Novarra as part of a corporate move to give justification to their 
irrepressible need to change the UA and trick websites into returning 
web content no matter whether those sites had a mobile version ready 
(and we all know that mobile sites are not born alone. It takes a large 
resource investment to create and maintain them).

Of course, the real problem is the fact that you do not need W3C to 
create and deploy abusive transcoders. You need W3C to wrap the abusive 
behavior in a shroud of legitimation. So, my point still remains: remove 
the "unless" clauses when it comes to UA spoofing from CTG. By this 
move, W3C would show the world that they can still police the worst 
misbehavers on the web (to some extent, at least).
Never condoning UA-spoofing would be consistent with what:
- the overwhelming majority of mobile developers want.
- the overwhelming majority of content owners want.
- people in W3C know to be right (and consistent with the rest of own 
BPWG: notably, static content is fully legitimate mobile content)
- many employees from transcoding companies know to be right when given 
a chance to express their opinions off the record.

Luca

Received on Friday, 19 December 2008 09:26:20 UTC