Re: test against Gmail

I'll consolidate several replies into one here:

- You wonder why anyone would think of mobile web and web clients as
the same. I don't think anyone is suggesting this and don't see why
this perception has persisted. The BPs and mobileOK are in fact
specifically about content tailored for mobile devices, and
specifically not about any other content. I think the BPs have almost
no bearing on how you choose to adapt content, or "one web" issues, as
a result (see previous post though).

- You could only serve mobileOK content to the mobileOK tester user
agent. I suppose you could do that to fake out the XHTML validator
too. What is the incentive to do so -- why invest time making a
mobileOK site and then not deploy it to the world?

- Your mobile site could consist of only a (mobileOK) error page, yes,
one that passes the tests. The problem with that scenario is not
mobileOK tests -- your error page may well be a great mobile error
page -- but with the fact that your site consists of one error page.
Some site! I don't see incentives for behavior like this. If you
believe  that developers will often look for the easy way out, then I
suggest that the easy way out here is to simply ignore mobileOK. Those
that adopt mobileOK are those that want to adopt mobileOK because it
adds value; we hope many developers feel this way.

- Of course, Google mobile web services have been tested and work
pretty well already, but nobody's perfect and there is always room for
improvement. Running even the preliminary checker reveals some
problems that should be fixed on, say, mobile Gmail. (And I'll try to
get them fixed.) If anything that demonstrates that even established
sites can benefit from mobileOK, and it has value not just for
new-to-mobile developers. I don't see a disconnect with Phil since he
was talking about something entirely different... maybe I
misunderstood which post you were referring to.


On 12/4/06, Luca Passani <luca.passani@openwave.com> wrote:
>
>
> I think there is a disconnect between what you and Sean says.
> I assume I am safe assuming that Google understands mobile well enough that
> they do not need the checker to create good mobile services.
> So, according to what you say, there should be no need for google to
> successfully run the checker on their site to make sure that their service
> is MobileOK. They have testes (and run!) the service extensively enough,
> that there is no extra value the chacker can bring.
> Yet, Sean is saying that he will push internally to make sure Gmail on
> mobile runs through the checker successfully.
>
> Where is the disconnect?
>
> Luca
>
> -----Original Message-----
> From: public-bpwg-request@w3.org [mailto:public-bpwg-request@w3.org] On
> Behalf Of Phil Archer
> Sent: Monday, December 04, 2006 12:14 PM
> To: public-bpwg@w3.org
> Subject: Re: test against Gmail
>
>
> Not so much a joke as a pointless exercise. Since the checker and the
> whole of mobileOK is about helping people make better resources for
> mobile, the only person they'd be fooling by doing this is themselves.
>
> However, Luca, you do expose the weakness in perceiving mobileOK Basic
> as purely machine testable. The tests can be done by machines, but a
> mobileOK trustmark will require at least some human interaction if it is
> to be certified - which is why mobileOK needs to be carried in some sort
> of label that can be indecently authenticated.
>
> Expect a charter for review on this topic a week today. Meanwhile keep
> your POWDER dry.
>
> Phil.

Received on Monday, 4 December 2006 18:24:17 UTC