RE: Followup: ACTION-541: MobileOK Scheme Document [was Summarize Jo's comment and my document and place into a W3C template]

Hi Kai

Thanks for figuring this out. For the benefit of everyone else, this
refers to the as yet non-existent MobileOK Scheme document.

I think we need to start from the top here in this discussion and go
back to basics. In particular, we need to be clear who is claiming what.
The self-claim use case is important. The third party claim and
certification bits are also important. But since I have not been
POWDERing for many a long month, I have no idea of what the expressive
features of the now-hopefully-nearly-agreed-scheme are.

I've added a number of other comments in line below.

I've got some other comments on your original which I will post under
separate cover.

I'm thinking that this stands every danger of becoming a complex beast
that no one other than a brave hearted few understand. It would be far,
far preferable for us to come up with some simple-to-use rules that
everyone can understand. However, I'm not sure that claims and
certification are open to a simplistic definition.

Jo


> -----Original Message-----
> From: public-bpwg-request@w3.org [mailto:public-bpwg-request@w3.org]
On
> Behalf Of Scheppe, Kai-Dietrich
> Sent: 15 February 2008 11:43
> To: BPWG-Public
> Subject: Followup: ACTION-541: Summarize Jo's comment and my document
and
> place into a W3C template
> 
> 
> I was asked to repost, what I had posted for ACTION-541.
> Unfortunately it was fitted with the original Action number and so was
> not associated to Action 541
> Also, it was not placed into a template, as discussion stopped.
> Basically the thread died.
> 
> 
> Original action:
> ACTION-532: to draft mobileOK usage rules and come back to the group
> 
> to be found
> http://lists.w3.org/Archives/Member/member-bpwg/2007Aug/0000.html
> 
> Jo's comments
> 
> http://lists.w3.org/Archives/Member/member-bpwg/2007Aug/0060.html
> 
> 
> My response
> 
> http://lists.w3.org/Archives/Member/member-bpwg/2007Aug/0065.html
> 
> 
> Here is the full text of my response, for the benefit of easy
reading...
> 
> Hi Jo,
> 
> I think most of what you had written is actually contained in what I
> propose.
> 
> > 1. Content that claims mobileOK at either level MUST indicate
> > that claim using the labelling mechanism that is to be
> > specified by the POWDER WG and which is to be elaborated by
> > the BPWG in a separate Recommendation.
> 
> Check
> 
> >
> > 2. The mobileOK logo is an optional additional human
> > perceivable visual notification that a claim is made in
> > respect of a given URI or group of URIs and that when the URI
> > is dereferenced using a suitable User Agent HTTP header,
> > content labelled using the mechanism identified in 1. will result.
> 
> Check

Right - the point I am trying to emphasize is that the logo is not a
claim. That it merely indicates to a human that a claim is present. This
obviously only works if the content is self certified by a Content
Provider.

In fact I don't think that what I wrote is right. As it doesn't allow
for a 3rd party claim, especially not for allowing a search engine to
make claims about content that is not labelled.

> 
> >
> > 3. Content in respect of which a mobileOK claim is made
> > SHOULD NOT include the mobileOK logo - as the logo is usually
> > extraneous to the purpose of the content and adds to the page
> > weight and latency of access to the content from mobile devices.
> >
> > (But how about decorating links to mobileOK content with a
> > teeny mobileOK logo or some other sign)
> 
> I am not sure if this is feasible, as it devalues the notification
that
> could be given if a single page is found via search engines.
> This will be a very important point to content authors and providers,
as
> it advertises the nature of mobileOK content.
> Furthermore this mechanism will help spread the knowledge of mobileOK
> content.

Well, from a search engine point of view, it is aware of a claim from
the reference to the claim contained in the retrieved content. It should
not use the presence of a visual indication that a claim is being made.
I'd think the Search Engine should then decorate or otherwise
distinguish the result to indicate the presence of a claim - or indeed
to make its own claim on behalf of the content.

Also I am wondering what is so very harmful about a mobileOK page
designed for mobile use containing a small logo, which is what this sets
out to prevent. However, there is a sort of chicken and egg in this in
that adding a mobileOK logo might alter the content in a way that makes
it not mobileOK any more. 
> 
> >
> > 4. Content from a URI from which a claim is made, when
> > accessed using a User Agent HTTP header other than one
> > identified in 2. MAY include a mobileOK logo as a sign that,
> > when accessed using appropriate (different) HTTP User Agent
> > Headers mobileOK content will result.
> 
> This is a bit convoluted and I am not sure how this would not be in
> conflict with your 3., which in turn makes me support 4. :-)
> 
The difference is that this one says that a page destined for desktop
usage says that a mobile version is available, whereas 3 says that it
_is_ the mobile version, and as originally written says that it mustn't
contain a logo. Which I now think is wrong.

> 
> >
> > 5. The mobileOK logo MAY be used to decorate URIs in printed
> > material and other visual media.
> 
> I have not listed this.

It might be useful from a licensing point of view to make it clear that
the logo may be used to draw attention to sites/pages outside the
context of the Web, especially when you put a URI on the side of a bus,
or in a newspaper or on TV or whatever. Think that would be an
especially important use.

> 
> >
> > 6. These usage rules do not apply to meta discussion of the
> > mobileOK logo, as long as it is clear that the logo is not be
> > used as a notification of the presence of a claim.
> 
> Which meta discussion are you referring to?
> 
This is a bit obscure, I agree. But if you have a Web page that says
"Hey check out this new scheme from W3C! When you see this logo, it
means that the page works well on a mobile". Now clearly the logo is
being used, but the page it is on may not itself be mobileOK, but the
intention is to allow the representation of the logo to tell people what
it means.

It also actually introduces some stuff in my mind at least about search
results that show that the things they reference are mobileOK may also
come into this category. And further, that a search result page that is
itself mobileOK may need to distinguish its own label from labelling its
referenced content.

Jo

Received on Friday, 15 February 2008 16:19:57 UTC