W3C home > Mailing lists > Public > public-bpwg@w3.org > September 2008

Re: ACTION-837 - Provide explanatory text for the addendum... ISSUE-272 a new name ISSUE-273 for which document? ISSUE-274 which texts are needed?

From: Jo Rabin <jrabin@mtld.mobi>
Date: Fri, 12 Sep 2008 09:39:16 +0100
Message-ID: <48CA2AB4.8070705@mtld.mobi>
To: "Scheppe, Kai-Dietrich" <k.scheppe@telekom.de>
CC: MWI BPWG Public <public-bpwg@w3.org>

Hi Kai

I think we are talking about different things. In my original mail I was 
making the point that that BP document defines some loose "what to test" 
notions that do not relate to the DDC. mobileOK Basic Tests defines 
machine tests only for the DDC - mobileOK Pro was to define human tests 
only for the DDC.

Now we have no more Pro we need to re-evaluate whether having an 
addendum that talks only about human tests for the DDC makes sense any 
more. In my view the addendum would make more sense if it supplemented 
BP 1.0 in a more general way, by presenting explanatory text on the BPs 
where necessary and by presenting better "what to test" material.

This hopefully has the effect of repurposing all the substantive 
material in the document today as well as allowing greater flexibility 
on BPs we might wish to add comment on, in the light of experience. I 
agree that this should not be in stead of a new revision of BP, but I 
don't think the level of comments we might want to make necessitates that.

I thought this was what we agreed in Sophia, but clearly this is not a 
universally shared view ...

Jo




On 11/09/2008 16:15, Scheppe, Kai-Dietrich wrote:
> Hi Jo,
> 
> Thanks for your input...
> 
>> a) mobileOK Basic Tests are both machine testable and also 
>> intended only for the DDC (though, that said (SM), there are 
>> tests that are universally applicable too - but we make no 
>> distinction in mobileOK Basic Tests between the two)
> 
> I am not quite sure which point you are making.
> Only that the "Mobile Web Best Practices 1.0 - Addendum: Tests requiring
> human interaction" is not machine testable, but does require a DDC.
>  
>> b) Each best practice has an associated "what to test" that 
>> is not limited to the DDC
> 
> The entire document has been set up a based on the Best Practices and
> the resulting Basic Tests.
> The assumption is that the same rules apply, i.e. the UA is the DDC.
> 
> Perhaps this needs to be explicitly stated.
> 
> 
>  
>> c) Much water has flowed under the bridge since the text of 
>> the BPs was carved into stone (at PR stage) [that's enough 
>> analogies and mixed metaphors for now, ed.] and there are 
>> some areas that require clarification, and some aspects of 
>> the tests in that document that require review and amplification.
>>
>> Therefore:
>>
>> We should take the opportunity to a) make the clarifications 
>> to the best practices themselves (some useful ones already in 
>> the addendum - plus some others such as the stuff on keeping 
>> page title short) and b) to clarify the tests in general 
>> whether or not they are machine oriented, and make sure that 
>> they have _general_ not just DDC applicability. While we are 
>> about it we might choose to also comment on which of the 
>> mobileOK Basic Tests are more generally applicable than the 
>> DDC and which might be parameterised to become so.
> 
> I would be very cautious here.  The main document is the BP document.
> If we have learned and need corrections then those should go into a new
> version of the BP document.
> 
> The Addendum specifically deals with human interactive tests and is, in
> my mind, not the correct place to put in potentially substantial changes
> to the main document. 
> 
> 
>> I think that there is a lot of good material in the document 
>> as it stands. I think also that repurposing it will require 
>> quite significant structural changes.
> 
> What do you propose and why?
> If the addendum is another test document, which it is, then I am not
> sure a structural change is warranted.
> If it is, by definition, we must reconsider the BP document as well.
> 
> The issue of subjectivity and inability to make binary decisions over
> the outcome of the test, has no bearing on its structure.
> 
> Beyond that I certainly have to change the references to mobileOK Pro
> once we have decided on a name.
> 
> 
>> I am willing to offer my time to make some of those changes 
>> if we all find the above approach agreeable. [however, have 
>> also unwisely committed to undoing the mess I made of Phil's 
>> effort on mobileOK Scheme, and also have the small matter of 
>> the CT Guidelines (or whatever they end up being called tomorrow].
> 
> Thank you for the offer, but let's discuss this to see if changes are
> needed.
> 
> 
> 
> -- Kai
Received on Friday, 12 September 2008 08:40:18 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:09:52 UTC