RE: Editorial issues with PR-mobile-bp-20061102

Hi Johannes

Thanks for your feedback on this document. I agree that the formatting of
the tests could be improved. As things stand the reason for some of them not
being introduced by the name of the best practice is that it is intended
that this should be inferred from the preceding test that does have a label,
or if there is only one BP then it refers to that. I did it this way to
avoid having a further level of sub-headings, however now that you point it
out I think it could be made more clear.

Since the document is now in its final phase it's probably best to keep your
comments for when a new version of the document is produced and see if it
can't be done better at that point.

Ref you point about the relationship between the tests outlined in the BP
document and those specified in the mobileOK document:

I think the main point is that the relationship is not direct.

The BP document gives some outline guidance as to how to test in a general
way given various different delivery contests. The mobileOK document tests
very specifically for the Default Delivery Context - and as such you could
expect it to be a significant elaboration of the general tests outlined in
the BP document.

A further public draft of MobileOK will be released very soon.

Thanks again for your feedback
Jo

> -----Original Message-----
> From: public-bpwg-comments-request@w3.org [mailto:public-bpwg-comments-
> request@w3.org] On Behalf Of Johannes Koch
> Sent: 03 November 2006 22:14
> To: public-bpwg-comments@w3.org
> Subject: Editorial issues with PR-mobile-bp-20061102
> 
> 
> Hi there,
> 
> here are some editorial improvements for PR-mobile-bp-20061102.
> 
> Sometimes several statements are grouped under one heading. In most
> cases the following tests are prefixed with the statement label. However
> some are missing. I inserted the ones that I think should be added.
> 
> 
> 5.2.6.3 What to test
> 
> <ins>LINK_TARGET_ID </ins>Human Test: Check for proper descriptions
> (e.g. no use of "Click here").
> 
> <ins>LINK_TARGET_FORMAT </ins>Machine Test: Check for links to non-HTML
> formats.
> 
> 
> BTW, HTML is not in a proper content format according to the default
> delivery context in section 3.7. Should this be XHTML (Basic) instead?
> 
> 
> 5.2.8.2 What to test
> 
> AUTO_REFRESH, REDIRECTION: the content attribute may also contain a
> number for the timeout, not just a URI. When used for a refresh, it is
> quite normal to have only a number in the content attribute.
> 
> 
> 5.3.2.3 What to test
> 
> PAGE_SIZE_LIMIT: Only mentions markup and images. What about CSS or
> script or other onload referenced resources?
> 
> 
> 5.3.3.3 What to test
> 
> Mentions a statement label "SCROLLING_LIMIT" which does not exists
> anywhere else.
> 
> 
> 5.4.4.2 What to test
> 
> <ins>TABLES_NESTED </ins>Machine Test: Check that there are no nested
> tables.
> 
> 
> 5.4.8.3 What to test
> 
> Mentions pixel/absolute values for font-size (CSS property), while the
> statement also mentions markup language attribute values. What about
> attributes that allow both numbers (pixles) or percentage values? What
> about width and height CSS properties?
> 
> 
> 5.5.1.3 What to test
> 
> AVOID_FREE_TEXT, DEFAULT_INPUT_MODE: What about type="password"? The
> type could also default to "text".
> 
> 
> 5.5.3.2 What to test
> 
> <ins>CONTROL_LABELLING </ins>Machine Test: Check if the label element is
> used in forms.
> 
> Should the label be explicitly associated with a control (label/@for =
> $control/@id)?
> 
> <ins>CONTROL_POSITION </ins>Human Test: Check whether the labels are
> properly positioned with regard to the controls.
> 
> 
> 
> Some of the tests differ from the ones listed in WD-mobileOK-20060712.
> But I think, as mobile-bp is the newer document, mobileOK will be
> updated accordingly, right?
> --
> Johannes Koch
> Spem in alium nunquam habui praeter in te, Deus Israel.
>                           (Thomas Tallis, 40-part motet)

Received on Monday, 13 November 2006 10:21:19 UTC