Re: [Editorial] Proposed text to explain that checkpoints are interdependent

"Hansen, Eric" wrote:
> 
> As I think about this issue, I wonder if it is fair to expect readers to
> infer interdependencies based on reading order.

I think it is fair to a large extent because this is a "document"
that has a reading order. We should tell them where they *must not*
infer anything based on reading order. For example, the order
of the Guidelines is not significant. The order of the checkpoints
is significant (e.g., we order them thematically and sort by priority).

> It seems to me that our
> approach has been to try to make important contextual information (e.g.,
> qualifiers, caveats, etc.) explicit. 

Yes.

> I wonder if we need to identify which
> checkpoints have these large cross-cutting functions and call them out in
> the notes or otherwise. (It might even be noted that such checkpoints might
> have been implemented instead as labels on other checkpoints, but for
> didactic or other reasons, they were included as checkpoints.) 

> I am not
> suggesting any major revolution in the document, but rather consider
> identifying those checkpoints that have this characteristic of acting like
> an overlay on many other checkpoints. Is such feasible?

It may be, but unless it has some normative status, I don't think
we should do this. Do you have a specific proposal?

 - Ian

> -----Original Message-----
> From: Hansen, Eric
> To: 'Ian Jacobs'; w3c-wai-ua@w3.org
> Sent: 3/27/01 5:25 PM
> Subject: RE: [Editorial] Proposed text to explain that checkpoints are int
> erdependent
> 
> Seems reasonable. See notes below.
> 
> > -----Original Message-----
> > From: Ian Jacobs [mailto:ij@w3.org]
> > Sent: Tuesday, March 27, 2001 5:16 PM
> > To: w3c-wai-ua@w3.org
> > Subject: [Editorial] Proposed text to explain that checkpoints are
> > interdependent
> >
> >
> > Hello,
> >
> > One reviewer of the document has noted that the document is
> > not designed so that the checkpoints can be read entirely
> > out of context.
> >
> > I think this is a very good observation. And it's true:
> > the requirements of some checkpoints "apply" to others,
> > the result of an attempt to reduce redundancy in the
> > checkpoints.
> >
> > For instance, we don't start off each checkpoint by
> > saying "In a device-independent manner, allow the
> > user to configure ....". We say this once (in checkpoint 1.1)
> > and thereafter 1.1 "applies" to the other checkpoints.
> >
> > I propose that we add a sentence or two to the beginning
> > of section 2 to explain this. This text might be appropriate
> > after the description of the structure of a guideline. Or it
> > might be appropriate after the description of the structure
> > of a checkpoint. In any case, here's the text:
> >
> > <PROPOSED>
> >   Note: Many checkpoints in this document are "interdependent":
> >   they include requirements that have been "factored out"
> >   of other checkpoints in order to reduce the size of the
> >   document. For instance, checkpoint 1.1 requires that the
> >   user agent be fully operable through the keyboard.
> >   Consequently, checkpoint 4.1 does not need to
> >   state explicitly that all input device requirements of
> >   the checkpoint must be satisfied through the keyboard;
> >   it does not start "Allow global configuration and
> >   control through the keyboard..."
> >
> New:
> First-time readers are likely to benefit from paying attention to
> contextual
> information such as guideline prose, guideline organization, checkpoint
> order, checkpoint notes, and associated techniques. The checklist [??]
> consisting of checkpoints themselves is expected to be a useful tool for
> people with some familiarity with the document to evaluate user agent
> conformance.
> 
> Old:
> >   As a result of this factorization, first-time
> >   readers are likely to find the checkpoints easier to
> >   understand in context than "on their own." For instance,
> >   guideline prose, guideline organization, checkpoint order,
> >   checkpoint notes, and associated techniques all provide
> >   context for understanding the purpose and scope of
> >   each checkpoint. The checklist is expected to be a useful
> >   tool for people with some familiarity with the document
> >   to evaluate user agents for conformance.
> > </PROPOSED>
> >
> >  - Ian
> >
> > Reference document: 23 March 2001 draft:
> >   http://www.w3.org/WAI/UA/WD-UAAG10-20010323/
> >
> > --
> > Ian Jacobs (jacobs@w3.org)   http://www.w3.org/People/Jacobs
> > Tel:                         +1 831 457-2842
> > Cell:                        +1 917 450-8783
> >

-- 
Ian Jacobs (jacobs@w3.org)   http://www.w3.org/People/Jacobs
Tel:                         +1 831 457-2842
Cell:                        +1 917 450-8783

Received on Wednesday, 28 March 2001 11:09:48 UTC