RE: Comments to UA Guidelines

I want to add the "voice browsing" view on point of regard:

With voice browsing (as in webspeak, our phone browser, and I believe IBM's
home page reader), point of regard is the thing you are listening to at any
instant.  Navigation keys take you to the previous or next element (point of
regard).  The user can request, for the current element, additional detail,
which is a spoken as a message, but does not change the current point of
regard. The user can also view the current element in finer detail (by word
or spell mode).

If the user is reading a full page (continuous read), webspeak tracks the
currently spoken element, and if the user selects a stop function, the user
is positioned at the element last read.

What constitutes an element is determined by our style sheet. In general,
block level elements (P, DIV, Hn, ...) are navigable elements, but this is
configurable.

Mark
> -----Original Message-----
> From: w3c-wai-ua-request@w3.org [mailto:w3c-wai-ua-request@w3.org]On
> Behalf Of Marja-Riitta Koivunen
> Sent: Wednesday, November 11, 1998 6:16 AM
> To: Charles McCathieNevile
> Cc: w3c-wai-ua@w3.org; ij@w3.org
> Subject: Re: Comments to UA Guidelines
>
>
> Some more comments!
>
>   Marja
>
> At 11:00 PM 11/11/98 +1100, you wrote:
> >Comment interpersed - look for CMcCN: and MRK:
> >
> >On Wed, 11 Nov 1998, Marja-Riitta Koivunen wrote:
> >> page 6
> >>
> >> ... user agents should avoid displacing the viewport away from
> the user's
> >> point of regard as this can disorient users.
> >>
> >> !!Don't understand. Isn't the viewport or several viewports
> always there?
> >CMcCN:
> >The point is that if the viewport is going to be suddebly moved (for
> >example by following a link) the user should be notified. I think that
> >the treatment of this is not quite sufficient, since moving the viewport
> >by following a link is a natural consequence, and should be expected.
> >Moving it becuase of a pop-up window, auto-refresh which redirects after
> >some arbitrary time, and such-like are problems.
> >
> MRK:
> For me it is still difficult to understand the wording. Does this
> mean that
> we should avoid displacing the content of a viewport or automatically
> generating multiple viewports without a warning when it is not
> initiated by
> the user (normally with activities in the users' point of regard or user
> defined settings in the user agent)?
>
> >MRK:
> >>
> >> For paper it is difficult to indentify the point of regard
> more precisely
> >> than the entire page. ...
> >>
> >> !!What about the insertion point? I thought that was point of
> regard also?
> >CMcCN:
> >The Insertion point may or may not be in the point of regard. Imagine
> >using MSWord - the insertion point can be placed somehwere, and
> then tyhe
> >user can scroll through teh document, so that the insertyion pointis not
> >in the current point of regard (where that is defined as the
> visible part
> >of the document).
> MRK:
> I thought the point of regard is still valid even though it might
> be out of
> sight. For instance if I select a sentence and scroll it away
> from sight I'm
> able to replace it with something else. However, in that case I
> really would
> like to return to the place where I can see the point of regard
> (and this is
> actually how MS Word does it).
>
> I guess the paper really means a piece of paper, not an
> electronic document
> paper as I first thought. But is that important as we deal with electronic
> documents?
>
>

Received on Wednesday, 11 November 1998 14:06:16 UTC