Re: the title and name tags for frames (migration)

The way user agents handle information  and present it to users is
important, and very often considered by designers.  Designers with
accessibility in mind think about how the screen reader presents the
information,  as well as the visual rendering, as well as the programmatic
execution, etc.  Sure the UAAG says ALL the information must eventually be
made available to the user, but the order of priority is important.

I recommend that we specify the "best" order in which to discover and
render information - for example on FRAME:
     1st render the title attribute,
     else then prefetch the title element of the page
     or else then the name attribute and/or URL of the page

> ...The title attribute is designed for a human-readable
>explanation of what an element does - the name attribute on a frame is a
>machine-readable one.

I agree.

>The situation is more complicated that this.  The 'name' attribute was
>there first, so it is not fair to say that implementations which display
>the 'name' attribute are not "good implementation."

I believe it is fair to say that is it "bad" if there is also a title=
attribute that was ignored.  If the title= was missing, then it should be
up to the user agent design to pick between pre-fetching the TITLE element
from the page - or - just giving some combination of name and/or URL to the
user.

Just as a visual browser loads all of the frame set, so did Home Page
Reader 2.5, and thus had available to it all the titles (if they were
there) of the pages that made up the frame set.  If the TITLE element was
missing, then the last resort was the URL of the FRAME.

>... it is important to work on current
>specifications where they are available, as HTML 4.0 has been for some
>years.

I agree.

Is this a PF or GL discussion thread??  Seems like it fits best in GL.

Regards,
Phill Jenkins
IBM Research Division - Accessibility Center

Received on Monday, 25 September 2000 13:27:32 UTC