W3C home > Mailing lists > Public > w3c-wai-ua@w3.org > October to December 1998

Re: Rating UA guidelines

From: Charles McCathieNevile <charlesmccn@yahoo.com>
Date: Mon, 12 Oct 1998 18:44:20 -0700 (PDT)
Message-ID: <19981013014420.10156.rocketmail@send104.yahoomail.com>
To: Jon Gunderson <jongund@staff.uiuc.edu>
Cc: w3c-wai-ua@w3.org

Jon Gunderson <jongund@staff.uiuc.edu> wrote:

> JRG: I think we are currently using "impossible", "difficult" and
> to describe our three levels.  I like them better than must, should,
> for UA.

And I think:
must/should/may applies to how important it is to implement the
guideline, which corresponds to whether the result will be impossible,
difficult or easy to use.

I had written:
> >2. The 'compatibility with assistive technology' sounds pretty woolly
> >to me. Are we talking about stuff that we ought to be expecting from
> >the DOM being passed from the browser itself to the assistive
> >technology, or  are we asking everyone to implement Microsoft's
> >Accessibility, or some other standard already in existence? It seems
> >to me that we should be specifying a standard interface for functions
> >we are not requiring of the browser itself.
And Jon replied:
> We cannot specify any particular technique, but the new techniques
> (to be made public to the group this week) can recommend a range of
ways to
> implement a particular guideline.  I think we want to let deveolpers
> their own solutions to a given guideline, but we do want to make it
> in the guideline the funcaitonality that is needed.  I think the
> document is critical in giving direction to the implementation
process and
> coordinating efforts between mainstrem browsers and 3rd party
> technology.  I think it would be very difficult, if not impossible,
to say
> to a develper you need to do it just like this.
my concern is:
if the developer is producing both the browser and the assistive
technology which companies like microsoft are perfetctly capable of,
then it is fine for them to state that their suite of products
provides a complete solution. If we are aiming for an open forum,
where people can produce pieces of the solution, then there needs to
be a standard that says how those pieces fit together (is that an
API?). If we take Jon's approach then I suspect that we will be
pleasing the big players at the expense of the little players. Since
smoe of the user groups here are fairly small, they are likely to be
interested as little players.

I know this is getting pretty airy-fairy, but it seems that we need to
think about it. My hope is that we can say to the developers
"implement the DOM as an interface..." and it will work.

Charles McCathieNevile
Get your free @yahoo.com address at http://mail.yahoo.com
Received on Monday, 12 October 1998 21:41:54 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:49:21 UTC