minutes from today part 1.

Very rough minutes indeed...

Present:
Charles McCathieNevile
Wendy Chisholm
Lisa Seeman
Dick Brown
Loretta Guarino
Donovan Hipke
Anuska Perkins
Matt May
Jason White
Len Kasday
Marshall
Cynthia Shelly
William
(check through - may be others)
--User Agent Capabilities

JW We need to distinguish between how they will work in guidelines and how
they will work in techniques. Charles put forward a few proposals. I think
Cynthis supported the option that we should make pragmatic decisions based
on each technology or feature

LK Symptom of trying to mix accessibility with acceptability. Availability of
browsers is one of a number of factors. It would be helpful to have a section
that said what constraints exist and tradeoffs

William Loughborough joins

LK for a given browser we can quickly decide what is accessible. Gets into
whether it is permissible to assume a particular browser...

LS There is a question of reasonable that gets hard to determine

Cynthia Shelly joins

WL Somebody writing and implementing policy isn't our problem (nor are we
judges of "undue burden, ...")

JW Charles suggested we set out techniques for determining when some
assumption is reasonable. Policy is important here, so the question of what
is reasonable/acceptable will enter into the discussion, and may be drawn
diffferently in a legal group than by this group.

LS Suggestion - have drier, more thorough, with example of how to apply the
guidelines on what we consider reasonable assumptions...

CMN We need to say which problems are technology dependent and which are just
impossible. We need to document the assumptions that we have made about what
software people do have, and what they can use.

LK We have two different sets of requirements in state of Pennsylvania. For
some circumstances we have to work on pretty much anything - can't assume
Operating systems or anything else. In other cases we have the chance to buy
software for everyone, so we know that they have a higher level of
software capability.

CS I think it is important to make our assumptions explicit - even if they
are not perfect we need to know what they are and people who read them need
to know what they are. When I was trying to implement this I needed to know
what the assumptions were. Difference between implementing an intranet /
internet site allows different assumptions. We might want to build a series
of intranet documents...

WL this is not relevant for first 4 guidelines. This is about graceful
transformation, device independence

DB There are questions like whether you have to support browsers without
scripts

WC Requirements document means we need to document our assumptions
(reads the requirements) Where do we begin?

CMN propose that we gather techniques for meeting requirements of
checkpoints, and that we add to them the browser reasons why. Then we have
enough data to interpret, like we are interpreting data to generate
checkpoints.

LK That leaves a lot of interpretation to be done

CMN Yes, but if we have better data we might be in a position to do that

JW We have extracted a lot out to techniques, and we couldprovide more data
there. Charles wasn't suggesting we go further and decide what reasonable
assumptions people can make about what browsers are out there - that is
beyond what we can do

CS Issue raised about thinking about what browsers being supported might not
be so hard in practise.

WL We have to get to the decisions of what lines we are going to draw - will
we enable text-only browsers or not?

DB What William said.

CMN We probably do have to make those decisions, but I don't know that we
have enough information collected to do it.

WC This is where we were with the TRACE guidelines - getting people to go
find out what worked and didn't, why and why not, which is why I wanted
testing. This is also the type of stuff we are going to need when we have to
show implementation. It is a lot of work and needs to be coordinateed. One of
the first things I am tackling is tabindex.

LS What happened to the database of techniques?

WC That is important - outstanding action item.

Action WC - chase up techniques database.

Wendy leaves the call

WL Checkpoint that talks about tabindex. Underlying assumption is spotty
because the idealised uses don't exist.

JW A technique that discusses accessibiltiy of scripts must document what is
needed and why.

Action WC ensure that information is provided for each technique on
technology assumptions

LS And the problems that people have

CMN I am assuming that is part of the same requirement (JW concurs)

[some discussion of how important it is that we make the decision about what
technologies people use]

WL Ultimately we have to draw some line.

JW There are some very minimal conditions. (Unfortunately VERY minimal).
After that it gets into proportions of people who have things, and we need
to be hesitant about drawing too many lines.

WL The minimal line moves.

JW HTML 2.0 is pretty well supported. after that?

CMN HTML 3.2 implements tables which have real problems in real, currently
used browsers.

JW In XML things tend to work better...

JW is there a way of deciding on minimum levels of support?

DB I am not sure that we should have the bar set as "practically universal".

LG Worried that keeping the bar too low removes the incentive for providing
decent support

JW Everyone should be encouraged to upgrade

CMN When I was pushing us to figure out how to draww the line now, I had a
prpoposal which said that there was a shortamount of time from when things
were available to when they could be assumed. I have backed away, but I think
it is important to us that people are using newer software and solutions

meeting closed...

-- 
Charles McCathieNevile    mailto:charles@w3.org    phone: +61 (0) 409 134 136
W3C Web Accessibility Initiative                      http://www.w3.org/WAI
Location: I-cubed, 110 Victoria Street, Carlton VIC 3053, Australia
September - November 2000:
W3C INRIA, 2004 Route des Lucioles, BP 93, 06902 Sophia Antipolis Cedex, France

Received on Thursday, 7 December 2000 17:17:31 UTC