- From: Ian Jacobs <ij@w3.org>
- Date: Fri, 18 Aug 2000 13:17:29 -0400
- To: w3c-wai-ua@w3.org
- CC: plh@w3.org, jbrewer@w3.org
Hello,
The User Agent Accessibility Guidelines 1.0 (18 August draft [1])
includes three checkpoints (5.1, 5.2, and 5.7) that refer normatively
to the DOM Level 2 specification. The latest version of that
specification today is the 10 May draft [2], a W3C Candidate
Recommendation.
As we near our return to last call, we must (re)consider how
to handle the fact that the two specifications are advancing close
to one another.
Background:
1) In general, for normative references to W3C specifications,
we should only refer to (stable) W3C Recommendations. Therefore,
if UAAG 1.0 becomes a Recommendation before DOM2, we should
not have a normative reference to DOM2 in UAAG 1.0.
2) We have already resolved in issue 277 [3] that if we
would have to wait for DOM 2 to become a Recommendation
in order for UAAG 1.0 to become a Recommendation, then
we would prefer to have DOM 1 requirements only.
Here are some scenarios to consider for moving forward:
1) We drop our requirements to DOM 1 and think about DOM 2
later, after both UAAG 1.0 and DOM 2 have had some experience
"in the market".
2) Suppose that UAAG 1.0 becomes a Recommendation before DOM 2 does.
We can lower our requirements to DOM 1 and include an
informative note that when the DOM 2 spec becomes
a Recommendation, we will republish a UAAG 1.1 Recommendation
(the number may not be UAAG 1.1, but let's use that to distinguish
it from UAAG 1.0). To facilitate this process, the Director has
suggested that the DOM 2 call for review to the Advisory Committee
might also say something like "We also request that you confirm
the publication of UAAG 1.1 with its DOM requirements increased
to DOM 2 requirements." In short, we could "piggyback" the DOM 2
Proposed Recommendation review and would not have to go through
any other reviews than that in order to be able to publish
UAAG 1.1 (unless, of course, the reviewing AC Members did not
approve for some reason). We should also inform the AC in the
(next) UAAG 1.0 Proposed Recommendation review that we intend
to do this.
However, there are two issues that this scenario implies:
a) If DOM 2 goes to Rec very soon after UAAG 1.0, then
UAAG 1.0 and UAAG 1.1 will be very close in date,
and that may confuse people. It's not worth publishing
UAAG 1.0 if we're going to publish the next one
a month later. What would be a suitable delay before
publishing UAAG 1.1?
b) If DOM 2 has just become a Recommendation, there won't
be many implementations of it. There will be a few,
because that is required in order for DOM 2 to exit
Candidate Recommendation status. But it means that
we are not likely to have any conforming user agents
until DOM 2 support is more widespread. DOM 1 became
a W3C Recommendation in October 1998.
One approach for addressing this is to say in UAAG 1.1
that the DOM 2 (now a Recommendation) requirements only
take effect in six months (or some other arbitrary delay).
3) Suppose that DOM 2 becomes a Recommendation before UAAG 1.0.
a) Issue b of the previous scenario applies.
b) If DOM 2 becomes a Recommendation shortly before UAAG 1.0,
then we need to be careful to say in the status section of the
specification at the spec goes through last call, candidate
rec, and proposed rec, what our intentions will be.
All of this means that we should decide how we want to proceed
(and what W3C process allows) before we go to last call to set
proper expectations for the future.
Other ideas about scenarios for moving forward welcome,
- Ian
[1] http://www.w3.org/WAI/UA/WD-UAAG10-20000818/
[2] http://www.w3.org/TR/2000/CR-DOM-Level-2-20000510
[3] http://server.rehab.uiuc.edu/ua-issues/issues-linear.html#277
--
Ian Jacobs (jacobs@w3.org) http://www.w3.org/People/Jacobs
Tel: +1 831 457-2842
Cell: +1 917 450-8783
Received on Friday, 18 August 2000 13:17:42 UTC