Re: PROPOSAL: User Agent Issue 190:

I agree with Mark - there is no need for the Mouse events in DOM2. (with the
possible exception of mousemove.) There is a clear proposal for what to do to
match the legacy event triggers in HTML 4 with the new DOM events. If we say
to developers "use that approach in DOM2" then we are encouraging them to get
it right. If we say "go for it - put in the activate methods but don't expect
anyone else to take them seriously either" then we are encouraging them to
propogate things that do not work, and we then have to explain to them how to
fix two lots of problems (as well as why there isn't a key events set but why
keyboard access is still critical).

Changing the names is neither here nor there. And I don't think anyone has
expressed dissatisfaction with the propogation or the mutation events (other
than the problems raised by Gregory, which properly belong in UA and WCAG,
about magically submitting things). Having a device-dependent set of
prefabricated events in the DOM is just a little like giving matches to
children with a history of lighting fires, along with the lecture on why not
to do it. Or like putting a big warning that Smoking Kills You on every
cigarette packet - for a tobacco company.

Charles McCN

On Tue, 15 Feb 2000 schwer@us.ibm.com wrote:
  
  >am I understanding you correctly in that the fact that there isn't any
  >difference with HTML, that makes DOM2 events OK?    DOM2 device
  dependencies
  >are not OK in my mind.  the i18n group made the right choice to pull
  >the keyboard events and I wish they'd removed the mouse events
  >from DOM 2.
  
  No. What I am saying is that the device dependecy exists in the HTML DOM.
  For example, HTML DOM contains mouse events. We all agree that device
  dependency is bad. What DOM2 does give you is a standard event model for
  developers to write to even though the event names may change. Changing
  event names is less of an impact than writing the entire event model from
  scratch and I would like to get people writing from a standard event model
  for:
  
  - event propagation
  - document mutation  (the mutation events are good)
  
  The problem is that if we say to developers "use DOM 1" they will be using
  HTML DOM 1 plus the device depencent events anyways because they have to
  (They still have to support things like JavaScript). What they will not be
  doing is supporting standardized mutation events and DOM event  propagation
  in DOM 2. These elements of DOM 2 are not likely to change in DOM 3 even
  with our changes for device independence.
  
  DOM 2 will also provide for:
  
  - DOM CSS and DOM Style Sheets which can help assistive technologies such
  as low vision and reading disorders
  - Iterators which will reduce increase performance when AT's need to access
  elements in the DOM
  
  Rich
  
  Rich Schwerdtfeger
  Lead Architect, IBM Special Needs Systems
  EMail/web: schwer@us.ibm.com http://www.austin.ibm.com/sns/rich.htm
  
  "Two roads diverged in a wood, and I -
  I took the one less traveled by, and that has made all the difference.",
  Frost
  
  
  menovak@facstaff.wisc.edu (mark novak) on 02/15/2000 02:36:35 PM
  
  To:   Richard Schwerdtfeger/Austin/IBM@IBMUS
  cc:   Jon Gunderson <jongund@ux1.cso.uiuc.edu>, "Gregory J. Rosmaita"
        <unagi69@concentric.net>, Charles McCathieNevile <charles@w3.org>,
        User Agent Guidelines Emailing List <w3c-wai-ua@w3.org>, WAI
        Protocols & Formats WG <w3c-wai-pf@w3.org>
  Subject:  Re: PROPOSAL: User Agent Issue 190: Reduce the scope of 5.1 to
                     say "write access only for that which you can do through
        the UI."
  
  
  
  
  hi rich
  
  At 12:47 PM 2/15/00, <schwer@us.ibm.com> wrote:
  >Mark,
  >
  >There are a number of features in DOM 2 that make life easier for ATs. One
  >is iterators and a core event model which does provide standard rules for
  >event propagation. DOM 1 is very weak. While I appreciate some of your
  >conerns regarding the device dependencies in DOM 2 this is not any
  >different from the HTML events provided for in HTML DOM 1.
  
  am I understanding you correctly in that the fact that there isn't any
  difference with HTML, that makes DOM2 events OK?    DOM2 device
  dependencies
  are not OK in my mind.  the i18n group made the right choice to pull
  the keyboard events and I wish they'd removed the mouse events
  from DOM 2.
  
  
  >DOM 2 also
  >provides for mutation events which tell the AT when the DOM data changes
  >which helps with caching. DOM 1 is totally dead in this area. AT's survive
  >out of the kindness of Microsoft's heart in that they provide these types
  >of events even though the are not in core DOM 1.
  
  I'm assuming DOM 3 will provide somekind of feedback here as well.
  Mutation
  events in this case, are nothing more than the data model informing the
  view/controller of changes.  While MS may have done this via the
  "kindness of Microsoft's heart" as you say, it doesn't hurt that mutation
  events
  like "download begin, download complete, etc." exist for the ability of
  scripting, DHTML and COM/CORBA/XPCOM to take advantage of, especially
  for an interface built on the fly.
  
  I still maintain that DOM 2 is not something the UA group should push
  too strongly, and I'm hopefull that DOM 3 will correct some of these
  problems, as well as not be to far off.
  
  mark
  
  
  
  
  >Rich
  >
  >
  >Rich Schwerdtfeger
  >Lead Architect, IBM Special Needs Systems
  >EMail/web: schwer@us.ibm.com http://www.austin.ibm.com/sns/rich.htm
  >
  >"Two roads diverged in a wood, and I -
  >I took the one less traveled by, and that has made all the difference.",
  >Frost
  >
  >
  >menovak@facstaff.wisc.edu (mark novak) on 02/15/2000 10:37:18 AM
  >
  >To:   Richard Schwerdtfeger/Austin/IBM@IBMUS
  >cc:   Jon Gunderson <jongund@ux1.cso.uiuc.edu>, "Gregory J. Rosmaita"
  >      <unagi69@concentric.net>, Charles McCathieNevile <charles@w3.org>,
  >      User Agent Guidelines Emailing List <w3c-wai-ua@w3.org>, WAI
  >      Protocols & Formats WG <w3c-wai-pf@w3.org>
  >Subject:  Re: PROPOSAL: User Agent Issue 190: Reduce the scope of 5.1 to
  >               say  "write access only for that which you can do through
  >      the UI."
  >
  >
  >
  >
  >hi all
  >
  >At 9:53 AM 2/14/00, <schwer@us.ibm.com> wrote:
  >>This is why we were pushing the DOM2 event model as P2.
  >
  >in my opinion, the UA group should leave the guildlines as
  >they are, referring only to DOM or DOM 1, as it may be.  Then
  >when the later releases of DOM are completed, the Tech DOC
  >can refer to them, as it undoubtedly will.
  >
  >to refer to the current piece-meal event model in DOM 2
  >is not only a bad idea, it is misleading to developers since
  >the model isn't complete.  if makes no sense to me to
  >rally support in the AT community for a spec that is as
  >poor as DOM 2 event model currently stands.
  >
  >
  >>It is unrealistic to expect the DOM WG to scrap their entire event model
  >>for accessibility. We should be able to improve upon it in terms of
  device
  >>independence. Having people start developing to the DOM 2 event model
  will
  >>not require them to rewrite the whole thing.
  >
  >No one on this list, as I've read, has suggested any such thing.  The
  >i18n group put a serious issue to the DOM  group, much as the PF
  >group did, and the event model that remained in DOM 2 is not very
  >robust.
  >
  >If the event model were removed from DOM, as I've suggested, then
  >having people follow the incomplete DOM 2 model may require
  >rewrites.   I'm hoping that those decisions are still being considered
  >and thus why the UA should not act on DOM 2/3 at this time.
  >
  >>I do appreciate your concerns.
  >
  >thanks, but continuing to push DOM 2/3 with the UA is not helping, in my
  >opinion.
  >
  >mark
  >
  >
  >>Rich
  >>
  >>
  >>Rich Schwerdtfeger
  >>Lead Architect, IBM Special Needs Systems
  >>EMail/web: schwer@us.ibm.com http://www.austin.ibm.com/sns/rich.htm
  >>
  >>"Two roads diverged in a wood, and I -
  >>I took the one less traveled by, and that has made all the difference.",
  >>Frost
  >>
  >>
  >>menovak@facstaff.wisc.edu (mark novak) on 02/11/2000 08:49:49 AM
  >>
  >>To:   Richard Schwerdtfeger/Austin/IBM@IBMUS
  >>cc:   Jon Gunderson <jongund@ux1.cso.uiuc.edu>, "Gregory J. Rosmaita"
  >>      <unagi69@concentric.net>, Charles McCathieNevile <charles@w3.org>,
  >>      User Agent Guidelines Emailing List <w3c-wai-ua@w3.org>, WAI
  >>      Protocols & Formats WG <w3c-wai-pf@w3.org>
  >>Subject:  Re: PROPOSAL: User Agent Issue 190: Reduce the scope of 5.1 to
  >>               say "write access only for that which you can do through
  >the
  >>      UI."
  >>
  >>
  >>
  >>
  >>hi
  >>
  >>At 5:26 AM 2/11/00, <schwer@us.ibm.com> wrote:
  >>>The DOM event model is an ongoing discussion in the WAI/PF. We are going
  >>to
  >>>submit requirements to the DOM working group regarding the event model
  >for
  >>>DOM 3 in March.
  >>
  >>
  >>can somone point me to where this *discussion* is taking place within
  >>WAI/PF???   I would very much like to be apart of this...
  >>
  >>who are "we" when you say "We are going to submit requirements to the DOM
  >>working group"
  >>
  >>
  >>>The reason I put it at P2 was because DOM 2 is not out yet. I also share
  >>>some device independence issues regarding the DOM 2 event model.
  >>>
  >>>If some developers could start adopting the DOM 2 event model, it should
  >>>reduce the time it would take to get the desired DOM 3 event model
  >>>implemented and also ATs would have something to go from.
  >>
  >>at present, I'd encourage developers to avoid the event model of DOM 2,
  >>assuming of course something more useable and robust appears for DOM 3,
  >>which
  >>would probably save the developers both development cost and grief!
  >>
  >>
  >>
  >>>We ought to be pushing for P1 with DOM 3 in the UA assuming the correct
  >>>changes are made between the DOM and PF working groups.
  >>
  >>it is unrealistic to push any priority over a document or spec. that
  isn't
  >>even yet written in my opinion.
  >>
  >>mark
  >>
  >>
  >>>menovak@facstaff.wisc.edu (mark novak) on 02/10/2000 10:15:30 AM
  >>>
  >>>To:   Jon Gunderson <jongund@ux1.cso.uiuc.edu>, "Gregory J. Rosmaita"
  >>>      <unagi69@concentric.net>, Richard Schwerdtfeger/Austin/IBM@IBMUS,
  >>>      Charles McCathieNevile <charles@w3.org>
  >>>cc:   User Agent Guidelines Emailing List <w3c-wai-ua@w3.org>, WAI
  >>>      Protocols & Formats WG <w3c-wai-pf@w3.org>
  >>>Subject:  Re: PROPOSAL: User Agent Issue 190: Reduce the scope of 5.1 to
  >>>      say   "write  access only for that which you can do through the
  >UI."
  >>>
  >>>
  >>>
  >>>
  >>>hi Jon and all
  >>>
  >>>I agree that all elements ought to understand and implement
  >>>the appropriate event model, so for the UA, I don't see my
  >>>next comment changing the UA process at the moment.
  >>>
  >>>However, I'm not in favor of making this a priority 1, since I'm not
  >>>a fan of the event model within DOM 2.  This is an on-going
  >>>(I hope ) discussion on the PF/DOM working group lists.
  >>>
  >>>After the events/event model are understood, this checkpoint
  >>>may need re-visiting.
  >>>
  >>>mark
  >>>
  >>>At 8:55 AM 2/10/00, Jon Gunderson wrote:
  >>>>It seems there is a consensus to merge 5.3 and 5.5 from my proposal
  into
  >>a
  >>>>single checkpoint of at least Priority 2 and maybe a priority 1 level.
  >>>The
  >>>>new checkpoint would require implementation of the event model
  specified
  >>>in
  >>>>the Candidate Recommendation of DOM2 for all elements.
  >>>>
  >>>>Jon
  
  
  
  
  

--
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
Postal: GPO Box 2476V, Melbourne 3001,  Australia

Received on Tuesday, 15 February 2000 22:11:18 UTC