- From: Doug Schepers <schepers@w3.org>
- Date: Wed, 31 Mar 2010 11:30:11 -0400
- To: XForms <public-forms@w3c.org>
- CC: public-hypertext-cg@w3.org, "www-dom@w3.org" <www-dom@w3.org>, Steven Pemberton <steven@w3.org>
Hi, folks- I discussed this issue with Steven Pemberton at our W3C Team Interaction Domain f2f, and he expressed serious concern that DOMActivate might be deprecated in DOM3 Events. One rationale he explained was backwards compatibility with XForms content; he was not satisfied with my response that XForms can continue to use 'DOMActivate', and that DOM3 Event even specified it more clearly (especially in relation to other events). I reasoned that deprecation (and possible future removal) from the DOM3 Events spec does not mean that XForms cannot use it; XForms can still reference DOM2 Events or DOM3 Events going forward. However, Steven felt that this would harm XForms, referencing an event type that is at risk of removal from the shared events defined in the DOM3 Events spec. I noted that DOM3 Events is only intended to define events that are shared between multiple technologies, and XForms is the only technology that relies on DOMActivate. One possible way forward which seemed to satisfy Steven was making DOMActivate optional without deprecating it. Currently, there are no normative, undeprecated, optional parts of the specification, and I am reluctant to introduce one on the grounds that it weakens the conformance criteria for the spec, i.e., the aim is to have all normative features of the spec implemented across UAs, so authors can count on the functionality. Retaining DOMActivate also sends a weaker sign to the content-author community about which event they should use for accessible content, click or DOMActivate. Though Steven and I have had some verbal discussions about the matter, I don't recall seeing any email response from the XForm group or community on the topic. I would appreciate if we could have some email discussion about this very soon, since we are planning to go to Last Call with the DOM3 Events spec as soon as possible, and I would like to have the matter resolved before then. Could someone from the Forms WG please chime in on the discussion? Regards- -Doug Schepers W3C Team Contact, SVG and WebApps WGs Doug Schepers wrote (on 2/1/10 3:38 AM): > Hi, XForms folks (& sundry interested parties)- > > I previously mentioned to you that the DOMActivate, DOMFocusIn, and > DOMFocusOut were being considered for deprecation in DOM3 Events. After > looking into the matter closely and talking with various invested > parties, I've now deprecated them in current drafts of the spec [1]. > I've explained some of the rationale for this in the HCG [2]. > > Note that this is deprecation, not removal... I took pains to fit them > into the more precise coherent model we've worked out for DOM3 Events. > Given XForms' use of DOMActivate in particular, I have put special > provision to explain its place in the event flow for activation > triggering and activation behavior [3]. > > Going forward, the XForms specs can continue to define DOMActivate [4], > DOMFocusIn [5], and DOMFocusOut [6] as they already do, and > implementations of XForms can maintain backwards compatibility with > existing content. > > Going forward, the XForms WG may wish to consider taking a fresh look at > events as specified in DOM3 Events, which is rather more detailed than > previous DOM events specs and tries to align more closely with actual > implemented behavior, to see if aspects of it might be useful to XForms. > > [1] http://dev.w3.org//2006/webapi/DOM-Level-3-Events/html/DOM3-Events.html > [2] > http://lists.w3.org/Archives/Public/public-hypertext-cg/2010JanMar/0001.html > > [3] > http://dev.w3.org//2006/webapi/DOM-Level-3-Events/html/DOM3-Events.html#event-flow-activation > > [4] http://www.w3.org/TR/xforms11/#evt-activate > [5] http://www.w3.org/TR/xforms11/#evt-DOMFocusIn > [6] http://www.w3.org/TR/xforms11/#evt-DOMFocusOut > > Regards- > -Doug Schepers > W3C Team Contact, SVG and WebApps WGs >
Received on Wednesday, 31 March 2010 15:30:26 UTC