what's the status of this project

82 views
Skip to first unread message

Steve Heron

unread,
Oct 10, 2012, 4:38:05 AM10/10/12
to mdv-d...@googlegroups.com
Hi, I attended Alex's  talk about Web Components and MDV at the London Ajax group last night - really good presentation. I have a couple of questions about MDV - where are other browser vendors with this , are they actually behind it or is it currently a google-only initiative ? Also, this mailing list seems pretty dead, is the project still under active development ? I'm hoping the answer is yes.
 


Rafael Weinstein

unread,
Oct 10, 2012, 8:40:56 AM10/10/12
to mdv-d...@googlegroups.com, Alex Russell
Hi Steve,

I'm glad you attended Alex's talk and got excited about browser-based
templating and databinding. The project is still very much alive
despite the low traffic on this list. Note that the code at this
project represents an initial design proposal to the W3C, not a
library intended for use in production.

Here's our current status:

In summer of 2011, we proposed that the W3C Web Applications working
group take up this work and the consensus response was a preference to
break the problem down into smaller pieces, which is what we have
done.

The first piece (a) was DOM Mutation Observers (which replace the
long-deprecated Mutation Events). This has already being specified in
DOM4 and is currently shipping in both Chrome and Firefox.

https://developer.mozilla.org/en-US/docs/DOM/MutationObserver

The second piece (b) is Object.observe (which allows for observation
of raw data). This was recently accepted into ECMAScript and
implementations are still pending.

http://wiki.ecmascript.org/doku.php?id=strawman:observe

The third piece (c) is the HTML template element (which allows for the
declaration of inert document fragment "prototypes"). The
specification for this is currently being drafted.

http://dvcs.w3.org/hg/webcomponents/raw-file/tip/spec/templates/index.html

------
The above three pieces are the rudiments of the problem. The remaining
work is to add the notion of bindings and template
instantiation/iteration, and we are actively working on a design for
this.

To the question of whether the larger feature will be supported by
multiple vendors: I can't speak for other browsers, but I'm personally
hopeful that once the above three primitives are specified, accepted
and shipped, there will be a good support for "completing" the work of
having a full solution to templating and databinding. Note that (a)
was driven as a collaboration between Chromium and Firefox and (b) and
(c) have mainly been collaborations between Chromium and Microsoft.

Adding features to the web platform currently means that they will be
available to a huge population of developers and also that they will
effectively last forever. This is a point of extreme leverage -- which
is great, but it also means that the work of getting it done and
shipped can take longer than one might expect.

Steve Heron

unread,
Oct 10, 2012, 8:51:18 AM10/10/12
to mdv-d...@googlegroups.com
thanks for the comprehensive reply. Exciting stuff. Couldn't agree more re. getting these features into the browser itself - now that the javascript libraries have shown us what works for developers, let's get a common, fast, native implementation that everyone can standardize on.
Reply all
Reply to author
Forward
0 new messages