On 9/21/10 7:08 PM, Mike Hanson wrote:
Hi - just wanted to note that Mozilla Labs people are here and listening.
There are people in the labs group that are very interested in web
application deployment into the main browser context. We are not
particularly focused on "widgets" per se (as has been noted there are
many runtimes for those, including Mozilla-platform-derived ones), but
we have noted the trend of full-browser-window applications and are
thinking about what it means.
Great.
Claes' note of Sep 10 [1], summarized the questions that we're thinking
about nicely.
I would say that our current leaning is towards the HTML5 feature stack,
with AppCache, local storage/indexedDB, and CORS/postMessage playing
prominent roles. My current feeling is that widgets that are stored
"forever" are better suited for a widget runtime than a general browser.
Agree.
That suggests that packaging will play a smaller role on the browser UA
since we will largely be working with manifests. API and network access
permissioning at install time makes a lot of sense, as does integration
with UX elements that help users manage a multi-application environment
(pinned tabs, notifications, etc.). I have a personal interest in
identity systems and would also like to see whether we can simplify
application-identity management at this level.
Me too, I'm very interested in application-identity management. What is
the appropriate forum to talk about this?
Marcos - the Widget Landscape document [2] identifies the difference
between Widget user agents and general browsers - but would you
highlight any of the widget specs as solving as-yet-untackled problems
in the general-browser application space?
No, the Widget specs don't handle this - and to be honest, the packaging
solution (zip) might not be the right technology for the job as it does
not stream well (unless you send it backwards, and do all sorts of hacky
things). Our design goals, which are listed in the Requirements doc,
focused around ease of use and reuse of existing technologies (read:
focused around zip and xml... it was 2006, after all...:)).
But this is why we (Opera and others), in 2010, proposed to keep "Widget
Embedding" as a deliverable for the Web Apps working group. We knew that
we would need to deal with it eventually, but we didn't have any
concrete technical proposals to make.
[1] http://lists.w3.org/Archives/Public/public-device-apis/2010Sep/0049.html
[2] http://www.w3.org/TR/widgets-land/
-Michael
--
Michael Hanson, Mozilla Labs (@michaelrhanson)
On Sep 16, 2010, at 5:30 PM, Marcos Caceres wrote:
Hi Nathan,
On 9/16/10 7:38 PM, Nathan wrote:
Marcos Caceres wrote:
On 9/16/10 6:10 PM, Nathan wrote:
Marcos Caceres wrote:
As above. I thought that was what we (Web Apps WG - Widgets) have been
doing for the last 5 years?
Maybe I've missed part of the specifications - are you telling me
that I
can package up an HTML,CSS,JS based application as per the widgets
specification, include a WARP, Digital Signature, set the view-mode to
windowed and that this will run as is, in the main browser context of
the main browser vendors (Firefox, Safari, Opera, Chrome, IE etc)?
Ah! ok. I get it now. No, that won't work right now (actually, that's
how we run them in our development environment for testing purposes :)
). But that is trivial and no one has really asked for that.
<snip>
Everything is currently pointing at an exponential increase in 100%
client side applications, from all angles, we've got client side
persistence, html5, canvas, ecmascript, gpu acceleration, a vast "web of
data", cloud storage, positioning the web as the data tier, a plethora
of standardized and supported APIs + media types - every element.
Right. It's all looking pretty sweet... maybe we don't need them
widgets things after all...:)
The
missing bit to tie it all together is for somebody to simply say "here's
how you wrap it all up and deploy", and that work has been all but done
under the banner of the widgets specifications.
Opera supports W3C widgets: use that :) For Webkit and friends, they
are open source, maybe you can just hack the support and convince them
to make it a standard part of their platform. Other people have done
this (e.g., Widgeon runs on Mozilla's code, and there are lots of
WebKit implementations of widgets). You can also email Apple and
Mozilla and ask them to add support widgets if you think its
important. I'd be interested to hear what they say - I'm sure they
would be receptive to the idea.
--
Marcos Caceres
Opera Software
