Re: “Closing the gap with native” headlight project

Hi Jonas,

Thanks again for your intro; comments and questions in-line.

Le jeudi 14 février 2013 à 20:13 -0800, Jonas Sicking a écrit :
> Some areas that I think are in great need are:
> 
> * Offline support

I wholeheartedly agree that our offer in this space is currently rather
poor; my understanding is that there is solid work proposed to replace
AppCache that should hopefully land in WebApps soon.

The other issue that many developers face is how to deal with the
limitations in storage space; there is work on the Quota API that is
meant to address this issue.

With all that, do you believe that there are additional steps that need
to be taken in this space? And more specifically, are there additional
resources that W3C ought to invest in here, either to address additional
use cases, or to accelerate the work on these existing use case?

> * Improved security models. Right now all web content is treated as
> "probably written by a malicious author" which limits what we are able
> to do.

While I agree that the current security model of the Web can be
extremely constraining, it has also served the Web pretty well (I
believe it is one of the reasons the Web is so ubiquitous nowadays). But
clearly there is value in exploring what the Web can do in a more
flexible security model.

This is what (as you well know :) the SysApps Working Group is looking
at; do you see there again a need for additional resources one way or
another to help make progress on this?

> * Compensation. Many developers are much more interested in writing
> software when they can get paid for it.

Not unsurprisingly :) Also, letting developers be directly paid make
some use cases less reliant on advertising, which some users will
certainly see as a benefit.

There is a separate headlight task force dedicated to Payment (see
http://www.w3.org/wiki/index.php?title=Payments_Task_Force ) which will
look at the monetary aspects of this, but handling the payment is only a
piece of what would need to happen in this space.

Quite a few people last week in Mobile World Congress asked about the
equivalent of application stores for Web Apps; I know that Firefox
Marketplace is one of the answers that Mozilla is pushing in this space:
http://www.mozilla.org/en-US/apps/

Now, from my early exploration in this space, "it's complicated™". I'll
start a separate thread on this since I think it's likely a domain where
this task force ought to look deeper into.

> * Hardware integration. Getting access to sensors and ports available
> on the device.
> * User-data integration. Getting access to pictures, music, contacts,
> calendars, etc, that the user has stored on the device.

A big part of the agenda of the SysApps and Device APIs Working Groups
are dedicated to these (and lots of it come from work pushed by
Mozilla); I'm curious here again if you see the need for new work,
accelerated work, etc.

> * Network performance. Websites end up both creating redundant
> connections as well as download redundant data currently.

There were some new discussions on this topic in DAP recently, with
Tobie proposing for instance that offering developers a way to schedule
large background download of data (which browsers could optimize based
on the network environment) would help, as well as other ideas:
http://lists.w3.org/Archives/Public/public-device-apis/2013Feb/0058.html

This certainly sounds worth further exploration, and it also seems to be
a topic on which mobile operators are (unsurprisingly) very interested
in.

There was Community Group set up to look at this space, but it hasn't
gained much traction: http://www.w3.org/community/networkfriendly/

I'm interested in hearing thoughts on how we could increase the pace of
this work; I'm idly wondering in particular if a workshop would help
getting people on the same page.

Dom

Received on Thursday, 7 March 2013 09:38:51 UTC