W3C home > Mailing lists > Public > w3c-wai-ua@w3.org > April to June 2011

Re: The UA is the new OS: redefining UA?

From: Markku Hakkinen <mhakkinen@acm.org>
Date: Tue, 7 Jun 2011 15:00:59 -0400
Message-ID: <BANLkTinqCDeSE+mQ487VcPwQ+q1Ozcfsng@mail.gmail.com>
To: simon.harper@manchester.ac.uk
Cc: w3c-wai-ua@w3.org
Hi Simon,

I should have cited our past discussions on this topic....  thanks for
bringing it up.  We did have earlier discussions on this topic in 2009

Your suggested 3 criteria are good.  What I think is missing is the
notion of embedded UAs and specifically UAs that are really devices.
Your 3 may be able to cover this, but I wonder if there is need for

The techniques document [3] defines the Primary User Agent, but calls
it, parenthetically, the "traditional browser." Primary user agents
can be far different than a "traditional browser."  I am of the
opinion that if we aren't explicit about the ways in which UAs are
being used, we are potentially giving those implementors a pass.

I also moved away from using Web content in my text and stuck with Web
technologies.  My concern is whether today's new crop of Web
application developers think in terms of Web content or view their
constructions as application programs.  How many of this new breed of
applications have content in the "traditional" sense as compared to
application programs with dynamically generated UI's (including Canvas
based apps), with no UA chrome, and in some cases initiated from a
locally stored start page and not even served up via HTTP?

The availability of device specific Javascript APIs (such as
accelerometers and GPS) and extensions to input type file that allow
access to device features  (such as cameras and microphones), also
redefines, in my mind what Web content for UAs is.  Is the following
example a UA that should conform to UAAG?

Social Cam, a stand-alone mobile app written in HTML5 and Javascript,
that allows you to take photos with your mobile phone camera, perform
photo editing (such as cropping and red-eye removal), erase photos by
shaking the camera, add cartoon-like captions to the photo, and then
automatically upload and post the photo to your favorite social media
site.  What exactly is the UA in this case?  Is it only the Webkit
engine that is wrapped by a platform specific app container? Or is it
the complete application and its UI?

And another example:

Movie Kiosk is a vending machine that incorporates a keyboard and
touch screen display, allowing consumers to browse available movies
for rental, view trailers, select, and then pay for movies that are
delivered via DVD or wireless download to a mobile device.  The Movie
Kiosk application, which controls all aspects of the UI and kiosk
functionality, is written in HTML5 and Javascript. Is the entire
hardware device a user agent?

Are Amazon Kindles or BN Nooks user agents?  They render and allow
interaction with XML and HTML based content.  The next ePub standard
update incorporates HTML5 [4].   I would therefore conclude that eBook
devices that support ePub are user agents.

And has anyone looked at Chumby [5]?  It is a small touch screen
internet appliance.  Should it be in UAAG Conformance?

User Agents... they're everywhere, it seems.


[1] http://www.w3.org/2009/03/05-ua-minutes.html
[2] http://lists.w3.org/Archives/Public/w3c-wai-ua/2009JanMar/0068.html
(start of an extended thread)
[3] http://www.w3.org/WAI/UA/2011/ED-IMPLEMENTING-UAAG20-20110525/#intro-def-ua
[4] http://idpf.org/epub/30/spec/epub30-overview.html
[5] http://www.chumby.com/

On Tue, Jun 7, 2011 at 3:08 AM, Simon Harper
<simon.harper@manchester.ac.uk> wrote:
> Hi Mark,
> I like this defn - discussion - I had the same concerns re UA defns at:
> http://lists.w3.org/Archives/Public/w3c-wai-ua/2009OctDec/0012.html
> I wonder how my attempts affect yours or vice-versa?
> Cheers
> Si.
> =======================
> Simon Harper
> University of Manchester (UK)
> More: http://simon.harper.name/about/card/
> On 06/06/11 17:57, Markku Hakkinen wrote:
>> Is the UA becoming the new OS?  Some developments to consider:
>> "Windows 8 apps use the power of HTML5, tapping into the native
>> capabilities of Windows using standard JavaScript and HTML to deliver
>> new kinds of experiences." [1]
>> "Webian Shell is a full screen web browser for devices that don't need
>> a desktop... The idea of the Webian Shell project is to re-think your
>> computer's interface as something much simpler which treats web
>> applications as first class citizens and does away with all the
>> un-necessary clutter." [2]
>> "At the core of each Chromebook is the Chrome web browser. The web has
>> millions of applications and billions of users." [3]
>> Are there implications in this trend for UAAG?
>> According to Google the ChromeOS accessibility support is via
>> ChromeVox [4], a self-voicing browser extension for Chrome.
>> Is it time to revisit the definition of UA? I believe we need to be
>> more explicit about how the user agent is evolving beyond just a
>> traditional Web browser.
>> To make a start, I have taken the introductory text from the current
>> draft [5] and revised it:
>> "A user agent is any software that retrieves and presents Web content
>> for end users or is implemented using Web technologies. User agents
>> include Web browsers, media players, and plug-ins that help in
>> retrieving, rendering and interacting with Web content.  The family of
>> user agents also includes operating system shells, consumer
>> electronics with Web-widgets, and stand-alone applications or embedded
>> applications whose user interface is implemented as a combination of
>> Web technologies."
>> The definition of user agent [6] is also a candidate for revision:
>> "A user agent is any software that retrieves, renders and facilitates
>> end user interaction with Web content, or whose user interface is
>> implemented using Web technologies."
>> I throw this out for comment. Is this a needed revision?  Are there
>> other suggestions?
>> mark
>> [1]
>> http://www.microsoft.com/presspass/features/2011/jun11/06-01corporatenews.aspx
>> [2] http://webian.org/shell/
>> [3]
>> http://googleblog.blogspot.com/2011/05/new-kind-of-computer-chromebook.html
>> [4]
>> http://www.google.com/support/chromeos/bin/answer.py?hl=en&answer=177893
>> [5] http://www.w3.org/WAI/UA/2011/ED-UAAG20-20110525/#introduction
>> [6] http://www.w3.org/WAI/UA/2011/ED-UAAG20-20110525/#def-user-agent
Received on Tuesday, 7 June 2011 19:01:51 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:49:40 UTC