W3C home > Mailing lists > Public > www-di@w3.org > February 2004

RE: Multi-channel content

From: Rhys Lewis <rhys.lewis@volantis.com>
Date: Wed, 18 Feb 2004 09:02:40 -0000
Message-ID: <D18551A6C7EA6241B960D4909EB75DF5CE1AF5@squid.volantis-uk>
To: "Kai Hendry" <hendry@cs.helsinki.fi>
Cc: <www-di@w3.org>

Many thanks for your mail and for opening up a discussion on this list.

As Rotan has pointed out, there is actually quite a lot of evidence that runs counter to both of your
propositions. The very existence of the W3C Device Independence Working Group shows that there is broad
industry support for the need for work to extend the capabilities of current web technologies to
support the huge variety of devices that can now be used on the web. In addition, you should be aware
that there is also customer demand for products that deliver content and applications over multiple
channels. Applications have been built that deliver content across multiple channels and multiple
modalities using techniques that are affordable for authors.

On the specifics of authoring content for multichannel use, DIWG has published a paper
(http://www.w3.org/TR/acdi/), also accessible via our public home page (http://www.w3.org/2001/di/),
relating to the challenges that authors face in attempting to provide content and applications across a
range of devices. DIWG's view is that current standards are not sufficient to support authors

On your comment about XHTML, while it is certainly possible to create XHTML documents that can be delivered to a range of XHTML compatible devices, this does not solve the problem of device independence. First, XHTML is not universally supported. XHTML documents are unusable on WML devices, for example. Not only are there literally millions of web capable, WML wireless handsets in daily use around the world, there are also millions of digital interactive television systems that are based on extensions of WML. And, by the way, there is a large community of HTML 3.2-based devices still being produced in the form of smart phones. 

Looking at XHTML alone also ignores the issues about how CSS is represented, if at all, on these devices. Where it is supported, there are subset issues. Some devices support CSS 1, others CSS 2, still others, CSS mobile profile. Some devices require stylesheets to be external, others require style definitions to be in line. And that is without even considering the wide variety of different detailed characteristics and bugs in various browsers. XHTML devices come in a huge range of sizes and physical capabilities, from small mobile handsets to PC's and workstations. Even if you could send the same
XHTML to them all, these differences in physical characteristics mean that the likelihood of a pleasing
user experience is diminished.

So even in the restricted world of XHTML, the variability is huge. The result is that the only way to guarantee that an XHTML document is usable on a wide variety of devices is to use only a tiny fraction of its capabilities, to make the pages very small and to avoid use of even basic features such as forms. XHTML forms are very device specific, by the way. You probably have to avoid any images or other rich media content as well, because of the variability of display size and support on different XHTML capable devices.

To me, that doesn't sound like a description of a 'well designed' XHTML page. It is also not very promising for delivery of applications, only the simplest, least formatted, fixed content.

Turning now to CC/PP, the provision of detailed information about devices to the servers that create the appropriate adapted content for use on devices is critical to the success of that process. You can find more information about the way DIWG views adaptation in our paper on principles (http://www.w3.org/TR/di-princ/). To give you some idea of the scale of the problem, as a commercial supplier of software that supports delivery of content across multiple channels, Volantis maintains a substantial body of information on close to 1000 web capable devices and device families. Other
suppliers have to do this too. The number of devices grows almost daily. Not only do recommendations like CC/PP and its applications, such as UAProf, address many of the issues associated with communication and maintenance of this information, the mechanism also allows dynamic changes in properties to occur. These can form the basis for future capabilities that allow users to tailor the way their device operates and hence the way that they receive content and interact with the web. This is seen as extremely important for support of accessibility and for situations where modality changes with use. Taking a voice-enabled browser into a library, or taking a display-enabled device into a car are examples where the user's environment may constrain the device modality, for example.

So, in summary it is very clear that extensions to existing approaches are needed to support content being delivered over mulitple channels to the huge variety of devices that are now web capable. The challenge is to support authors to create content that can be delivered this way so that the web remains as close as possible to a single, seamless entity accessible by anyone from anywhere with whatever device suits them. CC/PP is a key component of this kind of capability, as are forthcoming versions of XHTML and the recently announced XForms recommendation. DIWG is working on the additional modules that will help make creation of multichannel content more affordable for authors.

May I take this opportunity to wish you well in your studies and especially in the work on your thesis. It occurs to me that you might find it instructive to try and write at least part of your thesis in
XHTML that could be used across a range of devices from PC's to small mobile handsets with restricted
memory, and to try and get an acceptable user experience on all of them.

And finally, your referenced paper is, unfortunately, in postscript format. Whilst this is commonly used in acedemia, it is now much less common outside. I'm tempted to say that these days it is rather
device specific for postscript printers! I had to move it to a UNIX machine to look at it. Many people won't have the luxury of being able to do that.

It's usually considered best to provide HTML versions of attachments when posting to W3C lists if you
want the largest community to be able to access them. PDF is also a reasonable and fairly 
device-independent alternative. 

Very best wishes

Rhys Lewis, Ph.D.
Chair DIWG
Director of Software Architecture
Volantis Systems

-----Original Message-----
From: Rotan Hanrahan [mailto:Rotan.Hanrahan@MobileAware.com]
Sent: 17 February 2004 17:38
To: Kai Hendry; www-di@w3.org
Subject: RE: Multi-channel content

I partially agree.
A well designed document should be accessible by anyone on any device.
Where this falls apart is in assuming that XHTML in its current incarnation is sufficient to represent a document that can be accessible by anyone on any device.
The DIWG has explored the challenges, deficiencies, requirements and current techniques relating to this issue. I urge you to read this material. Perhaps, eventually (with some guidance from our Group and related Groups) we will achieve a document representation and adaptation strategy/mechanism that will achieve the goal of "write once, read anywhere". I personally doubt this will happen any time soon, and I encourage you to research the gap between what we have today (e.g. XHTML, CSSMQ, XForms, SMIL) and what we (probably) need in the future.
As for CC/PP, the need for CC/PP (or some technology that satisfies the goals of CC/PP) will become obvious as your research matures. But I hasten to point out that describing CC/PP as a "waste of time" is hardly a good way to engage with a group of highly dedicated people from around the world who have expended so much effort in creating CC/PP...
Dr Rotan Hanrahan
Chief Innovations Architect
(Member, DIWG)

	-----Original Message----- 
	From: Kai Hendry [mailto:hendry@cs.helsinki.fi] 
	Sent: Tue 17/02/2004 17:20 
	To: www-di@w3.org 
	Subject: Multi-channel content

	I am in the process of writing a masters paper [1] on the topic of
	device independence.
	I thought I should write to this list, to see if anyone shares my belief
	that we do NOT have to have :
	"multi-channel content, i.e. services available for various groups with
	various browsing equipment"
	I feel a little on my own here compared to fairly large amount of
	research on the opposite end. I wonder if someone can give me some words
	of encouragement, or tell me where I have gone wrong.
	Right now, I believe a well designed XHTML document should be accessible
	by anyone on any device. I also think CC/PP is a waste of time.
	[1] http://www.cs.helsinki.fi/u/hendry/work/thesis/topic/report.ps


Received on Wednesday, 18 February 2004 04:02:51 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 19:54:23 UTC