(If this got posted a number of times my apologies - JISCmail has been  
misbehaving on me)

Hi Richard,
> The talk by Helen Petrie and Chris Power at UKMW09 yesterday got me  
> thinking about the whole process of content negotiation.
>
Thanks for this.  It is nice to know that we have people thinking in  
this direction.

> A quick bit of research reveals that there is a W3C Recommendation  
> for Composite Capability/Preference Profiles [1] dating from 2004.   
> This says:
>
> "A CC/PP profile is a description of device capabilities and user  
> preferences. This is often referred to as a device's delivery  
> context and can be used to guide the adaptation of content presented  
> to that device."
>
> Sounds ideal, until you read the text and find that it only talks  
> about device capabilities, not user preferences.
Indeed, device preferences is but one component of it.  In order to  
have true personalisation for anyone you must adapt to preferences  
regarding the user and their wishes and the devices that are available  
to them wherever they are, and match these  to the content available.   
This is discussed in the area of adaptable and adaptive systems quite  
a bit, primarily in reference to mobile devices.

> And certainly not "user capabilities".
Building ontologies, classifications or other models of "user  
capabilities" is not really what people talk about in adaptable  
systems for people with disabilities these days.  This is an old view  
of modelling disability.  Instead, we are interested in their  
preferences.  As you rightly observe later in your mail, these could  
be anything for anyone regarding the what content they wish to  
receive, how they wish to interact with that content,  or how they  
wish to have it delivered through their devices.

> Why not define an ontology for use within CC/PP ...
I'm afraid that I must disagree here.  This is the 'holding the wrong  
end of the sword' problem I discussed yesterday.  With this statement,  
we are taking a technology and then saying "we must represent things  
in it".  It is important that we think about what we want to  
represent, and then implement it in some technology or markup language  
(one of which might be CC/PP).  Also, this has implications beyond  
just web resources, but also any digital resources, or indeed non- 
digital resources.

> To me, it makes much more sense to proactively adapt the content as  
> it is delivered to the user, rather than forcing them to make do  
> with the vanilla page served up to everyone else, and then expecting  
> them to bring specialised tools to bear on the problem of accessing  
> its content.
There are two issues here.

First, this type of adaptation, of rich personalised contents,  is  
definitely what we were advocating yesterday in our talk.  An  
information ghetto of low fidelity pages for people with disabilities  
is not an option we have ever advocated.  :-)

This type of adaptation is being discussed in several different  
domains, primarily in education and online learning.  Our EU project  
is currently doing this type of content personalisation.  There are  
several standards that have come before, and several coming now, which  
are all intended to deal with the labelling of people, technology and  
content.  All of these have serious problems at the moment, mostly to  
do with no one thinking about communication of meaning in the media  
and instead focussing on the delivery (and I say this as someone who  
has been, unfortunately, intimately involved in one of these standards  
for about a year).

Secondly, there is the problem of actually preparing the content  
appropriately.  Currently, few in the world of adaptation for people  
with disabilities is really thinking about how we prepare content to  
communicate a message.  Instead everyone is worried about how do we  
automatically transform what exists into some other form.  Our  
argument is: if we work at the other end, preparing several different  
content presentations that communicate the message of the original  
author, we can then worry about delivering that message through  
technology.

There are advantages and disadvantages to this proposal certainly, and  
we were very encouraged by the warm reception we received from people  
who spoke to us after the conference and the Jodi Awards last night.

> How nice it would be if the pages just came up bigger without my  
> having to tweak browser controls.
Again, this is adaptation in terms of how we deliver the content, not  
in the content itself.

> What do other MCG Affiliate Members [2] think?  It surprises me that  
> the web's powers-that-be expend so much care and attention on  
> setting up mechanisms to customize content for delivery on different  
> physical devices, while doing nothing at all to support the  
> requirements of the people who actually have to make sense of that  
> content.

I think that this is a somewhat unfair statement.  There are many  
people and organisations trying to work with the W3C's CC/PP work,  
such as the UAProf standard for user agents that extends it. There is  
further work, such as the IRIS project that have work on this problem  
as well, specifically for assistive technologies.  Finally, there is  
the excellent work in the MultiReader project that did turn the  
argument on its head and focused on the content and the delivery, not  
just the delivery itself. There are also several very clever people  
doing ad-hoc approaches of providing alternatives or enhanced contents  
in education and in culture sectors.

Also, be aware, that there are several competing efforts at the moment  
to be "the" answer on how to use CC/PP and related efforts.  The  
Ubiquitous Web Application Working Group (UWA) (which is a group that  
arose from a sequence of 3 other working groups in the W3C) is  
currently debating a lot of this.  I'm never one to defend standards  
organisations, however they have been struggling since the inception  
of CC/PP to get mobile providers on board with the idea of using it  
for personalisation.  Indeed, the creation of the UWA came about  
because the finally, after many years of work, got the technology  
providers on board.

We also have to remember, that up until the last year, maybe two,  
there has not been the type of ubiquitous mobile technology for  
presenting really rich multimedia content.  That has changed the game  
substantially.

As I mentioned, we, and our colleagues, have been doing this work in  
the context of education, from which we have learned many valuable  
lessons which I think is of great value to the ML sector. The  
presentation last night was a first step in gauging interest in those  
lessons, and given the extremely positive response we received, we  
look forward to pursuing this with all of you with vigour.

Thank you again for the thoughts.  And thank you to everyone for the  
kind reception last night.

Cheers,
Chris Power

****************************************************************
For mcg information visit the mcg website at
http://www.museumscomputergroup.org.uk
To manage your subscription to this email list visit
http://www.museumscomputergroup.org.uk/email.shtml
****************************************************************