W3C home > Mailing lists > Public > public-semweb-ui@w3.org > June 2007

RE: UCD Description for SWUIWiki "How To" section

From: Duane Degler <ddegler@ipgems.com>
Date: Sat, 2 Jun 2007 14:13:45 -0400
To: "'Hyowon Lee'" <hlee@computing.dcu.ie>, <public-semweb-ui@w3.org>
Message-Id: <20070602181031.0FDC4D4005@mail.itcinternet.net>

Some additional thoughts about context and experience...

If you set out to build a better mousetrap, you probably have a mouse, or
some passing experience with mice.

As analyst/designers (and researchers), we are often called on to build
products in unfamiliar domains. The processes we undertake - which in this
discussion are labeled "UCD" - is the way we gain an understanding of the
domain, conditions, and context to inform design. That design may be
evolutionary or revolutionary, but it begins with understanding.

UCD techniques help you in four ways:

1. Systematically explore and understand the environment, context, goals,
prior experiences, and constraints of a domain and the people that are
within it.

2. Expose, question, and clarify your own assumptions about the domain, the
people, and the intervention/artifact that you are imagining for that

3. Confirm your understanding over time by contact with the people in that
domain, to evaluate that your proposed intervention - and expected change in
the environment - has value and relevance.

4. Communicate your understanding, approach and assumptions in common
formats that other people can understand and learn from, whether in academic
papers, requirements/design documents, prototypes, etc.

Now, from the designer's perspective, it is vital to apply imagination and
your experience to designs, in order to achieve valuable break-throughs and
improvements. Saying that, how the design process is informed by context is

Say you were asked to design emergency controls for a nuclear power station
- if you had never been in one or even knew someone who worked in one, would
your first action be to sit at a whiteboard and design the innovative
solution in isolation? The same can be said in the multimedia domain - if
you have never stepped into a video editing environment, or more extreme had
never watched television, could you envision the types of tools and
facilities that would be required to overcome practical challenges in the

I suspect the people who developed the iPod had some prior experience using
MP3 players or a cassette Walkman or at least a boombox - some sort of
portable music device. They probably also had some experience with a large
number of CDs or records in their collection, thus with the cognitive
process of finding and selecting something they were personally interested
in. The widget designs are clever, but probably started from a contextual
foundation and then refined via experience and feedback.

Let us make explicit the implicit assumption in a lot of software design -
that because designers/developers use computers and software they can
perceive themselves as users, and thus "know" what is required from users.
It is not surprising to me that a lot of experimental projects I see relate
to creation of development tools, searching databases (e.g.
conference/abstracts), sharing/collaborating with professional project
colleagues, researching... all tasks that are common within the domain of
researchers and developers. For example, the web browser began in that

To relate this to SWUI - what problems are you trying to solve? What areas
do the particular techniques of the Semantic Web relate? What affordances
can be applied? What efficiencies can be gained... and in comparison to what
pre-existing challenge or interaction? What can we imagine, now that certain
enabling technologies are becoming available... and once imagined, how can
we communicate our understanding and the products of our imagination?

It is not either-or... Innovative design and UCD are extremely
complementary, in service of vision and improvement.


> -----Original Message-----
> From: public-semweb-ui-request@w3.org 
> [mailto:public-semweb-ui-request@w3.org] On Behalf Of Hyowon Lee
> Sent: Saturday, June 02, 2007 11:07 AM
> To: public-semweb-ui@w3.org; Eric Miller
> Subject: Re: UCD Description for SWUIWiki "How To" section
> Hi all,
> I'm a UI researcher for newly-emerging multimedia 
> techniques... this makes my job a little different from 
> conventional HCI approach that emphasises UCD.  
> For example, I start with plenty of brainstorming and focused 
> (but un-precedented) interface sketches, mostly only with the 
> promises of the technologies that my group offers me.
> I try to involve users, but mostly ONLY after the initial 
> prototype is done and working - so that I can bring in people 
> and get them use the system, collect opinions, etc.
> Eric mentioned examples of Apple's successful designs, where 
> it's not the result of UCD but the creative or novel thinking 
> of the designer or visionary.  I guess for a completely new 
> technology where there is no conventional practice or user 
> base, it works fine when somebody (who knows technology) can 
> come up with some new design and then follows that up with 
> exposing it to people in near-by domain or approximate 
> discipline, in order to get their feedback and refine 
> afterwards - combine Eric's 'X factor' with UCD follow-up.
> I guess now that Apple's new products are used by large 
> number of people, Apple can get user feedback and survey etc. 
> to refine it to make it better.  I'm sure there were many 
> small and large elements that the initial design didn't get 
> right, even if the overall UI was a great invention.
> Anyway on the outset, I agree with the importance of 
> designer's novel and creative approach (especially at the 
> initial stage of development).  Let me think about how I can 
> relate this to SWUI.
> Hyowon
> ----------------------------------------------------
> Dr Hyowon Lee
> Post-doctoral researcher
> Centre for Digital Video Processing
> Dublin City University
> Glasnevin, Dublin 9, Ireland
> Tel: +353 -1 -7005829
> http://www.computing.dcu.ie/~hlee/
> Email: hlee@computing.dcu.ie
> ----------------------------------------------------
> ----- Original Message -----
> From: "Eric Miller" <eric@squishymedia.com>
> To: "Roberto Garc í a" <rogargon@gmail.com>; <public-semweb-ui@w3.org>
> Sent: Thursday, May 31, 2007 11:54 PM
> Subject: Re: UCD Description for SWUIWiki "How To" section
> > 
> > Hi all,
> > 
> > Just a quick observation here.  I'm a practitioner, not a 
> researcher, so I
> > can't really speak to the lit, but I use UCD on a daily 
> basis.  And speaking
> > as a designer I've found that user involvement provides an 
> invaluable
> > starting point for the design process but shouldn't solely 
> govern evaluation
> > processes.  Users can be good at articulating their needs, 
> but aren't as
> > good at articulating specific solutions to their needs; 
> often because their
> > viewpoint is reflexively restricted to their personal 
> contexts and past
> > experiences.  Their involvement isn't as useful to the 
> practicing designer
> > when the users don't have experience with that specific 
> class of application
> > or task.  Like, say, the Semantic Web.
> > 
> > I saw that this issue was noted under "Methodology Issues" 
> on the SWUIWiki,
> > incidentally.  And the "Pathetic Fallacy of RDF" piece 
> touches on this too.
> > 
> > So I'd suggest that processes like UCD (as formally articulated and
> > implemented, and driven by actual people and/or personas) 
> are well-suited
> > for evolutionary refinement of existing classes of 
> applications with a
> > pre-existing critical mass of user awareness and 
> understanding.  But to
> > create effective user interfaces for new classes might 
> require a less
> > structured and more creative approach.  (My example here 
> would be Apple's
> > successful new product design process that is decidedly not 
> UCD-driven in
> > the traditional end-user sense; it's driven by Steve Jobs' 
> benevolent
> > dictatorship of design.)  Another interesting thread along 
> these lines would
> > be the "genius design" v. UCD discussion a few months back on IxDA.
> > Additionally, there might be a case made for adapting SW 
> technologies to fit
> > a more coherent interaction metaphor, rather than 
> struggling to bolt a UI
> > onto a fully-baked technical implementation.
> > 
> > Again, this audience would certainly know more about the 
> HCI lit than I
> > would, but I'd simply offer this thought.  To paraphrase a 
> great one-liner,
> > "Writing about design is like dancing about architecture."  
> There's a
> > subjective X factor in design that defies quantitative processes and
> > analysis.  And I'm willing to speculate that breakthroughs 
> in SWUI design
> > would be most likely to spring from creative processes 
> unencumbered by
> > constraining mechanisms such as formal UCD processes and 
> methodologies.
> > There will be more failures; but there's the chance for 
> something truly
> > innovative and successful to emerge.
> > 
> > Eric
> > 
> > 
> > On 5/31/07 1:26 PM, "Roberto García" <rogargon@gmail.com> wrote:
> > 
> >> Dear all,
> > 
> > Before adding the following text to the "How To" section of
> >> the
> > SWUIWiki, I would appreciate your comments about the following
> > description
> >> of User Centred Design (UCD). Do you think it
> >> is
> > appropriate?
> > 
> > --
> > 
> > 
> > 
> >>  User-Centred Design
> > 
> > The methodology for developing
> >> usable and accessible applications is
> > called User-Centred Design (UCD). It is
> >> based on an iterative
> > development process based on a detailed study of the
> >> users' needs, the
> > tasks they carry on in order to meet them and the context in
> >> which
> > they are performed (Norman 1986). There are many UCD
> >> development
> > processes proposals, but all of them provide a mix of
> >> software
> > engineering plus usability and accessibility engineering tasks.
> >> One
> > particular proposal, which combines both disciplines in a neat way
> > thus
> >> facilitating its adoption, is the usability and accessibility
> > 
> > 
> > 
> > 
> >
Received on Saturday, 2 June 2007 18:14:23 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 19:53:51 UTC