I'm sure some, if not many, of us are hungry to hear/read some reports from
UPA's recent Montreal soirée.
Dave (at) ixdg (dot) org
Dave (at) synapticburn (dot) com
AIM: bolinhanyc || Y!: dave_ux || MSN: hippiefunk at hotmail.com
Here is an extract of my conference report sent out to interested folks
within my company (Landmark). Since what follows is an extract of a
report, things may seem a little disjointed
--- cut here ---
UPA 2005 Conference Report
27 June – 1 July 2005
Montreal, Quebec, Canada
I found the 2005 UPA conference to be informative, enjoyable, and quite
likely helpful. Attendance was again up from the previous year: 525+
attendees, and there were more job postings on the bulletin boards and
more vendors this year, so I take that to be a sign of a generally
improving economic environment for usability.
There were four particularly good presentations that contained things
which I feel that I found quite useful:
• Making Software Learnable for Reading-Averse Users
• Bridging Internal Cultures: Getting Cross-Functional Buy-In
• Usability Analysis Visualization to Improve Communication and Build Trust
• Strategy and Tactics for Agile Design: A Design Case Study
Agile Usability Workshop
On Monday, I co-facilitated a workshop on Agile Usability with Jeff
Patton of ThoughtWorks and Thyra Rauch of IBM. We had 10 participants,
and the workshop was generally very well received.
In small groups, we discussed and cardstormed challenges to doing
usability in an agile development environment, some possible solutions
to those challenges, and then adaptations that participants have made to
traditional usability techniques/deliverables for use in an agile
Several people rather liked the “Evil DUG’s Top Ten Ways to Annoy Users”
poster and the DUG Tip Sheets as an adaptation.
Breakthrough Design Tutorial
On Tuesday, I took a tutorial entitled “Breakthrough Design: Innovation
that Works for Users” given by Larry Constantine. The tutorial was
engaging, will be quite helpful and is a modification/extension of some
of Constantine’s previous work on usage centered design. Several
techniques were presented that I would like to try here at Landmark
including card storming (which we also did during the workshop),
essential models, and abstract prototyping.
As an exercise during the day, we split into different teams, each
working on a design problem brought to the tutorial by one person on the
team. I ended up on a small team working on an RSS feed area for a
website used exclusively by members of Congress and their staffs. While
our design isn’t something that we’ll directly use at Landmark, I do
think that the design pattern could be used for similar things (like
Constantine did cause a couple of people to walk out during the break
because he advocated focusing on usage rather than users.
There were many interesting presentations, but four stand out as being
the most likely to be things that I want to use or try at Landmark. The
presentations are in the order that I feel we are likely to be able to
implement at Landmark.
Making Software Learnable for Reading-Averse Users
John Schrag of Alias presented six strategies that the usability team at
Alias have used to make software more learnable by users who don’t like
to read the online help or even read the text in dialogs and such. His
examples also showed the benefits of having graphics designers available
to the usability team.
Their techniques generally involve using professional quality graphics
to help draw the eye and illustrate concepts. Among the strategies
mentioned are: Welcome Screens, Just in Time Help, Graphical Error
Dialogs, Error Recovery Dialogs, Tricking the user into the help, and
Heads-up Display Feedback.
One of the reasons that the strategies work for them is that they do
have graphics designers preparing the graphics.
Another reason is that they do considerable testing on concepts, designs
Bridging Internal Cultures: Getting Cross-Functional Buy-In
Larry Marine presented what his consulting firm does to help clients
create successful products. One of his interesting comments was that
fittest to market wins all the time over first to market.
The most immediately useful part of his talk is the idea of a task
priority matrix which is used to determine development priorities and
get cross functional buy in. Each task item is rated by usability
(rating user experience), the product manager (rating business
need/goal), and development (rating technical feasibility) using a scale
from 1 to 3 (3 being the most important or the easiest to implement).
The final rating for each task is the sum of the individual ratings.
Then, the team determines a cut off total rating and focuses efforts on
the tasks which fall above the cut off.
Usability Analysis Visualization to Improve Communication and Build Trust
Rally Pagulayan from Oracle presented this talk on a technique that he
uses to improve the speed at which developers can see usability test
results and build the trust the developers have in the usability tests.
Essentially, during the usability test, he builds an affinity diagram on
a prominently located whiteboard by transcribing user comments and
difficulties on post it notes after each test subject. Developers can
come by and see how things are going over the course of a week long test
and see how users are having similar problems in certain areas. By the
end of the usability test, a large whiteboard is covered in grouped post
it notes which quickly give an idea of the places where users had
difficulties and the general scope of those difficulties (more
whiteboard area covered = a bigger problem).
Rally still prepares a traditional usability test report, but I feel
that in a more agile environment, the actual whiteboard (suitably
recorded) or a derivative version (on feature cards or electronically in
Canvas or Visio?) could be used instead.
A downside of the technique is that the number of test subjects per day
is lower (1 to 2 only) and the technique would not work as well in a
Strategy and Tactics for Agile Design: A Design Case Study
Desirée Sy from Alias presented how the Alias usability team has adapted
their activities to match an agile development environment.
Essentially, the Alias usability team have adapted their usability
activities to meet with a 2-4 week iteration cycle agile development
environment by scheduling and planning their activities to match the
shortened cycle. In general, the usability activities are one cycle
shifted forward (plan interaction design for cycle 2 in cycle 1). Some
designs may take more than one cycle to finish, but they still feed
things to the developers in time by carefully choosing what should be
developed (with the development team).
The team does have time to conduct usability tests during the cycles and
takes advantage of the in house users (testers) and local resources (art
Deliverables from the pair of usability specialists assigned to the
development team are also modified so that they are more appropriate for
an agile development environment: features are put up on features cards.
Usability test results are communicated in person and issues are also
placed on a whiteboard in the form of feature/issue cards.
The usability team does derive considerable benefit from having two
usability specialists assigned to each project, having two dedicated
graphics designers available to the usability team, and having a
dedicated intern developer available for producing prototypes. Desirée
was able to show successive high fidelity prototypes of Sketchbook Pro
developed by the intern developer.
A final note about the conference. I have historically added a little
note about design or usability (not presented at the conference) that I
observed while at a conference to my conference reports. This report is
The conference was held at the Fairmont Queen Elizabeth, a high end
hotel in Montreal. UPA conferences are traditionally held at high end
hotels or resorts except for last year, when the conference was held in
a Marriott. Almost every year, I see an email from the conference
organizers a couple of months before the conference, imploring attendees
to make reservations at the high priced conference hotel rather than the
reasonably priced alternatives nearby. This year, like every year
except last year, I stayed at perfectly fine and reasonably priced hotel
some distance away from the conference hotel. You would think that the
conference committee for the lead usability conference would make the
choice of conference hotel more amenable to more attendees by selecting
a decent business class hotel rather than a high end hotel.
Another irony about the Fairmont Queen Elizabeth is that this high
priced hotel seemed to have a lack of appropriate serving utensils
(tongs, cheese knives, and other basic serving utensils) and almost
always just placed regular knives, spoons, and forks for serving at the
buffet. The staff is probably quite adept at picking up sandwiches and
green been salad with a regular spoon and a regular fork, but for the
rest of us, tongs would have been considerably more usable.
--- cut here ---