Prototyping & GOMS

12 Jun 2008 - 11:29am
6 years ago
3 replies
605 reads
Santiago Bustelo
2010

CogTool, developed by Carnegie Mellon University's HCI Institute, has
a really interesting approach: perform a KLM-GOMS simulation over the
prototype.
http://www.cs.cmu.edu/~bej/cogtool/

Sadly, current version is frustrating to use, to say the least. I hope
they find the support and funding needed to take the project further,
but I won't expect a miracle... last release is dated more than a year
ago.

--
Santiago Bustelo // icograma
Buenos Aires, Argentina

On 12/06/2008, at 07:00, John Gibbard wrote:

> In terms of prototyping fortunately with tools like iRise (ahem),
> Axure, DENIM and Thermo jostling for position we're in the fortunate
> situation where - if we've got the time and space to do it - we can
> do hundred of iterations too. That's not to say it's everyone's
> bag and we do have to be careful that iteration doesn't just mean
> repeat but includes incremental, quantifiable improvement.
>
> John

Comments

12 Jun 2008 - 3:23pm
Bonnie E. John
2008

Hi, I'm the PI on the CogTool Project and I want to assure you that
we are continuing work and will have a new release in Fall 2008. Most
of the new features are already working, have been tested with
internal users on real projects, and we are now concentrating on
documentation, so the new release is really coming.

We have made a lot of usability improvements (yes, we've used
CogTool on our designs for CogTool as well as think-aloud usability
studies and other HCI techniques). So hopefully some of the
frustrations Santiago mentions have been taken care of in the next
release. As with any software development project, we are juggling
our resources and priorities, so we can't fix everything for the
next release, but if you send us requests and bugs, we'll do our
best.

In terms of using CogTool for building prototypes, we have made a lot
of improvements. Its very efficient now to make menu systems and
systems of related widgets like radio buttons and list boxes. And you
can export your prototypes to HTML to share with folks who don't want
to download CogTool or use as a basis for quick-and-dirty user
testing.

So please keep watching and let us know what you think.
Bonnie John

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=30188

13 Jun 2008 - 1:43pm
Todd Moy
2007

Bonnie -

could you give us a basic idea about what CogTool does--and how it might be
applied?

>From the site, I seem to believe this is used to model efficiency along a
single path. Assuming individual flows within a storyboard could be widely
variable in terms of completion time, would you use this tool to identify
those as areas to think more critically about? Also, do you believe this
could be applied to an environment that supports multiple ways of achieving
a goal? The tutorial leads me to think "no", but it's entirely possible I've
missed the point. :)

-Todd

____________________________
http://www.oombrella.com
oombrella /a/ gmail.com

18 Jun 2008 - 10:48pm
Todd Moy
2007

Josh,
thanks for that "crash course." It definitely gave me a better lens into
when and how the tool could be applied.

Cheers,
Todd

On Wed, Jun 18, 2008 at 11:20 PM, Josh Weihnacht <
joshwhynot at inputpattern.com> wrote:

> Having taken a class with Bonnie where we used CogTool, I'll give it a
> shot.
>
>
> On Fri, Jun 13, 2008 at 2:43 PM, Todd Moy <oombrella at gmail.com> wrote:
>
>> Bonnie -
>>
>> could you give us a basic idea about what CogTool does--and how it might
>> be
>> applied?
>>
>
>
> CogTool is basically about automating predictions of how long it would take
> an expert user to perform a task or tasks on a system. You build screens
> with various widgets on them and then use these screens to create one or
> more scripts of how a user would perform some task in the system. The
> program then estimates how long it would take an expert user to perform that
> task in that manner.
>
> >From the site, I seem to believe this is used to model efficiency along a
>> single path. Assuming individual flows within a storyboard could be widely
>> variable in terms of completion time, would you use this tool to identify
>> those as areas to think more critically about?
>
>
> It can certainly be used to model two different ways of performing a task
> on the system (say with a mouse vs a keyboard. or even two completely
> different flows). Then it will give you estimates of how long each task
> would take an expert user to perform, along with a graph of all the various
> events (moving the cursor to a target, clicking the mouse, recognizing that
> the screen has changed, etc). This can help you to identify where time
> savings could be found in order to make the task more efficient. It is
> certainly intended to help you think critically about ways to improve
> completion time as well as to compare completion times of different tasks.
>
>
>
>> Also, do you believe this
>> could be applied to an environment that supports multiple ways of
>> achieving
>> a goal? The tutorial leads me to think "no", but it's entirely possible
>> I've
>> missed the point. :)
>>
>
> Yes, definitely. Sounds like there is more work for the CogTool team to do
> on the tutorial.
>
> Hope that helps.
>
> best
> Josh Weihnacht
>
> http://www.inputpattern.com
>

--
____________________________
http://www.oombrella.com
oombrella /a/ gmail.com

Syndicate content Get the feed