EARL and IUSR

At 03:10 PM 2001-08-03 , Sharon Laskowski wrote:
>FYI.  We, at NIST, have been following the development of EARL and are going
to 
>try a couple of examples of the output of our WebSAT tool ( that does some 
>usability checking) represented in EARL   The Common Industry Format for 
>Reporting on User Tests (from the IUSR project, another project in my group)
is 
>a format for report on testing that a usability engineer does with users,
not 
>for automated checking, so the format contains information such as the user 
>demographics, the tasks, the metrics (such as time on task, and number of 
>errors), the data collected about these metrics,etc.  So, the CIF reports
more

>on the process and data analysis from testing with users.  You can read more 
>about it at <http://www.nist.gov/iusr%A0>http://www.nist.gov/iusr  and please
feel free to email us for more 
>info. on the CIF.  I'm not sure what the synergy would be except that
something 
>analogous could be build to report on testing for accessibility, but to
report

>on the output of automated tools testing, say 508, would be very different
that 
>anything in the CIF.  Sharon

Great!

EARL is not meant to be limited to expressing the results of automatic
evaluations.  Its ambitions extend to integrating the kind of information
expressed in IUSR CIF as well.  This doesn't mean that users sold on using
IUSR
would necessarily have to go back and learn EARL.  But we might be after you
for an 'authorized binding.'

IUSR reads like something that usability engineers can relate to.  But it
would
be more effective if that information can be freely joined with information
from other sources.  Check out Bob Grossman's droll assertion of "W's Law: the
interest in data grows as the square of the number of columns you correlate."

Alliance All-Hands PowerPoint Presentations
<http://fantasia.ncsa.uiuc.edu/media/2001Meeting>http://fantasia.ncsa.uiuc.
edu/media/2001Meeting/

IUSR implies a schema and this schema could be used to recast the information
in an RDF binding of the EARL model.  Depending on how formal you have been in
defining the IUSR information space, it may be a relatively mechanical
operation to get them interoperating.  Or there may be structural
primitives in
the IUSR scenario that we forgot to cover in EARL.  Which is what we need to
know early in the process.

One could use RDF 

 Working Paper SIDL-WP-1999-0126
 <http://www-diglib.stanford.edu/cgi-bin/get/SIDL-WP-1999-0126>http://www-d
iglib.stanford.edu/cgi-bin/get/SIDL-WP-1999-0126

or another logic-capable language 

 Model-Based Mediation with Domain Maps
 <http://www.sdsc.edu/~ludaesch/Paper/icde01.html>http://www.sdsc.edu/~luda
esch/Paper/icde01.html

to create a cross-schema-mapping beteween the EARL core vocabulary and the
IUSR
information model; and just process IUSR reports interoperably with any other
EARL-interfaced tool reports (both automated tools and tools used to capture
reports from the results of interactive evaluations, as with the GMD standards
work).

There's a lot there for us to continue talking about.  IUSR annexes in EARL,
other cross-fertilizations.

Al

Received on Friday, 3 August 2001 16:41:59 UTC