[Sigia-l] IA research?

[Anne Miller]

It always bothers me that people equate research with experimentation as the
only valid form of research. Experimentation is an entirely reasonable
approach to research but only when you have a well defined,
operationalisable research question. Arriving at a well-defined research
question happens only after you've employed a range of other usually
non-experimental methods to understand a problem space. 

HCI, human factors, anthropology, ethnography and so on are coming to and
perhaps usurping AI because the problem spaces we investigate lead us to
questions that involve categorisation (AI) issues. Thus we come to AI with
an understanding about where AI fits in a broader problem space; what we
don't have but are in many ways developing are methods and orientations for
dealing with categorisation problems. What does this say about AI? First is
says that AI has a place. Second it says that AI would benefit from looking
at broader problem spaces so that it can take charge of the development of
techniques development. Third, it may say that AI needs to better define
itself as a contributor in a multi-disciplinary research context so that AI
researchers can claim their seat at the research table.  

By engaging with other disciplines (sharing disciplinary secrets!!!) does AI
risk loosing it's identify?  Quite probably but then this is the nature of
scientific (r)evolution (Kuhn's Paradigm). It's not so much that AI will
loose its identity as much as that it will morph into something else through
the integration of its methods with other methods and approaches. This is
exactly what has happened to HCI. Below is the blurb for the 2005 CHI
Conference from http://www.hcirn.com/res/event/chi.php:

"Welcome to CHI 2005
CHI 2005 is the premier international conference for human-computer
interaction. Learn more about our conference theme of Technology, Safety,
and Community in the Conference Overview, and about submission venues in the
Introduction to the Technical Program. Please join us in Portland and take
advantage of the many opportunities to network and showcase your research,
design, and practice!

Compare this to the following blurb for the first CHI conference in 1982
after its split from the Human Factors annual meeting. From 1983 this
conference has been known as CHI'__: Conference on Human Factors in
Computing Systems

"Human Factors in Computer Systems
March 15-17, 1982, Gaithersburg, Maryland, USA
Attendees: 907
The proceedings appeared as a special issue of the SIGCHI Bulletin in
January 1993.T his predecessor meeting was indeed entitled "Human Factors in
Computer Systems" and not "Human Factors in Computing Systems" like the
later CHI conferences."

You can check out the evolution of HCI through the CHI proceedings. 

At the end of the day the body of knowledge is in my opinion less important
than the suite of tools, techniques and methods used to uncover knowledge.


          

Dr Anne Miller
Program Coordinator, Human Factors Online
www.uq.edu.au/human-factors-online
Key Centre for Human Factors & Applied Cognitive Psychology
Ph: 61 7 3365 4543
Email (preferred): amiller at humanfactors.uq.edu.au


-----Original Message-----
From: sigia-l-bounces at asis.org [mailto:sigia-l-bounces at asis.org] On Behalf
Of Peter Van Dijck
Sent: Monday, 22 November 2004 8:32 AM
To: aifia-members at lists.ibiblio.org; sigia-l at asis.org
Subject: [Sigia-l] IA research?

Apologies for cross posting and for pointing to a blog entry and for the 
long link.

I wrote a long rambling entry about (the lack of) IA research and 
innovation. Are we building a practice and a body of knowledge, or are 
we slowly dying out? Again, apologies for linking, but here goes:

http://poorbuthappy.com/ease/archives/2004/11/21/2167/

I'd like comments/ideas. Am I the only one who feels we, as a 
profession, need to get with the program? Oh well.

Cheers,
Peter


------------
When replying, please *trim your post* as much as possible.
*Plain text, please; NO Attachments

Searchable list archive:   http://www.info-arch.org/lists/sigia-l/
________________________________________
Sigia-l mailing list -- post to: Sigia-l at asis.org
Changes to subscription: http://mail.asis.org/mailman/listinfo/sigia-l