Deferential Design: Being Prepared for Re-Design
http://www.acm.org/~perlman/chi2001/deferential.html

Gary Perlman
perlman@acm.org
http://www.acm.org/~perlman/
OCLC Online Computer Library Center

In Perlman (2000), I described how attempting to address multiple access issues motivated an architecture that was more adaptable along anticipated and unanticipated dimensions. Designing a system that adapted to language changes, browser differences, user levels, etc. resulted in a system that was better able to adapt to change in general (e.g., disability access, new functional requirements). One conclusion is that an adaptable architecture -- a universal design -- is cost effective when viewing the whole life of the system. During initial development, universal design may be viewed as an unwarranted expense, but it can provide dramatic benefits when face with changing requirements and design. This is especially true for Web-based applications, which change on a time scale unheard of a decade ago. But it is also true for more traditional applications, for which the same methods have been applied (Perlman, 1989). My guiding principle has been that of deferential design, not to be able to get the design right the first time, but to be able to change what is wrong. The name also has a useful alternate meaning, of respect and esteem.

OCLC, being a non-profit cooperative of over 38,000 libraries in 76 countries, has a special interest and responsibility for universal access along many dimensions: economic, language, disability, etc. OCLC is also able to be more open than for-profit companies about its development practices and experiences (both positive and negative). During the workshop, I would be able to contribute information on the methods and technology we have used at OCLC for making the old FirstSearch service multilingual (Hysell & Perlman, 1999) and the new FirstSearch service more universal along several dimensions. I would like to spend some time differentiating methods (e.g., separation of function, language, and display) from different technologies (e.g., Java classes, Cascading Style Sheets, etc). I would also like to work on incorporating ideas of universal functionality -- functionality available across all applications -- with universal access (Perlman, 1992).

As director of the HCI Bibliography (hcibib.org) and the maintainer of the following internet resources of ACM SIGCHI and SIGCAPH:

I believe I can provide special assistance with identifying the state of the art in universal design, particularly for HCI.

References

  1. Hysell, D. & Perlman, G. (1999) Lessons Learned from Internationalizing a Global Resource, (HTML, with followup) Presented at: IWIPS99: 1st International Workshop on Internationalization of Products & Systems, Rochester, New York, USA 21-22 May 1999
  2. Perlman, G. (1989) Coordinating Consistency of User Interfaces, Code, Online Help, and Documentation (50k HTML) in J. Nielsen (Ed.) Coordinating User Interfaces for Consistency, pp. 35-55, Academic Press.
  3. Perlman, G. (1992) (HTML) A Vision of Universal Functionality for Tomorrow's User Interfaces, Keynote address in the Proceedings of OZCHI 92 Australian Conference on Human-Computer Interaction, pp. 1-14.
  4. Perlman, G. (2000) The FirstSearch User Interface Architecture: Universal Access for any User, in many Languages, on any Platform (229k Adobe Acrobat PDF) Proceedings of the first ACM Conference on Universal Usability, 16-17 November 2000, Arlington, Virginia, USA. pp. 1-8. New York: ACM.