Interaction Patterns
Call for Participation

What Are We About?
User interfaces - especially graphical user interfaces - take significant amounts of time to design and develop. The need to tailor user interfaces to the requirements of individual users' needs imposes a heavy design burden, and this, combined with complex to use development tools and inadequate test facilities, results in lengthy development cycles.

The adoption of human factors techniques and expertise, widely practised in larger user interface groups can lead to highly usable interfaces. However, the required expertise is often in short supply, and furthermore barriers may exist between the human factors groups and the development groups because of the different skills and backgrounds involved.

A common language for the development of user interfaces is missing. Although many attempts have been made to create rapid development environments, these frequently focus on the reuse of components at a low level, such as buttons, menus and so on.

Very little effort has been expended on bringing together the object development communities (in which are included the pattern community) with the human factors communities. In short, a good deal of effort is made developing 'easy to use' development environments, but little on ensuring that the results of these tools are tuned to the requirements of the end user of the resulting interface. User interface development tools and methods frequently focus on one environment to the exclusion of others, for example on graphical user interfaces excluding command line, audio based or tactile interfaces.

The concept of a pattern language describing the operation of user interface elements offers the hope that a reasonably standard and systematic set of observed phenomena in user interfaces may be catalogued for future reference. This will allow user interface developers and human factors experts to commuicate in a common, non-technical language and for toolkit developers to build user interface toolkits based around macroscopic concepts as well as microscopic details.

Pattern languages of this form offer a number of interesting opportunities to those of us working in user interface development. Firstly, they offer a language by which team members can communicate with each other - whether human factors or programmers and other developers. This may sound trivial, but in practice is a significant benefit. Secondly, the development of interaction patterns offers the possibility of developing toolkits based not upon low level constructs such as buttons, windows and menus, but upon far higher level constructs described by a pattern. My group, AT&T Labs UK, has been developing just such a toolkit since January 1998.

What Will We Do?

This workshop would have potential impact in a number of areas:
  • the development of an on-going interest group who can share ideas and work to the mutual benefit of all
  • the development of a standard terminology or language for interaction patterns and for user interface concepts generally
  • the exposition of core ideas required for the development of more rapid user interface development systems and toolkits.
  • a sharing of ideas and concepts between human factors experts and object development experts

How Will We Do It?

The format of this Hot Topic might take a number of forms:
  • open dialogue to frame the issues of primary interest
  • compare notes on work to date in this field, both published and unpublished
  • develop a taxonomy of interaction patterns, both published, known and suspected
  • with the help of ChilliPLoP newcomer session, introduce non-patterns aware human factors consultants to patterns, yet have a last interactions patterns session to report back to newcomers with an interest in this arena
  • review submitted papers subject to time constraints and priority agreement during ChilliPLoP
  • planning activities beyond ChiliPLoP

As Eugene Wallingford commented last year, without knowing who has proposed what in response to a call for participation, it is difficult to get too specific about the workshop plan. No doubt, therefore that the following plan will not survive contact with the enemy, as the saying goes! For what its worth, I propose the following outline structure:

Pre-Workshop Preparation
  • candidates submit position papers, with pointers to materials they use, in response to the call for participation
  • participants review materials submitted by others
Day 1
  • a short opening session to introduce everyone and to frame the goals of the workshop
  • a couple of demonstrations of interactions patterns
  • attempt to further develop a taxonomy of what interaction patterns might exist, and review work (and workers) to date, especially those detailed above.
  • break-out sessions to work on interaction patterns
Day 2
  • readers workshops on some of the elementary patterns out there and their use, with a focus not on learning the patterns but rather on what makes a pattern good for novices and on how to use them in courses
  • break-out sessions to continue work on interaction patterns
Day 3
  • workshop the materials developed the preceding two days for the focus topics [this is the climax of the workshop]
  • an open forum, especially targeted at human factors folks new to patterns to review the work to summarize what we we achieved.

Who Should Participate?
I hope to attract three main groups to this hot topic workshop:

Object Patterns folks with an existing interest or track record in interactions patterns who can focus the group and avoid repetition of existing work

Object Patterns folks new to interactions patterns, but keen to participate

Human factors experts with a desire to learn more regarding object patterns within the context of their field - joining the hot-topic on the last day, but hopefully contributing more after the event .

I suspect and hope that a balance between industry and academia and between human factors and 'hard core' develop folks can be achieved in the interests of balance.

What Are The Important Dates?

February 16, 1999
Submit your request to participate by this date.

February 23, 1999
We notify all applicants whether they have been accepted to participate or not. Workshop participants commence e-mail discussion to tailor the goals and activities of the workshop.

February 23, 1999
We will make all resources submitted by the participants available on the web for reading and discussion.

March 16-19, 1999
We gather at the Wickenburg Inn and (set out to) accomplish our goals!

How Do I Find Out More?

Please contact me by email at bbrook@uk.att.com

Note: PLoP is a trademark of The Hillside Group, Inc.