I have read Chris Crawford’s book The Art of Interactive Design: A Euphonious and Illuminating Guide to Building Successful Software. This book was quite a bit of an eye-opener for me. Not that I completely agree nor disagree with Chris Crawford, but mostly because it helped me articulate some of my thoughts. Also, while Chris Crawford writes about the design of computer interaction, I think that he explains many useful concepts that can also be used for programming language design and library interface design.

Chris Crawford’s main message with the book is that a new kind of “programmer” is needed. Namely, interaction designers. And he thinks that the interaction designers should be recruited from the Arts and Humanities.

Some of my favorite concepts from the book are:

• The Definition of Interaction:
  interaction: a cyclic process in which two actors alternately listen, think, and speak.

  Or in a more computer science jargon we can replace listen, think, and speak with input, process, and output. I think this definition is spot on, and it has helped me to think about interaction problems using this definition. And like for humans a polite and pleasent computer program is thoughtful, carefully listens, and doesn’t hold back important information.

• A Criterion for Interactive Excellence: Crawford gives a nice succinct formular for what is excelent interaction:
  interactive excellence = accessible states / conceivable states

  While the formula from a strictly matematical hair-splitting point of view does not make sense. I think it succintly describes what I have found frustrating–but havn’t been able to articulate–about some computer games and a certain office suite. Namely, that I get some idea about what ought to be possible (conceivable) but for some reason I couldn’t make the damn program do it (not accessible).

• Colour-to-dirt ratio: A concept originating from game design: Every new feature that a game designer adds to a game increases the game’s colourfulness. But every new feature also entails a certain amount of bureaucratic dirt. And a good game designer will ask herself: “Is the colour I’m adding with this new feature worth all the dirt it is bringing in?”. The example Chris Crawford gives is adding stock speculation to Monopoly. An example in programming language terms could be if we considered to add call-cc to SML. This would definately add colourfulness to the language, lots of features would be possible to add as libraries, for example, lightweight threads. However, call-cc also adds lots of “dirt”, certain optimizations are no longer vailid, for instance.
• Educational software should be based on simulation:I could not agree more. For many topics, simulation is a really good way of facilitating the learner to formulte her own theory about how something works. This is one of the reasons I work at Laerdal Sophus.

What I don’t like about the book is that Chris Crawford is beating some points to dead, back to life, and then dead again. In particular I think he is completely overdramatising the two culture problem which is the conflict between Arts/Humanities (AH) and Science/Engineering (SE). Perhaps it is because I’m a SE male who have all the power and is the source of the problem (according to Chris Crawford) and not part of the solution. Perhaps it is because the two culture problem is bigger in the US than in Denmark, which I do think is the case. I have seen references to this conflict in serveral US movies and series, and have sometimes wondered about it before I saw the problem described for the forst time in Chris’ book. However, I think is poor that the only suggestion Chris Crawford has for the SE male who wants to make successful interation design is to show more interest in more kinds of art. I guess I just don’t get it.

Nevertheless I really like the book and Chris Crawford’s in-your-face writing style. A sample of Chris Crawford’s writing can be found on GameDev.net as a featured article. The sample is a chapter from one of his other books on computer game design, and it gives a good illustration of his writing style, and is a good read to boot.

[I feel I have (had?) much more to say about The Art of Interactive Design, but this reveiw have been sitting in my Drafts queue for almost two months now. So here your have it. Maybe there will be a part two later…]