Re: Magical Thinking and Grimoires

[Sandra Charker <scharker -at- MASTERPACK -dot- COM -dot- AU>]

I have three responses to this message:

1. Arthur C. Clarke's proposition that "any sufficiently advanced
technology is indistinguishable from magic" (quoting from memory and no
I don't know where he said it except that it's in more than one of his
published works), so that grimoires are entirely appropriate. But
current technology is by definition not advanced, so that's not
relevant.

2. Bruce's users' notebooks full of unsystematic notes show "little
overview or understanding of the processes involved" because these are
exactly the users, or are users at exactly the stages, that we target
with procedures and task-oriented information. The notions of
information types and layered information provide a way to separate out
information so that these users can "cast spells" confidently, while
others can find a conceptual framework to apply if the magic stops
working, or just because they like to understand things.

I think that the amount of background information that's appropriate
often depends on the difficulty of the interface. It's not long since I
was working with an MVS-based "office productivity" system that I
understood quite well but couldn't use because it was riddled with
apparently arbitrary inconsistencies in behaviour. Whenever I got into
trouble people would try to tell me why these idiot things existed,
which usually meant describing 10 or more years in the history of the
software, the department, and the company, and then explaining why one
way or another was best. I thought learning incantations and pestering
the help desk was more efficient.

3. Bruce's observation seems to imply that both 'magical thinking' and
'focussed thinking' are intensely pragmatic, and that both are likely to
find straightforward instructional documentation satisfying. IOW, you
can't necessarily generalise from users' behaviour to their levels of
knowledge or their attitudes about a system. Conversely, you can't
necessarily predict users' preferences in documentation from their level
of knowledge about a business domain or even about their own jobs. Which
seems to mean that we're working with ever-increasing uncertainty about
the effectiveness of our own techniques and thought systems - task
orientation; layered information; user-centered design; usability
testing; information design etc. I don't mean these things are exactly
comparable to typing a command with certain parameter settings, but many
times we them with little reference to their theoretical frameworks.

I'm thinking a lot at the moment about how a software interface and
documentation can best satisfy people with different behavioural and
learning styles in organisations that are continually changing and
remaking themselves. Bruce's observation seems relevant somehow.

At the moment, my thinking is strongly influenced by a book called
'Spiral Dynamics', by Don E. Beck and Christopher C. Cowan. It seems to
provide a lot more insight and more useful guides to complexity than
other systems I've encountered. Has anyone else read the book or does
anyone know anything about it?

Sandra Charker

My words are mine, not Masterpack's

mailto:scharker -at- masterpack -dot- com -dot- au

Ph: +61 2 9937 1427  Fax: +61 2 9937 1499

Masterpack International
18-20 Orion Road
Lane Cove  NSW  2066
AUSTRALIA