I guess my message wasn't very clear, because two of three replies to it
misinterpreted it.
There are certainly many "scientific professionals in academia and
industry" who would find a zero-price CAS useful. But it is too broad a
category to be useful in a marketing analysis. It's a little bit like
saying that the target market for the VW New Beetle is people who drive
passenger cars. True, but useless.
That is why I think we need to sharpen the characterization of our
target markets: For example, as Dave suggests, small independent
engineering firms; as CY suggests, undergraduate physics departments (as
opposed to research university physics departments, say); as Robert
suggests, incidental or occasional users. I am not sure those are the
right categories -- I'm just giving examples.
Wolfram Research has already done some of the work of defining major
market segments for us. Their price categories divide the world into:
commercial / industrial / government users ($1900); academic users
($900); precollege and community college educators ($195); student users
($140). And of course they have add-on packages at extra cost for a
large variety of application areas: electrical engineering, control
systems, time series, etc.
To be sure, $140 sounds like a lot of money for a student. But I am
apparently out of date. A quick check at Amazon shows that college
calculus textbooks go for $100-200, as do graphing calculators. I also
learned recently that the going rate for AP Math tutoring is $80-120 per
hour... Given that a CAS is something you presumably use in many of
your courses if you're a hard science or applied math major, $140
suddenly sounds positively reasonable....
> ultimately our worries about marketing will come in the future,
> after the GUI situation is better under control
I don't agree. Defining our target market helps us figure out where we
want to put our development effort. What is the relative value of (a) a
slick GUI; (b) fancy plotting; (c) interval arithmetic; (d) solving
differential equations in terms of Bessel functions; (e) interfacing to
Excel or OpenOffice Calc; (f) easy extensibility? It all depends on our
target market.
-s