[Arisbe] Re: Inquiry Into Information
Jon Awbrey
arisbe@stderr.org
Mon, 27 Aug 2001 10:36:33 -0400
¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤
Howard,
A few morning afterthoughts on the questions of yesterday ...
Software for a complex application needs to have a generic model of the task.
Inquiry support software needs to have a model of inquiry, so that it can be
given its head to follow the drift of inquiry, and if not lead then at least
be able to nudge us in plausible directions without having continually to be
dragged along the way.
You do not need to have a grandmother cell for inquiry,
but you do want the horse to know the way to her house.
By looking at human, all too human inquiry, one is gathering data for the sake
of reverse engineering the way that Nature's Evolutionary Application Programmers
implemented the generic specifications for inquiry out of the particular materials
that were available during a particular planetary epoch. But finding the revelant
generalizations is nigh unto impossible when all you have to go on is a single case.
That is why logic and inquiry systems engineering have irreducible analytic, design,
normative, and speculative components as natural parts of their tasks.
Jon Awbrey
¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤