Why not look at complexity as being itself multi-dimensional, with one
axis the human comprehensibility, another the different allowed states
of nodes, and another the rules for interaction of nodes, another the
semantic specification etc.
It seems a good deal of the conversation is about why one
aspect=dimension is more appropriate than another, but they are all
significant.
Why not also look at the dual, simplicity, where that is closeness to
the origin and complexity is a weighted measure of distance from the origin?
Then perhaps one could decide if it is a good thing (simpler) to have an
average function, on the basis that it covers both tallying and dividing
as one operation.
That leads to juxtaposing ideas of simplicity against ideas of complexity.
Differing requirements may lead to different essential ways of doing
what yields the same result, for example, data may be assigned to a list
or an array or a relation depending on whether memory, speed or ease of
query access is dominant. So what is important about the complexity is
partly the human dimensions, and partly the degree of fitness for
purpose (as seen by a human).
Relative simplicity:
From some perspective it's easier, from another it may require more
effort, but when one is weighed against the other there is an advantage.
Synergistic simplicity:
A useful whole can be built up in many ways using essential or
relatively simple parts. The complexity of the combined pieces is
submerged in the interface and behaviour of the whole.
A standard deviation function may use the average function. Synergy is
achieved by a layering of simpler components.
That is if you choose to move along one axis further than along another
the trade-off between simplicity and complexity changes.
In any case the references by various contributors to other ideas have
been quite useful for me an i have some reading to catch up on.
Regards,
Gerry Jensen
02 9713 6004
Richard Karpinski wrote:
Come on guys. Systems are complex only with regard to what you want to
do with them.
Is a book or a computer more complex than a brick or a block of wood?
That depends. If you want to read it, then many answers are possible,
but they are very different from those that arise if you just want to
carry it to the next room, or pack it into the trunk of your car. What
if you want to analyze their chemical composition? What if you want to
construct and legally sell a thousand copies of one?
In the absence of your plan for the object, you can't expect to agree
on how complex it is.
Perhaps you think that because your two objects are both disembodied
computer programs in symbolic form, the comparison of complexity is
easier. Nope. 'Pends on who is trying to do what to them. The skills
and knowledge of the actors matter. The task to be undertaken matters.
The environment in which the task will be performed matters.
So? So don't even ask their complexity. Instead ask the questions that
matter like, "How soon can YOU transform either of these into
something that also does THIS?" Just because that is a question which
potentially has an answer still doesn't mean that you'll find that
answer, but now at least, you MIGHT be able to get one. And it MIGHT
be right.
--
Richard Karpinski, Nitpicker extraordinaire
148 Sequoia Circle,
Santa Rosa, CA 95401
Home: 707-546-6760
http://nitpicker.pbwiki.com/
------------------------------------------------------------------------
_______________________________________________
fonc mailing list
[email protected]
http://vpri.org/mailman/listinfo/fonc
_______________________________________________
fonc mailing list
[email protected]
http://vpri.org/mailman/listinfo/fonc
