Re: Is OO anti-Math? (Re: Whose Fish?)


Don Roby wrote:
topmind wrote:
Don Roby wrote:

AndyW wrote:

On 14 May 2007 07:32:03 -0700, Jordan Marr <jnmarr@xxxxxxxxxxx> wrote:



On May 11, 11:02 pm, AndyW <a...@xxxxxxxxxx> wrote:


On 11 May 2007 08:10:23 -0700, Jordan Marr <jnm...@xxxxxxxxxxx> wrote:




So you're saying that the more mathematically oriented people could
probably abstract the problem down to a purer, leaner form in their
code, without needing the extra overhead of modeling objects around
the problem domain?

jordan


What I am suggesting (and it would require a study) is that the more
'compex' or higher order math a person learns the more likely they are
to tend towards creating procedural systems and the less likely they
are to be able to deal with OO techniques at a conceptual level.

This would certainly require a study, and I find the suggested theory
pretty dubious. Of course I would, being a mathematically oriented
person working as a software developer and producing object-oriented
systems. ;-)

Math involves some pretty abstract concepts, so I don't see the
expectation of a decreased ability to deal with abstraction in software.

If you elaborate on why you expect this, it might be an interesting
discussion.


Whether the above is true or not, I think one has to agree that OO is
kind of the anthesis of math.

One certainly does not have to agree to this. I do not.

It tries to deal with things as
mirroring some kind of physical representation (objects, things,
little machines that talk to each other) and navigate them as if
walking around a room or hallways (getNext), etc.

A set of interacting state machines is not quite the "antithesis of math".

Nor is the use of list or tree styled data structures and traversal by
iterators.

A more mathlike
approach would do something like (A x B), applying transformation A to
all B. It does not strive to mirror the physical world of things if it
is not necessary to solve the problem. For example, a transformation
to one item versus the same transformation to 1000 items may not be
treated different in math-think.


OO can be used to model real things. This is perhaps one of the
historical roots of the paradigm, but this is not all it's useful for.

Math also has historical roots in modeling, measuring and counting real
things. It has evolved to model, measure and count non-real things,
such as models, counts and measures.

But it still carries the philosophical baggage of "physical stuff"
modeling and hierarchical noun taxonomies (or mutually-exclusive
subtypes). You can work around it, but it FEELS like a workaround
still.


In both, unimportant stuff for the current goal is omitted from the
models. That to my mind defines the word 'abstraction'.

To your mind only. Sets and relationial are a better abstraction than
polymorphism to my mind. Sets better fit how I see biz info and see it
change over time and offer me easier ways to study such info from
different angles. Sets allow a more "mathy" way to work with info
than polymorphism.


What you're trying to describe with your (A x B) and transformations
blather is likely also mathematical, but not clearly any more so than
OO. Nor does it clearly describe something contrary to OO so far as I
can understand it. Perhaps you've omitted too much from your model.

Food for fodder...

(I smell a coming debate on the meaning of "math"...)


Debate all you like. I expect I'll stop here.

We'll just have to agree to disagree. I find OO rather low-level,
messy, primative, and Go To like.

-T-

.

Relevant Pages

  • Re: Is OO anti-Math? (Re: Whose Fish?)
    ... without needing the extra overhead of modeling objects around ... Math involves some pretty abstract concepts, ... approach would do something like, applying transformation A to ... This is perhaps one of the historical roots of the paradigm, but this is not all it's useful for. ...
    (comp.object)
  • Re: my electrical box
    ... now, in keeping with the general flow, math seeks to answer why 2+2 = ... modeling knot work, and knot work math modeling is extensive google ...
    (rec.bicycles.tech)
  • Is OO anti-Math? (Re: Whose Fish?)
    ... Don Roby wrote: ... Math involves some pretty abstract concepts, ... mirroring some kind of physical representation (objects, things, ... approach would do something like, applying transformation A to ...
    (comp.object)
  • Re: Eric Gisse, your comment was ridiculous
    ... >>transformation, ... the question is of math operations on equations. ... So what does that light ray look like from B's viewpoint? ... If observer B is moving v from observer A, ...
    (sci.math)
  • Re: Eric Gisse, your comment was ridiculous
    ... >>transformation, ... the question is of math operations on equations. ... So what does that light ray look like from B's viewpoint? ... If observer B is moving v from observer A, ...
    (sci.physics)