[tpm] Clarification- Challenging Pure Pearl Object - Multi-D
cj at enersave.ca
Sat Feb 9 15:55:52 PST 2013
Honestly I am only looking for a co-author. Someone like me who has
a fair idea of how to do Object correctly but isn't an expert on it
or anything else related to anything else. I don't care how much
anyone knows about anything else, just a fair idea of how pure Pearl
objects can work correctly to be extremely fast and efficient. I
forgot to mention there is a very fast efficient gaussian solver I
need to write. I have a fast solver in C with you guessed,
pointers. It's a nightmare because the solver is in unity offset
matrices and C gives you 0 offset pointers to work with. But it works.
I am at a cross roads in my career. I have a brilliant idea and I
know exactly how I want to do it and I want to do it in pure object
oriented Pearl, the correct way the first time. First my abstract
for a paper I know exactly how to write if only I completely
understood pure Pearl Objects.
I have been doing exactly the same thing since 1995 with, wait for it:
dos C. With a limited understanding of pointers never mind
objects. I knew pointers were the key to understanding memory which
was somehow important. I quickly learned how important. Sometimes
it took days to learn how one misplaced pointer could screw things
up, eventually. And then it took days to figure out how to fix
it. Then I worked for a guy who understood that pointers were not 2
dimensional, nor 3 dimensional. In fact multi-dimensional. Cool, I
know I can use this someday and let it lie.
We continued to use our same 2 dimensional model to represent a quasi
3 dimensional model. It was very effective, and very fast and has
made my a good portion over my career. I know have a way to solve a
problem that bothered my since I started making money with this
model. It isn't as accurate as some people would like. Who cares,
it has made me a fair amount of money and has made other people a lot more.
I know what the next step is and I want to "control it". All I need
is to become an expert in Pearl fast, very fast. The best way to
become an expert on anything is to write a paper, present some nice
pictures at a conference and you are off the races. That is what I
did in 2001. I co-authored other papers but let the other person do
the conference thing - ever two years or so since then. So he gets
the benefit, because he completely understand and uses this model and
the tools I developed.
Here is my abstract:
A Perl Object Oriented Program to model a multi-dimension matrix.
We have developed Perl Object, Multi-D, control a 2-d matrix model to
represent the same 3-d system.
It is as simple as that. I know all the math, I know the complete
structure of the Object. I know all the inputs and outputs to each
object within the object right down to the base object and how it all
interrelates to product the same output object. I also know the
complete structure of the output object and how all its tiny little
objects represent all the tiny little objects in the input
Object. So in essence, I have a fairly good understanding of the
object. And I have it all laid out and named and what each object
represents in 10 files. Text files all object oriented. I need the
Object Multi-D to first read these files to create itself.
Does this sound like an exciting, challenging pure Pearl Object? I
need a co-author. I have been to the Mongers meeting with another
Pearl problem about this same Object which I already knew but didn't
know what to do with it that might get someone interested.
Can I come to another Mongers meet, this time with a nice slide show
with pictures and matrixes and solvers and structures and all that stuff.
The curve is just about to start because I started it, the same as I
did in 2001.
Christopher Jones, P.Eng.
Suite 1801, 1 Yonge Street
Toronto, ON M5E1W7
email cj at enersave.ca
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the toronto-pm