The main source of invention is not "math wins" as described on
http://www.vpri.org/html/work/ifnct.htm since the world would be speaking
math if it were really the source of inspiring more inventions that improve
the world's standard of living. Math helps add precision to tasks that
involve counting. Attempting to move from counting to logic such as in
statistics sometimes leads to false conclusions, especially if logic is not
given priority over the tools of math. For human value, readability is
required, so computer language improvements must focus on natural language.
Human language itself has problems seen in large projects such as Ubuntu
where contributors from around the world write in their own language and
tag their code with favorite names which mean nothing to the average reader
instead of words which best explain the application. Thus a major
improvement for world computing would be careful adherence to a world wide
natural language. We know cobbling together a variety of languages as in
Esperanto fails. While English is the world standard language for
business, Hebrew might be more inspiring. In any case the use of whole
words with common sense is more readable than acronyms.
The first math language Fortran was soon displaced in business by more
readable code afforded by Cobol's longer variable names. In Smalltalk one
can write unreadable math as easily as readable code but Smalltalk may have
a few legacy bugs which nobody has yet fixed, possibly due to having
metaphor or polymorphism design errors, where the code looks good to
multiple programmers but fails to perform as truly desired in all
circumstances. Further reluctance to use commonsense whole words on some
objects such as "BltBlk" present a barrier to learning directly from the
code.
One way to reduce these errors is to develop a set of executable rules that
produce Smalltalk, including checking method reuse implications. Then one
could make changes to a few rules and the rules would totally reengineer
Smalltalk accordingly, without forgetting or overlooking anything that the
programmer hasn't overlooked in the rules. There is also room for a more
efficient and more natural language. Smalltalk is supposed to be 3 times
faster to code than C and Expert systems are supposed to be 10 times faster
to code in than C. So a better language needs development in two
directions, easy to understand Expert rules using common sense whole words
and a built in library which enables "Star Trek's Computer" or "Iron Man's
Computer" level of hands free or at least keyboard free function.
There are three basic statements in any computer language: assignment, If
then else, and loop. Beyond that a computer language should provide rapid
access to all common peripherals. Expert systems tend to have a built in
loop which executes everything until there are no more changes. Some
industrial process controllers put a strict time limit on the loop.
Examining published rules of simple expert systems, it appears that random
rule order makes them easier to create while brainstorming, it is possible
to organize rules in a sequential order which eliminates the repeat until
no changes loop. Rule ordering can be automated to retain freedom of human
input order.
Several years ago I worked with a Standford student to develop a language
we call Lt which introduces a concept of Object Strings which can make
rules a little easier. Unfortunately the project was written in VBasic
instead of Smalltalk so I've had insufficient ability to work on it since
the project ended. Soon I'll be working on converting it to Smalltalk then
reengineering it since it has a few design errors and needs a few more
development cycles educated by co-developing an NLP application.
Here's a simple Lt method which is very similar to Smalltalk
game
"example Lt code"
| bird player rock noise |
'objects
rock exists. player clumsy.
'facts
player trips : [player {clumsy unlucky}, rock exists]. 'a
if x w or x y and z
noise exists; is loud : (player trips, player noisy).
'a and b if x or y
bird frightened : noise is loud.
'a if x
(bird ~player has : bird frightened.
'case: if b then not a else a.
bird player has.).
^
'answer rock exists, player clumsy, player
trips, noise exists, noise is loud
'bird frightened
Now to complete the project without corporate resources, it is necessary to
select an NLP application which is both more powerful and physically
smaller than IBM's Watson which won against Jeopardy's best players. The
most powerful NLP text in history is the Bible which is only 4 Mb instead
of Watson's 4 Tb. Bible analysis can be very rewarding - getting software
to develop the footnotes in the free New Testament from
http://www.biblesforamerica.org would be a first step, rewriting them based
on my discovery of the true church being what Jesus practiced before Paul
appeared would be progress. After that level of achievement, it should be
easier to apply the system to any problem.
This level of NLP mastery in or external to an outside and indoor robot
could be used to end poverty, illiteracy, crime, terrorism, and war around
the world by growing and serving food, educating and entertaining a family
with the same language and religion cradle to Ph.D. in some ways similar to
this video http://www.hulu.com/watch/69831/outer-limits-family-values,
especially note the ending. While there are amoral inventions such as
smoking, all good inventions come from a desire for absolute perfection of
spirit and being resulting in faith for improving some aspect of life.
Anyone wanting to work on this must agree to not allow any use of the
system in the forthcoming rebuilt Temple in Israel for many reasons,
especially the Creator's warning in the Gospel.
Love Truth,
Kirk W. Fraser
_______________________________________________
fonc mailing list
[email protected]
http://vpri.org/mailman/listinfo/fonc
