> Interesting how a topic far removed from programming, such as the above article, can influence
> software development.
Discrete mathematics, set theory, functions, relational logic etc. have always been at the heart of computing and software development.
Bob
_____
From: BlackBox [mailto:BLACKBOX{([at]})nowhere.xy
Sent: 13 January 2009 22:11
To: BLACKBOX{([at]})nowhere.xy
Subject: [BLACKBOX] Frameworks
Folks,
I am writing an extensive application and am starting to see
the rational for the Blackbox framework.
I have never liked message passing since it is difficult to follow
who is doing the sending and who should do the interpretation
of a message.
In my current application I have been forced to implement a form
of message passing. However, unlike Blackbox that uses Views
for its message hierarchy, I have a collection of nodes (models
if you will).
My messages are called Symbols (since I am working in the domain
of language). Symbols also have Coders which transform symbols
of one type into symbols of another type. I had kept the coders and
symbols separate until I read an article on xxx.lanl.gov
http://xxx.lanl.gov/PS_cache/arxiv/pdf/0812/0812.1946v1.pdf
CATEGORICAL FOUNDATION
OF QUANTUM MECHANICS AND STRING THEORY
by A. Nicolaidis
Theoretical Physics Department
University of Thessaloniki
54124 Thessaloniki, Greece
nicolaid{([at]})nowhere.xy
January 4, 2009
wherein he discusses Category theory and relational logic developed by C. S. Peirce.
Peirce defines a 'subject' as the sum over all the 'relations' (transformations) that generate
the subject.
Ah, says I. I can group my a symbol coder with its symbol since the symbol now becomes
the output of all the inputs to the coder: Many in, one out. With this new understanding
the amount of code I had to write was greatly reduced and unified.
So a symbol is both a message and the interpreter of messages. If it receives a message
(symbol) that it doesn't know, it ignores it (in the same way that Smalltalk and Objective-C do).
Some coders will take a sequence of input messages before they emit a single output message.
Other coders will take a single input message and produce a sequence of output messages.
So, in general one can have: many in, many out.
Interesting how a topic far removed from programming, such as the above article, can influence
software development.
(by the way, I think the above article is significant for the physics community)
Best regards,
Doug Danforth
---- To unsubscribe, send a message with body "SIGNOFF BLACKBOX" to LISTSERV{([at]})nowhere.xy
---- To unsubscribe, send a message with body "SIGNOFF BLACKBOX" to LISTSERV{([at]})nowhere.xy
Received on Tue Jan 13 2009 - 23:23:47 UTC