RFC on CR-Algebra

[Giancarlo Niccolai]

I decided to publish my draft on CR-Algebra on Academia.edu for comments. If you take a look at the document, what I have been talking about in the last posts should make more sense.

https://www.academia.edu/s/c4be2bbc5b/concept-relation-algebra?source=link

This is the base of what we may call "concept-relation programming", which I exemplified here: https://groups.google.com/forum/#!topic/falconpl/VeGC1dMBNeg

Everyone's invited to take a look.

[Klaim - Joël Lamotte]

Hi

I cant see anything without being logged in.

I don't want to create an account.

Please paste it in pastebin.com or on github?

--
You received this message because you are subscribed to the Google Groups "FalconPL" group.
To unsubscribe from this group and stop receiving emails from it, send an email to falconpl+unsubscribe@googlegroups.com.
To post to this group, send email to falc...@googlegroups.com.
Visit this group at https://groups.google.com/group/falconpl.
For more options, visit https://groups.google.com/d/optout.

[g...@niccolai.cc]

Hi Joel, sorry I didn't see you before. I am publishing an update based on feedback and further analysis, currently proofreading it. Should be ready on Monday night, I will also share it with you.

Sent from my Huawei Mobile

To unsubscribe from this group and stop receiving emails from it, send an email to falconpl+u...@googlegroups.com.

[Klaim - Joël Lamotte]

Paul sent me a copy,

then I realized I can't understand tit all because of lack of maths understanding (or more exactly math language).

Although I'm familliar with graph theory, I'm far from being a specialist or even math-fluent when talking about them.

Joel

[g...@niccolai.cc]

In that case, you can help out with the programming language side. I am writing it in the wiki area of git hub (falcon2 project, it is easy to find) .

While I was working at that, last week, I realised that an axiom of the cr_algebra theory that is not very limiting in math (the composition must be commutative) turns out to be very limiting as a programming language device (I. E. We would be forced to ignore the position or rank of the parameters of the input functions in node/concepts).

Rather than saying "we use this theory, but with this exception" I preferred to improve the theory so that we can use it directly.
Removing that axiom has turned out to improve the theory a lot, and has also made other assumptions and limitations to fall, with the result that now the math maps the usage I have in mind for the CREL programming paradigm much better. Having math behind this idea is important, as it will help automatic code analysis and massive parallelization.

Otoh, I had to rewrite most of the math expressions, which took a lot of time. The good news is that I have been forced to make them simpler (the necessity of taking ordering into account required stronger math, and this means more intermediate definitions).

It was a bit like refactoring a program breaking a big bunch of code into smaller, more readable functions.

Ad said, I am proofreading it and I think I should be done by Monday.

To unsubscribe from this group and stop receiving emails from it, send an email to falconpl+u...@googlegroups.com.