Page 1 of 1

XST: Extended Set Theory

Posted: Mon Apr 24, 2006 7:26 pm
by therandthem
Hello, my name is McKinley. I have been studying the relational data model for a few years. I have been following Rel development since the beginning. I recently watched the video which I found very informative. At the prompting of the video, I started to investigate Extended Set Theory (XST).

I want to discuss what your findings are on XST as discussed here and here. It is simple enough to understand that it helps formalize data structures beyond Classic Set Theory (CST). What can it do, however, to replace indexing? That part is lost on me, but I will keep looking for answers. Not much is written at this point.

Kind Regards,


Posted: Tue May 09, 2006 1:18 am
by Dave
My apologies for the delay in responding...

I'm not sure XST can replace indexing, at least not universally. However, it may provide a suitable basis for implementing index data structures such as BTREEs, hash tables, and the like.

I suspect some XST operators may internally benefit from indexes in order to optimise performance, but it remains to be seen whether these will have to be implemented as primitives, or whether they can be optimally composed from primitive, non-indexed XST operators.

Re: XST: Extended Set Theory

Posted: Tue Jul 28, 2009 3:09 am
by mrankin
Can somebody post a link to the video? I'd like to see it.


Re: XST: Extended Set Theory

Posted: Tue Jul 28, 2009 7:34 pm
by Dave
Unfortunately, the video was lost from the server some time ago during an upgrade. I might still have a copy of it somewhere, but it's somewhat outdated now. I'll probably create a new one at some point.