2011/03/13

Structuring the work.

 Last time was my first post of my first blog, and already I see I made two mistakes
  1. I should probably add, ojalá (or Insha'Allah which is surely where it comes from) if ever I say "...that's for tomorrow"
  2. Seems Jena is not the back-end.
For a while yet I still won't be able to give top priority attention to this blog due to a not-yet-closed contract.

As regards the role of Jena :
"The primary use of Jena is to help you write Java code that handles RDF and OWL documents and descriptions."  Ian Dickinson -- http://www.iandickinson.me.uk/articles/jena-eclipse-helloworld/

Well, you will see I'm pretty far along with that, but I can see I still have much to learn.


I've realized that the best way to structure this work is in three separate blogs, since the material here is going to be interesting to three probably unrelated groups.

This main one, Monetary Ontology Walkabout, will be brief; most of it will point to new items in the other two. It'll be for those who want to follow progress and jump in when I get to something that's important to them.

The second, Monetary Ontology Walkabout - Technicals will be very geeky, but I'll avoid too many buzzwords, and provide lots of tutorials as well as all the source code.  The idea is to offer something people can really use, so I mean to make it easily accessible.

Finally, Monetary Ontology Walkabout - Ontology Stuff, doesn't really exist yet.  This will be for money geeks -- those who want to actually work with monetary solutions but may prefer to have someone else deal with the Information Technology details.  It will tend heavily toward the ontology of money

Most of the work, for now, will be in the Technicals blog, since we are building the foundations on which the rest will reside.

Indeed, I just now posted a detailed setup guide there, with links to a series of tutorials, as Wiki pages, hosted on a wonderful service I've used for over two years with complete success, www.xp-dev.com

Please visit the main Wiki page, and the source code page.


Next steps?

We got to this point 'coz of all the RDF stuff in the Jena tutorial :
%JENAROOT%\doc\tutorial\index.html which leads to this kick-off course :
%JENAROOT%\doc\tutorial\RDF_API\index.html

Apparently, I've done things a bit out of order. The Jena tutorials seem to push you to understand "A Semantic Web Framework" (which is their subtitle of course).  That isn't exactly what I want to do!  I'm more interested in the overall architecture.  I guess I need a map of the whole territory before getting to know the land around a single water hole.

In the next post we'll try get a map of the whole territory.

2011/03/02

Hello World

Hi!  Thank you for coming to visit.

Over coming months I hope to create an ontology of money able to achieve wide spread acceptance.

It will build upon a very crude attempt I made three years ago, very kindly linked to (https://financialcryptography.com/mt/archives/001047.html) by Ian Grigg in his superb blog "Financial Cryptography".  I put it together as an in-at-the-deep-end leap into Protégé (http://protege.stanford.edu/).


That version doesn't have much more value than to act as an intro for interested parties, as in, "There, that's what I'm talking about."

So ... that diagram lets you see what I am talking about, but probably leaves you wondering what the point is, so here's a bit of a Q & A session:

Q: Monetary Ontology. Fine  ...  and Walkabout?

A: A diagram is nearly useless.  Much more is required.  I want something people can use.  For that to happen, an ontology needs to be tested and proven correct.  That involves transforming the ontology into a database structure definition (schema), preparing a database with that schema, loading up it with real world data and then testing that a reasoning engine will always answer queries with realistic results.  This is all new territory for me, so like a pubescent aborigine, I now set out Walkabout to make the vision a reality.  I have the songlines in my head.  I invite you to watch my progress.

Q: What's the point?

A:There are as many definitions of money as there are economists.  Economics poses as a science, but if there exists so much ambiguity about what money is then how scientific, really, is the rest of economics?  Economists have the fine luxury of having authority without responsibility!  ... and what an unholy mess they have made!  This initiative grows out of the conviction that money based on solid foundations is not at all hard to understand; a mumbo-jumbo priesthood is only required when a fraud is tarted up as arcane mystery.


The point is to make it clear what money is, and make it possible to prove the definition by testing it.  The language with which one defines a ontology is extremely limited, only certain kinds of things can be said.  I have dropped out of many discussions about money because the language used was so loose and contradictory that agreement on anything was impossible.  The benefit of a constrained and rigorous language intended to facilitate describing the real world ought to make agreement on many things quite possible.

Q: What's the route?

A: Well from past experience in other areas I think it is wise to work from the back-end to the front-end.  That is to say make sure I know how to :
  1. prepare a simple database schema
  2. load some data into the database
  3. connect to the database from an application
  4. see that query results can be validated through a reasoner and properly reflect the real world
  5. augment the simple database schema by transforming an augmented diagram
  6. repeat 2, 3 & 4



So my first task is step #1.  


Having spent a few hours poking around, it seems that Jena is a suitable candidate for a back-end.


I downloaded the latest version from sourceforge ( http://jena.sourceforge.net/downloads.html ) and found the first tutorial here at %install_path%/doc/tutorial/RDF_API/index.html.

So that's for tomorrow....