Thursday, May 19, 2005
concept of standardizing unique definitions for words
communication to http://groups.yahoo.com/group/jena-dev/
I understand Ben. You are a member of the advisory board and as such your advice always counts.
You said:
Paul,
It sounds like a smart plan if you can find the angel investor.
However, 3 months seems tight, I think you'd do better to spread out
the investment over 5-6 months and involve fewer or cheaper people. 3 months isn't very long to implement
anything significant in an impressive-looking and finely-tuned way.
-- Ben
when I send to you part of a message that contained the following:
The
first step, I imagine would be for the angel investor to talk with each of the
CEOs of the companies who will contribute the components that are to then to
deployed together ... and the also make discussions with a senior person at
Customs as to whether or not Customs would allow the system to be deployed at
little or no cost to Customs.
This
gets political, as it is possible to have two or three of the US Senators
involved in advising that Customs will cooperate if there is no cost to the
government, in an innovative approach to the problem of selectivity and
targeting commodity movements in world markets.
The
issue is the color of money. All money
that is supposed to be used to solve the problem is vacuumed up... and so one
way to make the first deployment is to have a independent economic process do
the work and provide it at no cost to a key client.
The
budget for making a deployment in this way is likely within 300K and could be
done within 3 months.
……
The fact is however, that the differential ontology system can deploy immediately and it is in the deployment that the "real work is done". There is very little coding that is needed.
First step is to bring the existing COTS systems to a group of Customs analysts and provide these existing COTS systems, in mass, to this group. It is my inside knowledge of the situation that let us know that the 1.6 Billion spent at eCP is largely wasted because the system is designed to do something other than solve the selectivity and targeting function required by Customs to “know” what is crossing our national borders.
So we solve the real problem and give this solution free to Customs. The prime contract may even get credit for the solution. Who cares? Two things happen.
1) the people of the United States are protected better
2) the investor gets a deployed tool set that he/she has developed a branding language over and some long term contracts and new patents
A real certification program, with paper certificates from an institution that has done this type of thing for a while, (www.kminstitute.com ) is given - as well as an understanding of how the Roadmap’s technologies work, individually and together. The investor gets a stake in this certification program (and given the need for certification – this might be bigger than the software integration stake.)
I have not emphasized this 80-20 issue as much as I should. But 80% of the effort is in getting existing product to be used within a methodology. It is knowledge management 101. We are not developing one more enterprise system from the ground up. The eCP prime contract is doing that at a rate of a little over 1.3 M per day.
Well we are developing an enterprise system, but no one is paying for this and it is not happening while important people sit waiting for the software engineers to finish.
I should say that Knowledge Foundation Inc's Mark 3 system would be a major added component to the Roadmap, and would be deployed along with the other components.. In the spirit of the Knowledge Sharing Foundation concept...
http://www.bcngroup.org/area2/knowledgeSharingFoundation.htm
we want to be able to deploy within a single highly visible environment all of the best tools, provide educational experiences for those who wish to learn, and then let the market decide. This environment is a few beta sites in a system with literally 1,000 additional deployment sites. And this is only US Customs. A system for the analysis of all world wide trade data is the prize.
Let me be very clear about this.... the problem is not that there does not exist semantic extraction and ontology mediation software that works. The problem is that the clients do not know how to use this software and the prime contractors do not care to do anything other than time and materials software engineering. The problem is also that most of the existing Semantic Web software is not actually functional - and goes through major revisions almost monthly. But the fact that most people are looking at the wrong technology is not the same as there not being good software. The best software puts the software engineers out of business... because the problem is 80-20.
The Roadmap approaches this situation in a vastly different way. We have a science/technology advisory committee, of which you are one distinguished member. We have an extended community of though leaders, many who I do not publish the name of.
We have a historical account as to where information technology has made some errors, and we have selected the best software components that are consistent with the "second school". We have a movement that can challenge the first school (read, Tim Berners-Lee’s notion of a Semantic Web). We may even have TBL himself come to endorse many of the principles that differentiate the second school from the first school.
Doug Weidner's KM certification organization is positioned to deliver the second school curriculum and to provide government employees with knowledge of the tools set and the underlying methodology.
So it is not about new code development. It is about teaming agreements, patents and a social movement.