Some of us feel that long ago the validity of underlying technology, SOAP or REST, etc; was set aside by the establishment of interest.
Fielding's thesis defining REST is at:
and is excellent reading, as is the founding documents of SOAP. SOAP has a longer history then does REST.
Several years ago, I thought that REST had some interesting potential as a means to preserve "state" and to treat service computing and fulfillment as if a finite state machine that lived in a distributed space. There were alternatives, such as CoreSystem and multi-core processing with distributed and localize elements.
For me, the REST architecture seems to match well to core tasks defined by global SOA processes. The devil is in the details and the "gaming" that has been done by industries attempting to gain advantages. Again, competition by itself is not necessarily the origin of all evil. In this case; however, . . .
Observations about where we are not is a good illustrator of what is wasteful about current e-Gov projects.
What is the lesson learned?
Lesson Learned: The federal government cannot manage the power of IT consultants to game and to "not solve problems" but rather to institutionalize the act of working on the problem. This non-management leads to huge waste in our efforts as a research and development community.
In e-Gov activities, as in other activities such as sponsored by DARPA and NIST; the platforms have depended on established corporate interests. AI and super-computing lead the way in defining esoteric technology what is simply wrong minded. It is not how fast the multi-core processing equipment can push data, but in the quality of the data, that the most interesting results are obtained. And the reification of RDF triples into well specified OWL with inferences, well this is now seen to yield good results about the non-resilience of data models produced using OWL ontology.
Again, the focus is on billable hours and new projects, rather than solving core technical challenges and making the results easily understood.
Only slowly and with great resistance has there been occasional clarity. Why would REST be considered as a basis for computing service requirements and fulfillments? Why would SOAP? These issues could be approached objectively by setting aside the profound influence of established interest. However, the entire funding infrastructure is designed to reward marketing and to not allow the type of objective science that John Sowa and others have advocated, here in this forum.
We have the marketing folks driving the entire spectrum of underlying technology development. This is producing wrong minded results and leading to more waste and a delay in getting SOA to function globally and in emergencies. The current expenditure for e-Gov (OMB) since year 2000 is 338 billion. Wow! (reference: Jane Fountain - Harvard 2005) .
In some cases, such as the work attempted and completed by Rex, and those that work with him, this work does focus on capabilities and openness. OMG (Object Management Group) provides another example of some good initial ideas (that internet data design should have models) where later developments were governed by marketing. This governance has lead to the quagmire we find e-Gov in.
The best path forward is to redo things that we understand have been done poorly. Wrong!!!!!
I mention DARPA not because there has been no DARPA led advances at all, but because RDF and OWL seems to some of us to have unfixable flaws as a knowledge representation language. I ask the question, "How can one really talk about service oriented architecture without a viable and flexible ontological modeling capacity. Protege is still an unusable "research tool", with no interest by the academics in completing a stable ontology editor. The role of topic maps as a means to represent the ontology of services, generally speaking, has been permanently sidelined. Why?
Illuminators such as Tim Berners-Lee, and Roy Thomas Fielding might help all of us by focusing the community's attention on the issue of profiteering by the established IT consulting industry. How does profiteering work in today's government IT consulting environments.
The issue of the results of massive profiteering is a general systems theory issue, and there could be ontological models (topic maps, KIF, OWL or "n"-ary ontology) of the details, and government action taken to correct what is a profound abuse. This is what I have called for in the Resilience Project Working Paper.
The core issues can be approached as if a manhattan project.
Paul S Prueitt
On Jun 29, 2007, at 9:29 PM, Paul Prueitt wrote:
I'm not sure that our approach would be welcomed given the appearance that the agenda presents of established interests settling into established niches.
On Jun 28, 2007, at 11:19 AM, Rex Brooks wrote: I'm not sure that our approach would be welcomed given the appearance that the agenda presents of established interests settling into established niches.
_________________________________________________________________
|