|
SOA CoP, please make comments and join a discussion by emailing to
Regarding a possible panel to be added to agenda at:
Susan,
I would like to make a response directly to technology similarities and differences between DBA pureXML and paradigms like coreSystem, and perhaps Klausner will make a more detailed response from his perspective. Others may join and perhaps we can organize a specific proposal for the panel or panels.
I am cc'ing the SOA e-forum and ask that Steven Newcomb and anyone else who would like to send me a private message to be included in an email discussion.
DBA pureXML uses a data encoding innovation that stores XML in a tree structure natural to the pure concepts of XML, thus providing one means to create near perfect marshaling and unmarsheling between readable and process-able XML and storage or even transmission structure of data. Indexing optimization is then layered on top of this. We are all getting really good at this.
However, the underlying problem still exists related to how information comes to be, meaning how information structure and meaning becomes inducted; and how information might be used in a reasoning engine, ie allowed to be process by a deductive logic.
These are issues addressed by both the XTM (topic maps) and the OWL (web ontology based on RDF tripes). What we find is the OWL, and to a lesser extend TM, has difficulty with a class of problems related to the induction of structure that can then be used to make inferences.
Versioning has to do with the assignment of syntax and semantics (meaning) to data student so that as the world, or our perception of the world shifts, the pre-specification of meaning to structure stay in alignment with some necessary correspondence between our abstractions and the real time moment. This problem is central to great debates, and is not resolved even always in natural linguistics, as a comparison between Whorf and Chomsky's linguistic theories easily demonstrates. However, there are ways to gain clarity on this by looking to the nature of induction itself. The viewpoints are strongly held on all sides, so I will not go into my work on this issue.
What I would like to do is to suggest that the relational data base itself continues to have problems that impact the re-versioning of both structure and function (meaning) IN REAL TIME AND WHEN THERE IS NOVELTY.
The conjecture is that there are alternatives which have arisen and been discarded not because of some real utility function, but due to certain paradigmatic limitations within current industry practice.
So, in summary; for what ever reason, there is entrenched inability to address a class of concerns related to informational structure necessary to match the expressive power of natural language, but within the concept of a distributed semantic net.
The objective is to realize and enhance collective intelligence.
Paul S Prueitt
On Jan 23, 2008, at 10:59 PM, Susan Malaika wrote:
Paul
I'm busy between and 9 and 11 eastern tomorrow. Earlier or later in the day would to talk.
I work in DB2 pureXML and specialize in industry formats.
Loking forward o talking.
|