Posted Oct 2, 2005 9:37 UTC (Sun) by efalsken (guest, #32813)
[Link]
Modern technology and improvements in object-oriented technologies are finally enabling us to open the door to high-performance object-oriented data systems. In a sense, object-oriented data can be much more relational and dynamic than current RDBMS systems. I highly advise reading some of the papers available on ODBMS.org, and evaluating this new generation of data systems yourself before criticizing an entire technology group. Modern ODBMS systems are not a throwback, and couldn't be more different from the era that you mention. If you had read some of the information available on ODBMS.org it would be pretty clear.
ODBMS.ORG Launches Portal on Object Databases
Posted Nov 19, 2005 15:49 UTC (Sat) by leandro (guest, #1460)
[Link]
Modern technology and improvements in object-oriented technologies are finally enabling us to open the door to high-performance object-oriented data systems.
OO data problems were never about performance, but conceptual integrity. Lacking a clear, solid base, OO data management is doomed to complexity, lack of expressive power and an utter failure in efficient ad hoc data manipulation.
ODBMS.ORG Launches Portal on Object Databases
Posted Nov 19, 2005 16:35 UTC (Sat) by leandro (guest, #1460)
[Link]
Some more stuff...
In a sense, object-oriented data can be much more relational and dynamic than current RDBMS systems
Which sense?
The only current RDBMS systems are AlphoraDataphor and perhaps Leap (there are minor implementations in the guise of libraries, like Duro), for all SQL systems violate fundamental principles of the Relational Model. OO is the same, but even worse.
I highly advise reading some of the papers available on ODBMS.org, and evaluating this new generation of data systems yourself before criticizing an entire technology group.
Why should I do so, when OO lacks theoretical foundations and even the proper elements of a data model? It may be useful, but it will never fulfill the promises of the Relational Model, which are finally coming into fruition. In the end they will be relegated to the dustbin of IT History, just as many other ad hoc, ill-defined, complex, limited and overhyped technologies. No amount or quality of development will be able to fix the OO mess, short of discarding OO-ness and switching to the Relational Model.