Re: [xtm-wg] A challenge on "the graph"

[Lars Marius Garshol <larsga@garshol.priv.no>]

* Sam Hunting
| 
| Agreed. On reflection, it seems to me that the node/properties
| approach, IFF it is identified with an OO approach (an object seen
| as a bundle of properties) may seem intuitive to some developers,

The spec should not, I believe, identify it with an OO approach, as
that would require implementors to choose a particular approach. In
any case, doing so would not affect the intuitiveness of the approach
in the eyes of developers.

| but shuts out:
| 
|      * RDBMS approaches (eg, Jack's)

A nodes-with-properties approach does of course NOT preclude RDBMS
implementations. The idea that it does is mistaken. Translation from a
nodes-with-properties model to an RDBMS model is fairly
straightforward: all the columns you will need are in the model, so
all you have to do is work out what tables you need.

Jack Park, do you see any difficulties in using the infoset-based
approach as a guide to an RDBMS implementation?

BTW, Ontopia also has an RDBMS-based implementation.

|      * non-OO programmatic approaches (eg, those who deal with huge
|        graphs and don't want the overhead of objects)

This is not ruled out either. As long as all the information in the
infoset is available, such an approach would work just fine.

In any case, those who deal with huge graphs would do far better to
use a database.
 
| Only a graph representation can encompass all these representations,

That is quite simply not true. It would be just as true to say that a
graph-based spec would preclude object-oriented implementations.

--Lars M.


------------------------ Yahoo! Groups Sponsor ---------------------~-~>
Do you have 128-bit SSL encryption server security?
Get VeriSign's FREE Guide, "Securing Your
Web Site for Business." Get it now!
http://us.click.yahoo.com/EVNB7A/c.WCAA/bT0EAA/2n6YlB/TM
---------------------------------------------------------------------_->

To Post a message, send it to:   xtm-wg@eGroups.com

To Unsubscribe, send a blank message to: xtm-wg-unsubscribe@eGroups.com 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/