LOL! C. Date was actually on to something there, I mean his arguments for the objects being analogous to data types (as opposed to the rows, columns, tables, or relations) is fairly strong and still very pertinent... at least I think so. Imagine using SQL to return an object set rather than a data set and I think that we are much closer to the right track for .NET.

Also (not entirely on topic), C. Date's description of a data type being able to be defined as anything, even a table... a table contained within the row column intersect (cell) of a table is very analogous (visually anyways) to placing the VFP grid control in a column of another VFP grid control. I remember the first time I realized I could do this in VFP, it blew my socks off.

The reason I said this final point is not entirely off-topic is that many of the objects that one would return from a query in an OO database model would contain members that were also objects.

It's fun to play around with it in my head anyways. Thanks for taking the time on your previous reply and this one. I share your penchant for learning new languages, but new languages day for me is Saturday, so I will wait until then to revisit one of the languages you mentioned. <g>

>>You're reference to Chris Date reminded me of an article I had read some time ago. It may be pertinent to this discussion and what Martin has in mind...
>>
>>http://www.dbmsmag.com/int9410.html
>
>Jeez, Craig... You're making me feel so old and rusty... 8-}