> I stepped thru the code and it seems that the internal dataset is
> not specifying the fields explicitly in the sql statement it's
> making. If it did that then we get away from writing the sqls with
> required fieldnames :-) (sounds nice)

TIBOTable is doing a "SELECT * FROM ATable", it does that for
compatibility with TTable.

Unless you are completely certain that all fields are required then
"SELECT *" is wasteful (both in terms of network overheads and server
and client resources). It is exactly this sort of thing that IBO
tries to encourage you away from.

I might also add that I think there are better possibilities than
those offerred by SQL DBMSes, but the fact is that that is what we are
stuck with (until someone pays me to write a better RDBMS ;-).

IBO offers some of the best facilities available to try and force
SQL/Client/Server into something useable at the client. But you wont
be able to take maximum advantage of these capabilities unless you get
away from TIBOTable.

--
Geoff Worboys - TeamIBO
Telesis Computing