> This is really important thing. On my opinion creation of visual
> components is an absolutely defferent task than creation of db engine
> components.

...

> but you agree that visual components are not the most important part
> in IBO.

There are issues relating to the interaction of visual components and
the dataset components often overlooked by database vendors. This is
where the native IBO controls out run other possibilities. They may be
visually less fancy than those provided by DevExpress, IP etc etc, but
they are setup to interact with the native dataset components more
fully than is supported by TDataset derivatives.

I dont know if the native control set originally came about because
they were needed for D2 or not BUT now the controls are a critical
part of all my applications - not just because they are needed to work
with the native datasets but because of the features provided by the
combination that I cannot get from TDataset and associated controls
(whereever they come from).


It should be obvious that your statement:

> If consider IBO visual components more effective then IBO MUST
> contain its own report builder components.

is false, since there are many of us with very complete applications
right now. The requirements for reporting are very different to user
interactive controls, as such there is less need for interaction
between report components and the dataset.

There is a pipeline to support ReportBuilder and there is capability
in FastReport to support native IBO datasets. So there was been no
urgency to write IBO specific report builder components - there are
good ones already available. And if you dont like these you can still
use the TDataset variations of IBO to access any other report builder
that you want.

--
Geoff Worboys
Telesis Computing