Subject Re: Connectivity options. Was Re: Re: Digest Number 10
Author Jason Wharton
>From: "Rob Schuff" <robertsc@...>
>
>What would be gained by "merging" IBO and IBX? IBX is not mature. Do it
>have something that IBO does not have? Not that I am aware of. You can
>also be sure that Borland will continue to develop IBX. Since IB is going
>open source, development of IBX fits right in with their current game plan
>(ie, "Wake up and smell the Linux"). My guess is they fully intend to get
>the most marketing value they can out of the is, and frankly why shouldn't
>they? I suppose they could try to buy IBO.
>
>There will be no IBO/IBX/BDE fight. I'd be willing to bet the BDE is dead
>(though it will die a slow death). I wouldn't count on any future
>development of it. Wintel stuff will be OLEDB/ADO. That leaves IBO/IBX
for
>x-platform. IBO has a **huge** advantage on the wintel platform and I
>suspect Jason will want to maintain that edge by getting IBO ported as soon
>as he can. Hopefully he will be getting some beta copies so he can start
>porting early.

Another advantage that I have with IBO is that I am not bound to port only
to Inprise development environments.

If there is a PASCAL based free compiler (which there is) and I can move IBO
onto it you can bet that I will be right in there making it available to
that community (which I am doing).

Do you think InterBase NewCo is going to be as inclined to do this? Probably
not. IBX is bound to the VCL data access layer so they are not in a position
to take Inprise code and port it to another language. IBX will be Kylix
only, count on it.

How well do you think that the open source community at large is going to
receive InterBase if they can only use it via. Inprise development tools?
This is bound to be an issue in coming months...

The Native IBO data access layer is built on top of TComponent and
TPersistent. I can move it anywhere I like! The question that remains to be
answered is how I choose to license IBO in the non-Windows platforms.

I am very confident that the LINUX community will like the IBO approach to
working with data. My challenge is in figuring out a way to present it to
them... I admit that I still have a lot of "culture-gap" to bridge over...

FWIW,
Jason Wharton