data cartridge development - workload?

Has anyone out there developed their own custom/proprietary data cartridge, using the extensibility interface?

I have been asked to look into this at work, specifically some specialised indexing.

My first observation is that the API looks big and hairy.

If anyone with experience would care to brain-dump on difficulty, obstacles, workloads and timescales, I'd love to hear about it.

After all the effort Oracle put in to creating that API, some *has* used it, right?

        BugBear (Hi, Nuno!) Received on Tue Aug 28 2001 - 10:34:38 CDT