Hi John,

Thank for dropping in. I feel this project is worth it. You mentioned C++ and fll-s. The first and last time I coded C - K&R version - was 1992 and that was to build a couple of flls. But gosh I'm too old for this game now :-)

My intention is to consider duckdb as batteries for python in the way rushmore was for our dear fox. Of course it is REALLY processor-and-ram hungry stuff not the memory-thrifty engine that the fox team output at that time:-)!

>2 things I haven't been able to establish:

>1) Where is data stored - single file or multiple files?
The stuff is really data-centric and I have started to consider it as a replacement to our decision support stuff ie in-memory rushmore-augmented index VFP cursors. But you can of course persist it into a .db format?

>2) None of the examples includes credentials. Does it have encryption/security?
Good point. I have not found any support at the current stage. They focus on performance on very large data - the data-science size - at the current junction.

I have asked - see my first introductory UT thread on the duck - whether they could consider ODBC as a communication channel. Hannes was pretty responsive. He added immediately to the project. Do not hesitate to get in touch with the team on the subject behind this friendly open source project.

Cheers, Daniel