>Done so in Dabo, by Ed Leafe and Paul McNett... I felt I had to do only one optimization - build the seek() lists for >your cursor on requery, not every time you have a .seek() call. IOW, instead of building an index on demand, >build it on requery. Requires some planning ahead (or not, you just add the field's name into the list of >indexables).

>And their bizobjects can do joins on each other... Well, when foxers write something, it has to have the good >stuff, right?

I use Postgres with Dabo and have had nothing but success. In some ways I think Dabo's data access is better than Fox's data access. For one it is simple to access different databases. No special data tools - access MSSQL, MYSQL, POSTGRES and Oracle. It all works the same. Which means I can (and I have) changed the backend from MSSQL to MYSQL without changing a line of code. Just change the connection type in one file. For the in memory data Dabo uses SQLite. And yes just as Dragan said, you can index SQLite - make subqueries - do anything SQL will allow. Need to take advantage of something special use direct SQL statements.

To this day I haven't found anything better or simpler to use when accessing data. The builtin CRUD stuff makes Dabo a pleasure to use. Just like Fox used provide many of us.

Like all frameworks there are issues. Mostly, stuff that I don't use. There is a WYSISWYG designer that I haven't used in many years. Don't use the command line debugger very often. Then there is the issue of keeping up with the changes in python and wxPython. There are other issues but I'll say that I am more productive with Dabo than I ever was with FoxPro.
Johnf