Tell me again why people insist on using over priced garbage like PervasiveSQL instead of PostgreSQL?

I suppose I was just using it as an expression. There is a good reason: TAS basically only supports btrieve. You know, that database access product from the 80's that some people are still torturing themselves with. There are TAS versions which support SQL now but I'm pretty sure it's the one ISTech isn't using. Quite frankly btrieve is an idea that made sense 20-25 years ago and quit making sense about 10-15 years ago.
 
I wanted to write a wrapper around btrieve calls and translate them to postgresql. I even got some of it done. But the btrieve API is so old fashioned and stupid that I couldn't bring myself to go through with it. Essentially all records are stored as one big binary blob. I'd have to store them as blobs in postgresql and then nobody would be able to directly access any information with SQL. Oh, I could try to reverse engineer how btrieve stores all its data in the blob but that would slow down database access. There might be some way of storing it as a blob but with metadata about the structure and then create triggers or something and maybe store the info twice, once as a blob for btrieve and then again as seperate data. Sounds like a mess doesn't it? The only reason I'd do it is to spite Pervasive for dragging this dead horse into the 21st century... But that's not that good of a reason.