> ... and if it is a little bit more modular and extensible it would take over the world. Been there, done that, got the T-shirt. I don't disagree with any point, and might add that all of the points have been made in one way or another, but possibly several posters, over time. They are definitely not wrong. Possibly two of the biggest advantages of SQLite are: - 1. The actual "lite"-ness, and - 2. The local-ness of data. Note that Both these points have severe drawbacks, especially when measured against other databases, but that's ok, it's drawbacks we embrace because it allows the advantages that come with the same. Put differently: SQLite can go places and do things none of the other engines can, and yes, by that same token there are many things it will never do as well as even a mediocre client-server system. It's a different focus, by design. Having said that, you are still not wrong, a lot of the described items could certainly make it into SQLite with good effect. Some of which may start to encroach on the "Lite"-ness, but I think the biggest reason why lots of it are not implemented is simply a question of time vs. need vs. utility. Nobody questions the usefulness of a Helicopter, but there is a reason not every household has one.