Jim Starkey wrote:
[...]

> Attached are the RC4 transform, an abstract "cipher" transform
> for polymorphic ciphers, and an update of the AES transform
> with optional CBC (cipher block chaining) to make Geoff happy.

If you think CBC will "make Geoff happy" then you obviously
understood almost nothing of what I wrote ... but then that
was pretty much what I expected.

I have no problems with a good implementation of RC4 as a
stream cipher, but I do not understand Firebird's requirements
well enough to know whether the properties of some of the
alternatives could be desirable. I suspect that WEP's problem
was having a developer that didn't bother to learn about the
best way to implement encryption before starting to use it,
and that does sound rather familiar doesn't it.

--
Geoff Worboys
Telesis Computing