At 7:05 PM -0500 3/12/06, [EMAIL PROTECTED] wrote:
Let me amplify this by pointing out that I have zero interest
in taking SQLite in the direction of The Third Manifesto. Those
who want to do so are welcomed - encouraged even - to fork the
tree and go off in their own direction. Just do not expect me
to follow along, please.
Richard, let me ask you this, though, to clarify your position.
Would you be willing to build a distinctly new, and different-named
product, reusing what you can of the SQLite code base (that you know
so well) in the process, mainly the virtual machine and b-tree etc,
which implements a set of specifications I provide, if you are paid
your standard custom work rates for the time spent?
I just wanted to know whether the whole idea I was raising was
distasteful to you period, or whether you simply didn't want to make
SQLite itself work that way.
I should also clarify my own position, that I *have* started my own
project to implement the ideas I had. It isn't on Sourceforge, but
it is hosted on other public repositories and is released via CPAN.
The reason I am raising these issues in the SQLite community is that
I like SQLite and I think it provides a good and mature point of
departure for implementing a fast and efficient performing version of
what I am otherwise making on my own.
But moreover, that I was preferring to pay someone else who was a lot
stronger in C than I am to do this work, rather than doing it myself.
All this said, I think I will shelve this discussion for now. Then,
when I actually have a working implementation of my proposal, I will
draw your attention to it later to see if anyone is interested in
seeing how things actually work out in practice; it may not be the
same as you preconceive.
-- Darren Duncan
