(sorry, this is a little lone)

... doesn't seem to me like the right way to go.

Do you know what an open source project is? Every read the cathedral
and the bazaar? It means there are people who, voluntarily for the
larger part, invest their own energy to work on a product and improve
it. They might even do so in their own right at first, without any
sort of coordination. I always understood that was the direction the
new InterBase was meant to go.

Now, the sources are there, there are minor things missing. They are
minor things because it'll take days or weeks, maybe months to
reconstruct them, but it'd take years to reconstruct a rdbms of the
quality and the tested reliability that InterBase has. Of course it
would be nice to have those things, but aren't there much more
important things to do?

The most important thing, IMO, is to set up a community. By that I
don't mean any fancy web pages but a base of volunteers who invest
their time. I also don't mean anyone who's already listening in on
this list. There are thousands of interested programmers out there who
have only heard of InterBase and what did they get as "open source"? A
chaotic source tree that'll build on no normal development system at
all, doesn't contain any docs at all and the architecture mailing list
is fuzzing about how to get the sources for some pdf files. So what
will they do?

Products like Apache or Linux itself are always used as arguments to
support any open source effort. But those products are successful
because there are loads of people actively supporting them. Of course
there are Apache consulting firms around, so a consulting firm for
InterBase might be successful, too. But they'll have a hard time
staying alive, let alone pay people for code development. Me, I'm not
going to be spending any more money on InterBase support than I ever
did before. That is, zip. That shouldn't sound like an offense to the
folks from whats-the-new-company-called-now, it's just a fact that
holds true even more for all those future OpenBase users.

Now, if everyone here wants to be part of an open source community,
why not get down to some real work now? There's been talk about
features already that could cost two men a year to implement. Yet,
there's still no build instructions, no updated make files, or
anything else available, although some people on these lists seem to
recall enough to provide some of that. Who do you think you can catch
this way?

Pardon me if I offended anyone,
might be my way of saying (or trying to) what I think,

Oliver Sturm

--
God is real, unless declared integer.
--
Oliver Sturm / <sturm@...>

Key ID: 71D86996
Fingerprint: 8085 5C52 60B8 EFBD DAD0 78B8 CE7F 38D7 71D8 6996