On May 14, 2003 05:59 pm, Artur Anjos wrote:

> Paul,
>
> > Hmmmm, theres an idea for the Firebird project to make some extra
> > cash, a CD with Linux O/S, just enough Linux to make the OS usable,
> > no compilers, no GUI, leave out all of the usual junk that comes with
> > it, plus Firebird, and maybe webmin for administration.
>
> This remind me that a webmin module for Firebird is something that
> could help us. A simple module for a start, to allow the user to
> start/stop the server, changing firebird.conf parameters, and so on.
> One of small things I'll like to see at add-ons tree. :-)

Agreed..... Hmmmm.... FBManager hasn't really sorted itself out yet,
perhaps it could consider this direction. I'll take it to the group and
see what they say.

>
> We must expect much more web-developers with version 1.5. The increase
> of speed in connections will help us on this.
>
> I don't know about the market for this, but I think a lot about it. I
> have lots of experience with NAS (Network Atached Storage) devices [my
> company distributes SnapServers for some years now], and I always ask
> about the market for a (maybe inventing the expression now) NAD device
> - a Network Attached Database. A black box with a small operating
> system in rom, Firebird included, optimized for database use.

A NAD device, I don't know, I think that what we should have is a
program, it creates an image file, this image could be placed on a
floppy, CD, FlashROM, HDD partition or other bootable device. When the
system boots, it loads a kernel and decompressor, we create a RAM disk
and boot to the RAM disk, which contains only enough Linux (or FreeBSD)
to boot, and run FB1.5. System builders and VARs can then get the boxes
and use the program to build the system. The reason for staying out of
the box business, is simple. Who's market, we have folks from all over
the world dealing with Firebird. I would never pay US prices for
anything computer related. I can often buy the same thing in Canada for
the same price as in the US at that is with a 30% smaller dollar!

> I think Firebird will be the only *possible/reliable* solution for this
> kind of device if there is a market for this. Of course we will need
> other things, like a very good ODBC driver.
>

I agree, our ODBC drivers need work, for one thing there are too many
different drivers, and none of them work as well as they should, the
commercial ones work very well, but defeats the purpose of having a free
database, when you need to pay for the drivers. The free ODBC drivers,
while improving, still are not good enough. They need to merge those
into a single project at some point, so that everyone working on a free
ODBC driver, are working on the same one.

Paul S