In Firebird-general@yahoogroups.com,
"paulruizendaal" <pnr@j...> wrote:

> Why do write that FB/IB has a market share of only 2% ??

(this has turned into a bit of a novel,
if you don't like literature, skip to the next post!).


It was a figure that I saw mentioned/referred
to on a site - my apologies I can't
remember or find the URL. That would be share
of the *_commerical_* database market - i.e.
compared to the likes of Oracle, M$, IBM, Sybase,
Informix and others. It was something to
do with Interbase - as distinct from FB.


As regards FB, I've seen a lot of rubbish written
about "market share" - people (supposedly journalists
who should know about their topic, or at least have
done basic research), can come out with utter
crap.

I found this gem at

http://www.osnews.com/story.php?news_id=10498

___________________________________
Many databases are available for the mid-market,
but only five have an installed base of 1 million
or more and have made a commitment to the mid-market
user: Oracle's Standard Edition 1 ("SE1"), IBM's
DB2 Express , Microsoft's SQL Server and two
open-source products, MySQL-5 and Firebird.
____________________________________


I'm sorry, but where the hell is PostgreSQL?
Informix? Solid? I'm too tired now to search for
all the alternatives, but if ever there were
two db's for the "middle market", they would
be PostgreSQL and Informix.


Same article
______________________________
Installed base is of course not relevant as a feature
of the software product itself, but it is important
as a measure of how real people have made real-world
choices, based on features and cost.
______________________________________

The above I agree with.

> See research by e.g. Evans, that says 28%
> in a survey with a skew unfavourable to FB/IB:
> http://news.fyracle.org

You're trying to tell me that 28% of
software developers use Firebird. Sorry, but
that just won't wash. Quite frankly, I would
be surprised if 28% of software developers
had even *_heard_* of Firebird.

Many of these so-called "surveys" ensure that
the zealots of a particular product who have
the time to waste vote 100's of times


I have only once ever seen a job advert
which mentioned IB/FB - I got the job, only
to quit two weeks later (company issue!). And
I'm not just talking about jobs here in Ireland,
I regularly cruise jobsites abroad (am considering
moving), so I'm not entirely unknowledgeable.

> Oracle, DB2 and SQLServer (excl. MSDE) have
> an installed base of about 1 mln installations
> each. FB/IB has an installed base 3 times that size.

Let's boil the problem down. How the hell do we know
this? Oracle, IBM and MS are at least publically
quoted companies who have a legal obligation
to provide honest (at least as far as
accountants are concerned) reports to the
SEC (or whatever it's called in Merka). Mind
you, so does Borland, and try guessing IB figures
from their filings.


One can then glean from such figures (perhaps)
an *_idea_* of how many deployments of x,
y and z they have - does MS split its revenue
figures for Widoze, SQL Server and Office? Does Borland
split Delphi, IB and JBuilder? Don't think so!

> A recent survey by Carlos Cantu et al. covered
> 100,000 installations.

Give me an URL and a reference as to how they
got these "100.000" sites together. How
much self-selection goes on? "Oh yes, I'll
answer your survey, and pass it to my
colleague..."

> That is 4 times more than all Sybase
> installations in the world.

Take a look at www.dice.com, which AIUI, is
an American job search site - search for

Oracle 12951
Sybase 2067
Interbase 3
Firebird 9

Sorry, but those figures are telling me
something about installed base and
importance and market share of product.
And it doesn't all have to do with
difficulty of admin either.

> Perhaps you should broadcast to the Oracle
> newsgroup that FB/IB has *not* failed to
> hit the big time.

Sorry, take a look at the numbers - real
ones, not from vague "internet surveys".

Send out a questionaire - "How many times
did you download porn last month", and you'll
get precious few responses above 0. Point:
Many surveys are meaningless.

> Borland failed to make a shitload of money

True, which if you read the URL that I posted
in the comp.databases.oracle.servers thread
you will understand *_my_* analysis as to why
they failed to do that.

Oh, and BTW, if IB (and I mean IB *_before_*
the birth of FB) was doing even moderately
well, what prompted Borland to Open Source
it, in particular? There was no talk of
Open Sourcing Delphi or JBuilder. The only
rational guess that one can make is that
IB as a division of Borland was (how to
put it kindly) "struggling" - maybe paying
its own way, but not putting back the revenue
into the company which merited its
upkeep.

*_My_* interpretation of the aftermath of
the open source debacle, is that a number
of important Borland clients said, "Right,
if you drop IB (into which we've put years
along with using your dev tools to work
with it), we're dropping Borland altogether.

Having said that, I think (or at least get
the feeling) that IB lost clients over the
OS issue - before, they would have no
problem quoting high (or relatively high)
profile clients (some American tank,
a Stock Exchange or two) - I have tried
to get similar info in the recent past
and failed. One would have thought that
they would have been only too glad to
funish high profile client details.

> -- which is something
> quite different and is only good from a
> user perspective.

Mozda da, mozda ne. Tallán igen, tallán nem.

Maybe yes, maybe no.


If they had made a "shitload"
of money, maybe they could spend more
promoting what, for my money, is the most
elegant db solution out there.


Believe me, I think that the FB project has
real potential to produce an elegant super
db solution for the "middle market", and if
the project leaders were prepared to organise
better in terms of instrumentation (a
l'Oracle), they could move even higher, but the
figures for usage you have quoted are nothing
short of ridiculous.

We'd be better off being honest, rather than
trying to kid ourselves with "surveys" that
purport to show that IB/FB are taking over the
world. The project should work on reliability,
consistency and instrumentation. My problem with
the FB project is that not enough attention
is being paid to the latter - even IB is now
starting to go down that route - killing
queries, regulating transaction times. Stuff
that is needed, but doesn't seem to be on
the event horizon.


Sorry for the long rant, but I felt that there
were things that I needed to get off my
chest.


Rgs.


Paul...