Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 08 Sep 1999 01:55:39 -0600
From:      Brett Glass <brett@lariat.org>
To:        "Jordan K. Hubbard" <jkh@zippy.cdrom.com>
Cc:        freebsd-chat@FreeBSD.ORG
Subject:   FreeBSD market share statistics
Message-ID:  <4.2.0.58.19990907234944.047c0980@localhost>
In-Reply-To: <12874.936232439@localhost>
References:  <Your message of "Wed, 01 Sep 1999 16:43:10 MDT." <4.2.0.58.19990901152642.047b0250@localhost>

next in thread | previous in thread | raw e-mail | index | archive | help
At 05:33 PM 9/1/99 -0700, Jordan K. Hubbard wrote:

> > >   Is the notoriously well clued-in
> > >    Gartner Group's (NOT) claim that Linux is the only non-Microsoft OS
> > >    to gain marketshare any more credible?
> >
> > Yes. They're paid big bucks and are in a fiercely competitive market
> > where customers demand accuracy.
>
>Haha.  You have a lot more trust in people who are paid big bucks than
>I do, clearly, and also are assuming that their results aren't simply
>OLD at this point.  When was that report generated?  What figures were
>the person or persons generating the report using at the time?  It's
>quite possible for highly skilled, well-paid people to make mistakes
>even with the best of intentions or we wouldn't have lost a shuttle
>and the U.S. government would operate like a swiss watch (one could
>correctly deem international politics to be "fiercely competetive" and
>the US government is certainly paid some of the biggest bucks of all).
>And no, this isn't simply a straw-man argument, this is a direct
>refutation of your claim that being in a competetive market and being
>paid big bucks is some sort of instant ticket to Objective Truth(tm).

As Petruchio says in Shakespeare's "The Taming of the Shrew," "Thou
know'st not gold's effect." If The Gartner Group loses a single client
due to an inaccurate report on its market segment, that's tens of
thousands of dollars (maybe millions) out the window. This is a very
good incentive to be thorough and accurate; it also means that the
resources are available to compile an accurate report. Gartner has
a very good reputation for coming up with good numbers with which
one can successfully gauge markets and predict trends. I daresay
they're a more reliable source than Walnut Creek's download
numbers, which are skewed in favor of FreeBSD and certainly
exceed the actual installed base.

>...I don't
>think you're even bothering to try and look for data beyond what's
>already been chewed up and regurgitated by the likes of the Gartner
>Group.

Not so. I've simply cited the best numbers I've found -- including
David's OWN numbers, which support my analysis.

>David asked for numbers and so far all you've given us is
>somebody's canned report and a single URL from the notoriously
>inaccurate "queso" stats (and for more info on why I think it's
>inaccurate, do a slashdot search for "queso" and read up on the
>flames).  That's not citing stats, that's looking for one or two
>numbers which fit your agenda and then stopping there.  It's poor
>science and it's a poor argument.

As I've pointed out elsewhere, "queso" overestimates the number
of BSD systems and thus produces results which are skewed
in BSD's favor. Yet, even given that handicap, the BSDs are shown as
falling behind.

> > Even David G., whose projections of FreeBSD's growth seem to me to be
> > overly optimistic, admits that the gap between Linux and FreeBSD
> > is growing. On April 15th of this year, he wrote:
>
>Now here's an issue we can actually agree on.  OF COURSE there's a
>growing "gap" between the Linux and FreeBSD camps - I've pointed out
>many times that it's definitely the Linux people who are currently
>enjoying their 15 minutes of fame and given that they're in the
>spotlight, only a fool would suggest that they're going to get the
>most hit points right now.  That "gap", however, is not the issue.

I disagree; I think it's an important issue. It's a predictor of when
we'll see, for example, native ports of commercial software such as
WordPerfect Office, TripWire (the up-to-date version), Borland Delphi,
CodeWarrior, etc. for FreeBSD. The manufacturers of all of the above
products say that, due to the gap, they have no plans to provide
native FreeBSD ports. And what about such mainstays as TurboTax?
For Linux, one day, maybe. For FreeBSD? Not unless it catches up.


--Brett Glass


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-chat" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4.2.0.58.19990907234944.047c0980>