Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 15 Jan 1997 21:43:14 -0500
From:      "Brian J. McGovern" <mcgovern@spoon.beta.com>
To:        hackers@freebsd.org
Cc:        webmaster@spider.parasoft.com
Subject:   RE: Development and validation tools...
Message-ID:  <199701160243.VAA01304@spoon.beta.com>

next in thread | raw e-mail | index | archive | help

(Note to Webmaster@parasoft - This is in response to Jordan Hubbard posting
about your companies request for interest in its profiling tools for FreeBSD.
I thought it might be good reading for whoever is handling user responses to
their queries. Could you please forward this to them? Thank you. - BJM)

Jordan,
	I suppose I could just as easily send this email to them as to you,
but I think the mailing list is a good sounding board for anyone considering
writing commercial code for FreeBSD (or any OS for that matter).

	The problem of commercial software becomes the great chicken and
the egg problem... People who write software don't want to create a new product
for a platform, until they know they have a market for the product. The problem
is, they won't have a market until the product is done. 

	As a good example, in many cases, I've never been able to wedge
FreeBSD in to a commercial application, simply because most managers feel
its "not a commercial grade product". Unfortunately, since no one will give
it a try, it will never become one because "no one else is using it, so
how good can it be?". Later today, we'll also all be shooting ourselves
in the foot, because everyone else IS doing that. Fortunately, I finally have
a manager who doesn't have his head where the sun doesn't shine, so hes giving
it a chance. Of course, he has high demands of what he expects from it. 
Fortunately, its been able to perform so far. Anyhow, I digress...

	I can not approach this company and say "I'd buy three of your
packages", because I can not approach my management and say "I'd like
budgetary approval for three copies of a software package that doesn't
exist yet". I can't get approval to buy it until its done, and they won't do 
it till they have enough people to buy it. I smell Mobius(sp?) loop here...
(not only does it keep going around in circles, but its single-sided, as well)

	As a more positive example, I think Maxis did the right thing when
they ported SimCity to FreeBSD. I liked the game. The only downside was that
it was a few years too late. I had exhausted my SimCity days on my PC/XT
years before... Right now, I still have the floppies in my 3.5" case, but
all they're doing (and will do for the forseeable future) is to collect
dust.

	Now, I can say that if Maxis were to release a new game that I haven't
played to death, preferably something multi-player that runs on a 486/33
or faster, and cost between $20-$30 per license, I know I'd buy at least
two, probably three, right off the bat. Plus I'd probably try to sell it to
my friends, so we could play over a modem link or the Internet. My friends
and relatives all love multiplayer games, and we always but a ton of licenses
when a good one comes out.

	As to the other people that are stepping forward, I would evaluate
their software as any other. Currently, I use CVS for source control, ddd for 
debugging, gprof for profiling/speeding things up, X with a ton of Xterms
for my development enviornment, jove for my editor, and a handful of other
utilities that I can get for free to do my development. And I usually come out
with a good product. Now, if they had a working product, and they could show
me why their product is any better than what I currently use, I'd ask for
approval to buy a copy for eval. If it then lived up to its claims, I'd 
probably buy a half dozen or so to equip our department. But, until I have
a product, I can't go any further than to say "Well, I'd _like_ to see it".

		-Brian



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199701160243.VAA01304>