Return to BSD News archive
Newsgroups: comp.os.linux.advocacy,comp.unix.bsd.freebsd.misc
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.mel.connect.com.au!munnari.OZ.AU!spool.mu.edu!howland.reston.ans.net!ix.netcom.com!netcom.com!kevinb
From: kevin@frobozz.sccsi.com (Kevin Brown)
Subject: Re: Linux vs FreeBSD
Message-ID: <kevinbDJInpr.GL0@netcom.com>
Sender: kevinb@netcom7.netcom.com
Reply-To: kevin@frobozz.sccsi.com (Kevin Brown)
Organization: Frobozzco International
References: <489kuu$rbo@pelican.cs.ucla.edu> <4aajus$nd@dyson.iquest.net> <kevinbDJGqtq.Fup@netcom.com> <4akkkj$14b@dyson.iquest.net>
Date: Wed, 13 Dec 1995 08:32:15 GMT
Expires: Wed, 13 Dec 1995 08:00:00 GMT
Lines: 104
Xref: euryale.cc.adfa.oz.au comp.os.linux.advocacy:30282 comp.unix.bsd.freebsd.misc:10430
In article <4akkkj$14b@dyson.iquest.net>,
John S. Dyson <root@dyson.iquest.net> wrote:
>In article <kevinbDJGqtq.Fup@netcom.com>,
>Kevin Brown <kevin@frobozz.sccsi.com> wrote:
>>
>>End result? In the free sector, the compiler ends up evolving
>>primarily as a result of the needs of individual consumers who
>>contract someone to implement the features they need, and therefore it
>>acquires the features that the consumer group as a whole desires.
>>Otherwise they buy the commercial version and deal with the well-known
>>consequences of that. This is likely to result in a free compiler
>>that keeps up well with the commercial versions in terms of truly
>>needed features.
>>
>Except one thing -- assumptions are being made that the GPLed compiler
>is "good enough". It is true that the gcc compiler is pretty good --
>but compare it to a Micro$oft compiler on an X86. It is amazing what
>that compiler can do... I doubt that anyone would want to spend the amount
>of research and money required to make gcc do what the Micro$oft stuff
>does (if they need to make it public.) For example, look at the intel
>Pentium patches -- they help a bit, but don't really do that much
>good.
Certainly the GPLed software in question (the C compiler in this case)
has to be "good enough"...otherwise people won't bother to use it.
And you're right, my arguments do depend on the free software being
"good enough".
Consider, too, that the same condition holds true for the BSD version.
For a company to use the BSD version as a base, it has to be "good
enough" sans a few needed features. When that condition is met, the
company can take the BSD version and hack it until they get what they
want out of it, and then create a proprietary distribution from that.
A customer who needs a compiler and finds that the GPLed version isn't
"good enough" can, if he so desires, hire someone to *make* it "good
enough", provided he decides it's worth the expense. Again, it
depends on the alternatives.
Either way, the free software, be it under the BSD copyright or the
GPL, must be of reasonable quality to start with, otherwise nobody
will bother with it (even in the free sector!).
>It is my position that GPL hinders in many ways as much as it helps. In
>fact, if I was creating a start-up company -- I would like to keep
>things private as long as I could, if I really had something to sell.
Right. That's because the GPL doesn't work well with the
software-as-a-product model. The consequences of that are that
companies who are interested in making money from the GPLed product
become service providers, rather than a provider of a product (in
actuality, they also provide the GPLed product, but that's not what
they make their money from).
>It is just good business. In fact, I could release the source under
>contract (or put it in escrow) so that my customer is not left hanging
>if I go out of business, but it protects me from another startup.
I have to wonder how many software companies actually care enough
about their customers to do that...
>I see that
>GPL can hinder startup businesses significantly. Yes, there are companies
>that make their business supporting GPLed software. But again,
>compare gcc vs Micro$oft C V4.0... For being proprietary and costing about
>$200-400 (I forget how much), it isn't all that bad...
This may be true...I can't really comment on that, since my experience
with recent M$ compilers is essentially nonexistent (the last M$ C
compiler I used sucked dead bunnies through a straw. Has it really
improved *that* much?).
In any case, the companies doing "software as a service" are probably
fairly small in number and small in size. Cygnus is probably the most
visible one.
It's likely to take time for GPLed software to really catch on. I
realize it's had quite a bit of time, but the business world moves
really slowly.
Then again, perhaps nothing will change, in which case we'll be left
with the same choices we have now.
>If I had a
>revolutionary idea (fat chance :-)), it would *very* unlikely be embodied
>in a GPLed document. And if the software had to be GPLed, I probably would
>not waste the effort to do it (If it was complex or expensive in time to
>do.)
Fortunately, there's nothing that forces you to release a completely
original work under the GPL...it's your choice.
But if you were doing a revolutionary new product, why would releasing
it under the BSD copyright be any more appealing? Chances are, if
you're interested in keeping it to yourself or making money off of it
as a product, you'll put it under a strict copyright just the way most
software vendors do...and free software becomes a non-issue at that
point.
--
Kevin Brown kevin@frobozz.sccsi.com
This is your .signature virus: < begin 644 .signature (9V]T8VAA(0K0z end >
This is your .signature virus on drugs: <>
Any questions?