Re: What about an OS monopoly under GNU license?

[Brett Glass <brett@lariat.org>]

At 11:23 PM 3/26/98 -0500, P.A. Petricone wrote:

>Wait a sec, I am amused by the thought that MS may have pinched some GPL
>code. More seriously, I have a growing problem with compiled software and
>no source. Source should be as available to the consumer as a shop manual
>or a house plan.

It'd be nice! But even shop manuals for cars don't go down
to the same level that source code does; it'd be the 
equivalent of providing detailed manufacturing instructions 
for every part. 

>:As in laws in which the true agenda is not made explicit, this
>:portion of the "preamble" does not explicitly state the anti-consumer
>:intent of the license. Note that it sidesteps the issue of the
>:"ACI-freedom" to reuse the code on one's own commercial software.
>:
>
>Yet it does not forbid it. 

Yes it does; in fact, it's quite explicit. You must make the source
available for free -- in other words, you must give away your product.
In which case, you are no longer selling software.

>I would guess, without the actual numbers, that there is a LARGE
>contingent of commercial applications developers worldwide working on GNU
>and like platforms who's startups may not have been possible if their only
>option for a platform were proprietary UNIXs and who most often give back
>improvements to GNU software for others traveling the same road.

There are also a large number working on platforms that use the Berkeley
license. I'm not sure how the numbers compare, but in general the BSD-derived
systems are more often recommended for mission-critical networking applications.

>Hmm... Caldera, RedHat, Infomagic, Walnut Creek, etc, etc... I guess they
>are all wasting their time. 

When these companies publish versions of Linux, they are only charging
for the service of recording the software on physical media. They may
have commercial add-ons for it, but that's another matter.

>Well then I guess we should all pack it in and subscribe to the MS model -
>much less oppressive. I think you are reading more into GPL than is really
>there in PRACTICE. It seems to me that FSF is an effort to provide and
>maintain the structure of a software development environment free from
>commercial interference. 

I would hardly consider using the code to create innovative new products
"interference!"

>As such, it meets it's objectives. While you are
>there, you are welcome to stay and use what you need as long as you support
>the objective of where you are. 

Commercial software developers find themselves most unwelcome in the GNU
environment. This is why it is important to promote the use of other
licenses which, in turn, promote commercial software development rather
than discouraging it.

>:Note the copious use of the word "free." Commercial software vendors
>:need not apply -- period.

>Parsing this a bit, one example I see is that you could, with the author's
>permission, revise their code to run on a BSD platform, thus moving it to
>the free-er license. 

Only with the permission of EVERY author -- providing that you could find
all of them and they all agreed. A better approach is to use a more
permissive license from the get-go. This would promote the development
of commercial software. The GPL discourages it.

>Also FSF claims it is approachable for the same. 

Again, not if they believe that the software might then be used
to produce a commercial product. Just try calling and asking! You
will get a decidedly cold shoulder.

>So we should blast FSF because their name says it all. They are just
>another repository for folks who need facilities like public libraries to
>get something done.

Nope. They're akin to a public library that insists that if you

realize some benefit as a result of reading a book there, 
you must give them all of that profit or benefit.

>The GPL states without question that anyone has the ability to contact the
>author regarding re-use of code.

Indeed. However, they don't make that very feasible. Again, code may
have DOZENS of authors -- often does, in fact. And they're often not 
all listed.

>Come now, you are asking for a hybrid economy - free to you, +ACQ- to me.

No, it is the FSF that is asking for a "hybrid" economy -- free to individuals,
inaccessible to those who would like to earn a livelihood by publishing
software.

>:In the interest
>:of consumers, these authors should be encouraged to license their work
>:under the least restrictive terms possible. Ironically, many parts of
>:Linux were in fact released under the Berkeley-style license. The
>:developers of Linux could not have appropriated the code were it not
>:for that license.
>:
>
>Does that apply to the current kernel? Here is the 'Linux' header from the
>GPL
>
>--------------snip---------------
>
>   NOTE+ACE- This copyright does +ACo-not+ACo- cover user programs that use kernel
>services by normal system calls - this is merely considered normal use
>of the kernel, and does +ACo-not+ACo- fall under the heading of +ACI-derived work+ACI-.
>Also note that the GPL below is copyrighted by the Free Software
>Foundation, but the instance of code that it refers to (the linux
>kernel) is copyrighted by me and others who actually wrote it.
>
>   Linus Torvalds
>
>------------snip----------------

>From your comments above, it appears that you do not understand what is meant
by Linus' note. "Using" kernel services means calling the OS from a program,
not incorporating code from Linux code into one's program. The two are very 
different.

>:But the point is irrelevant to the discussion at hand.
>
>I think the point is just as irrelevant as your John Lennon/Stallman
>remark which was uncalled for. 

Actually, I believe it is quite germane. Even Richard would most likely
agree that it characterizes his attitudes regarding intellectual
property quite accurately.

>:What we're concerned about here is the promotion of
>:the license which will, in turn, do the most to promote innovation and
>:competition in the industry.
>:
>
>I was not aware we were designated to make that consideration.

We all can. And it's especially important that the CPT do it.

>Suppose for discussion BSD was the global license model. There remains for
>me problems of sustainability. How do you propose we select over time those
>that will be committed to writing all the carefully polished free code
>components for you to build commercial apps - lottery? ...

There's no need to "select" them! Many people are working on and
releasing code under Berkeley style licenses right now. The use of such
licenses, rather than the restrictive GPL, should be encouraged.

--Brett Glass