Re: Copyright Misuse Doctrine in Apple v. Psystar

[Alan Mackenzie]

In gnu.misc.discuss amicus_curious <ACDC@sti.net> wrote:

> "Alan Mackenzie" <acm@muc.de> wrote in message 
> news:go0t2l$12pe$1@colin2.muc.de...
>> In gnu.misc.discuss amicus_curious <ACDC@sti.net> wrote:

>>> "Alan Mackenzie" <acm@muc.de> wrote in message
>>> news:gnv0cu$28je$1@colin2.muc.de...
>>>> In gnu.misc.discuss amicus_curious <ACDC@sti.net> wrote:

>> Just as an analogy.  Tax swindlers (whether legal or not) take without
>> giving, a bit like "code swindlers", who take open source code without
>> giving their bit back.

> Well, that is hardly equivalent.  Someone cheating on taxes is working
> with money which has a common valuation and some rather strict laws
> with clear metrics.  Someone using a GPL program is in no way obliged
> to do anything in return and 99.99% or more do not, if you consider,
> say, how many people have actually contributed anything to Linux versus
> how many use it directly.

No, the two aren't equivalent, they're analogous.  Anyone may use a GPL
program; what she may not do is make enhancements and distribute them as
her own.  That's the principle, grounded in practical experience, that
allows free software to exist and flourish.

>> When they take commonly owned source code, built by the labour of
>> thousands, yet aren't prepared to give their bit back?  People like that
>> have never been much loved.  Imagine a village festival where everybody
>> brings their produce in for a massive feast.  Now some people won't have
>> anything to give, but they'll be there with their neighbours.  Fair
>> enough.  But a prosperous farmer who is too mean to contribute, yet comes
>> to the feast anyway won't be liked, will spoil the occasion, and might
>> even be excluded.  This is analogous to the person who adds his bit onto
>> free code, presumably to make money, but won't contribute it back to the
>> project.

> There is a big distinction, I believe, in terms of the value of that source 
> which you want "given back".  If someone has merely fixed a few defects in 
> the original, it is reasonable to do that and the person is most likely 
> motivated to do so.

There's nothing "mere" about fixing defects.  It's a highly laborious,
highly skilled task, as well as often being mind-numbingly tedious.  If
it were "mere", Microsoft would have fixed nearly all of its bugs.

> If someone invents a truly useful functional addition, which they would
> have to do in order to make any idea commercially viable in the first
> place, that is more in line with a new work rather than a logical 
> extension of the original regardless of the way that the original might
> have been incorporated into the new work.

Not really.  That's not the way software development works.  You don't
get brilliant artists effortlessly putting the magic icing on the humdrum
cake laboriously put together by dull drudges.  That's an
over-romanticised picture, which might have been true, a bit, in the
1980s but certainly isn't now.

Software development is pure drudge work, with a tiny amount of
brilliance in the mix.  It takes far more creativity to produce the cake
than the icing.

> Your analogy suffers, too, in that the FOSS festival is advertised as
> "bring a dish if you want to share, but come one, come all for the
> feast!"  Perhaps the prosperous farmer only wants a dab of mashed
> potatoes to put on his own plate of prime rib that he brought for
> himself, and, as soon as he does, everyone demands a piece of his meat.

Possibly so.  That farmer, by participating, will earn far more in status
and goodwill than his cost in meat.

> I don't really see much danger in any of that happening, as I have said
> elsewhere, but I think the philosophy of the GPL is wrong at its core.
> If FOSS were about sharing, that is something that can be supported,
> but it has become about donating and with the court actions it is
> coming closer to the taxation of which you speak.

I'm not sure what you mean by "wrong" here.  The GPL works.  It persuades
tens of thousands of people, including me, to contribute software which
is generally useful.  Licences like BSD don't.  Not much.

>>>> No, actually, OS/X isn't dependent on Apple's hardware.  It will run
>>>> happily on a Psystar box.

>>> But not legitimately, I understand.

>> That's under contention in an American court at the moment.  Should it
>> be legitimate for a software manufacturer to dictate what sort of
>> hardware his software package may be run on?  I don't think it should
>> be.

> It is a simple matter of contracts.  You are not allowed to use a $17
> DVD to show a video picture in a theater and charge admission, for
> example, or are not allowed to use a 99 cent track from Apple and
> broadcast it on a commercial radio station.  If Apple wants to restrict
> the use of their product to specific circumstances, they have that
> right to do so.  The GPL has the same right, of course, and it is only
> a matter of what civil action might be taken in the case of a
> violation.  With Apple, they can show that they lose a certain amount
> of trade in their own hardware and that has a value to be assessed
> against Psystar.  With the GPL, it is much harder to show that any
> damage has actually accrued to the copyright owners.  We get into the
> arena of hurt feelings and some loss of prominence rather than any 
> financial impact, which is the traditional way of assessing damages and 
> compensation.

It's anything but simple.  If I buy a DVD, I can play it on whatever DVD
machine I like - I'm not forced to buy the movie company's super delux
offering.  Yet with OS/X, when I buy a license for a single copy, I'm
required to buy a super delux Apple machine to run it.

With the GPL, the damage is clear - without credible enforcement of the
GPL, hackers won't create free software, so allowing companies to violate
it will lead to a loss of value in that body of software.

For Apple, the damage isn't clear.  Yes, they'll sell fewer Apple Macs,
but on the other hand, they'll sell more software licenses.  And, of
course, some of those using OS/X on a cheap PC clone will later "upgrade"
to a snazzy Mac.

Then again, how is it "damage" to Apple?  The fact that they've got to
compete with their hardware as such, rather than tying it to their
software?

>> There's no reason why FreeBSD couldn't be packaged up in a box and sold,
>> with installation support, for 70 Euros the way GNU/Linux is (or was
>> until recently).  Red Hat and SuSE have done this profitably for 15
>> years.  Or even why you can't buy FreeBSD, with full support, for several
>> hundred dollars per box.  Maybe you can, but I've not heard of it.

> If you could do it at a profit, you should do it, but that is the crux of 
> the problem.  There would be a need to establish a market for the package 
> and that would cost a lot of time, effort, and money to establish.

Yet it was done for GNU/Linux, several times.

> Then, without any sort of protection for your product's code, many
> others could duplicate the package and offer it at a lower cost since
> they would not have to recoup the costs of educating the consumers.

This strategy was tried by Oracle about a year ago.  They took Red Hat's
stuff and offered it to the market at about half Red Hat's price.  It
didn't work.  What Red Hat's customers buy isn't the code itself, but
the quality of the service and support behind it.

And I think you mean "restriction" here, not "protection".  The product's
code wouldn't be in any danger, being adequately backed up, hence
wouldn't need any protection.

> But how do you differentiate Linux, or freeBSD for that matter?  Answer: 
> price.

Not really.  By the support deal offered with it.  Or by the target
market - Red Hat, Ubuntu, Debian and Gentoo are all aiming their versions
at different sorts of customers.

> Now many people are price conscious and it is becoming more and more
> of a concern in many places, but low-price suffers a lot because almost 
> everyone has an inherent belief that "you get what you pay for" and
> paying less means getting less.

I doubt that is true any more.  People download illicit music and film
at zero cost, and it's often of a higher quality that the legitimate
equivalents, because it's not encumbered with digital restrictions.
People, presumably, are using zero-cost web based software from Google.
And, of course, the quality of high priced software (e.g. from Microsoft)
is legendary.  ;-)

> When it comes to a luxury good or even what is termed a "shopping
> good", price is a lousy differentiator.  In a commodity situation, it
> might be more useful, but computers have not moved to that class yet.
> So the public image of Linux, where there even is one, is that it is
> a low-end substitute for Windows.

That is changing.  Slowly, but surely.  For example, Munich and Vienna
administrations are converting (or have converted) to Linux.  My nephew
(aged 20) sent me an email out of the blue saying how he'd installed
Ubuntu without problems and it was great.

>>>> They should get the profits from providing high quality products to
>>>> appreciative customers.  You don't have to misappropriate common
>>>> property to achieve this.  Again, gen up on the history.  It will help
>>>> you understand the GPL.

>>> They didn't misappropriate anything.  They added their own work product 
>>> to what already existed and charged a little extra for its use.

>> OK, "misappropriate" needs quote marks here.  They broke the gentlemen's
>> understanding that surrounded the code base, namely that the code base is
>> commonly "owned"[*] and each has an obligation to maintain and enhance it.
>> As a result of this breach, some coding ended up having to be done twice,
>> a stupid waste of productive capacity.

>> [*] The copyright was actually held by MIT.

>> The GPL was a remedy for this problem, making what used to be a
>> "gentlemens's understanding" into a firm legal obligation.

> That is one way of looking at it.  Another is that the GPL prevents an
> entrepreneur from offering a superior product at a fair value and so
> stifles creativity.

You don't get much more entrepreneurial than Richard Stallman, and it
was the GPL which unleashed a veritable torrent of creativity.

> If I have something better than the free version, I cannot sell 
> it, the GPL says, rather I have to offer the improvements to anyone and
> everyone.  In exchange for not having to expend a little "productive 
> capacity" to replicate some existing, free to the world to use, code, I
> have to offer my unique innovation for no return.

Yes.  Except, as described above, that's a falsely romanticised picture
of software creation.  There aren't really any "unique innovations" in
software.  Maybe one or two per generation (i.e. 25 - 30 years).  And 
even then, they tend to be steady developments of ideas rather than
sudden innovations.  The GUI WIMP interface was such a development.  The
LISP language a generation earlier.  diff and patch, perhaps.  Don't know
what's around now which is strikingly innovative.

The point is, free software doesn't work without the principle of
reciprocity.  Otherwise BSD style licenses would have become prevalent.

>>>>> Someone has to be smart enough to see an improvement and talented
>>>>> enough to implement it and dedicated enough to promote it.  They
>>>>> deserve the reward.

>>>> Who's arguing?  But if they've put in 0.1% of the effort, they deserve
>>>> 0.1% of the reward.  They tend to keep all of it, though.

>>> Things are worth what people value them at.  If I add something to
>>> Linux that is worth $100, do I not deserve the whole hundred?

>> I suppose that's fair enough.  If the system with your addition brings in
>> revenues of $100,000, and your bit is 0.1% of the system, then it's fair
>> for you to keep $100 and forward the other $99,900 to the other
>> developers.  But outrageously cumbersome and impractical.  Instead you're
>> expected to contribute your code, so that everybody can take advantage of
>> it, just as you took advantage of theirs, and you keep the entire
>> $100,000.

> That might work for a system situation where the customization effort 
> provides the product value and that is perhaps the best chance that
> FOSS has in the world since any individual set of changes is only used
> once.  If it looks like a product situation, though, where the change
> will be used in many separate instances, the cost of using the FOSS
> original is vastly overshadowed by the value of the improvement.

Then build it on a BSD infrastructure, or on a proprietary
infrastructure.  It's hardly as though GPL software has a monopoly on
many application domain.  Where's the problem?

>>> In any case, I am willing to concede that there are original works that
>>> are open source and always have been.  But the steak is sold by the
>>> sizzle and all of the sizzle these days seems relatively proprietary.
>>> And then it is followed by an open source effort to duplicate the
>>> feature, function, and look and feel.

>> You might be right, there.  But the solid, basic, durable programs are
>> mainly FOSS.

>>>> Practically anything decent to do with software development originated
>>>> as free stuff, or if older, from the Unix tradition.

>>> I like .NET and Visual Studio, of course, and if they originated with
>>> Unix, they have come a long way since.

>> I said anything _decent_.  ;-)

> Well, be that way, but, when you get a chance, download the free, as in 
> beer, Visual Studio Express C# version and MS SQL Server 2008 Express, and 
> build yourself a highly secure client/server system in a couple of hours to 
> see where it can take you.

I doubt I'd be able to run them, because they'd be built for MS systems
and they'd be lacking the source code.  But can you really build
something highly secure on Microsoft infrastructure?

Another thing, I'm not really much interested in Yet Another C-like
language.  C++ and Java are such bad languages (I do have specialist
experience here) and C is 35 years old, and all of them are painful to
use.

-- 
Alan Mackenzie (Nuremberg, Germany).