Free Availability of Sources

[Kirk Wylie]

At 08:16 PM 7/1/99 +0200, Francois-Rene Rideau wrote:
>>: Kirk Wylie
>
>> What about the case where Free
>> Software has failed to provide a tool that non-programmers want, or where
>> the Free Software alternative isn't as preferred as the closed-source
>> alternative?
>I know no such case. Where has a free software effort _failed_?

Note the phrase that I have used. "failed to provide a tool". This means
that either there is some need unfulfilled by Free Software, or some
project has started which has not resulted in that tool being available.
The simple fact that some consumer needs go unfulfilled by Free Software
alone is enough to establish that there are some cases where Free Software
has not yet met those needs, and might not meet those needs in the future.

Perhaps I mistakenly used a charged word.

>I know lots of cases where free software hasn't been there yet,
>where some free software effort was stopped
>by lack of funding or documentation or standardization.

Those are the cases of which I was referring. But don't forget that some
free software efforts have been stopped by lack of interest. Sometimes
people no longer want to work on a product that they worked on in the past.
When a project is small enough, one or two people quitting a project can
lead to it being abandoned.

>If consumers become aware of Free Software, and demand free software,
>these barriers will fall down.

There is a massive logical leap in this statement. Simply because people
demand something does not mean that it will be provided, except through
market-based means. This is the reason why closed-source software exists: A
group of people wants a piece of software, someone produces and sells it to
them. The reason why the system works is that the people are directly
placing a value on a particular piece of software, and thus manufacturers
know how much to spend on making it.

>> There are cases today, and
>> will be cases in the future, where there is no Free Software to solve a
>> particular problem that users have, and exhorting them to write it
>> themselves is useless if they are one of the 99% of humanity who have no
>> idea how to read source code, much less write it. And users should neither
>> be expected nor required to understand or write any piece of software they
>> use.
>That's what I call an _opportunity_ for a market of free software
development.
>In a free market, difficulties are opportunities.
>In a "protected" market, difficulties are the hell of consumers.
>Remove governmental enforcement of proprietary licenses,
>you'll have a free market.

No, you won't. The problem is that what you are suggesting isn't a free
market, it is an eliminated market, because no one has the opportunity to
provide something which is proprietary. Moreover, no one has the ability to
sell software alone, so the market has gone away.

For example, say that I have a piece of software that I wish to sell. The
whole reason that someone would want to buy it is that it functions faster
than someone else's software, because I have been very clever at coding it.
If, however, my source code has been made available to the world, then
anyone else can duplicate my efforts.

Eventually all source code will be essentially the same homogenous entity,
because there is no advantage in differentiation. The only way to
differentiate is through packaging and supplementary services, such as
consulting and support. This takes away the benefit of the consumer to buy
something else.

Where can products compete? In your idealized scenario, all products have
reached the pinnacle of perfection. And if someone comes up with something
that is better, what advantage does he have to create it at all, knowing
that it will never benefit him in a market environment?

Note that the things that make software unique make this a uniquely
possible scenario: because there are no barriers to entry whatsoever
(unlike Hardware), there is nothing to prevent someone else from abusing
the open source scenario.

Yes, a gift culture does exist to some extent that will provide many
things, but then it is no longer a market.

>> Or the situations where a non-Free application is superior to a Free
>> application in the minds of the users of that software
>Is _currently_ superior. So what? Supporting free software will always
>be a better choice in the long run. If you plan to die or otherwise
>cease all activities within three years, indeed, you may not gain much
>at supporting free software. Else, you're only mistaken (and/or crooked)
>at choosing proprietary software.

But what about the times that it is not? Logically I should expect that at
some point Free Software will be better at a particular task than non-Free
software. Should people not be allowed to use the superior software before
the Free Software catches up? I should hope that consumers are allowed to
use whatever they want, for whatever reasons.

But as I firmly believe, requiring source code be published essentially
eliminates proprietary software. If you want that for reasons irrelevant to
consumer choice, that's fine. But don't claim that it's some kind of market
force driving you to do so.

>> [...] you have to be
>> willing to accept that there will be situations where people choose to use
>> closed-source applications, regardless of the reasons, even if they know
>> that on a conceptual and engineering level that solution will not be as
>> good, in the long term, as a Free solution.
>There exist such things as security regulations:
>you may not drive on the wrong side of a public road;
Because you harm others when you do so.

>you may not drive when drunk when on a public road;
Because you harm others when you do so.

>you may not sell a car on the consumer market lest it have standard
equipment;
Because you harm others when you do so.

<...SNIP...>

>Well, let me add as a _possible_ additional law:
>you may not make sell hardware or software
>on the consumer market lest its source is freely available.

The problem here is that you need to get closer to the problems involved.
If you are going to have any particular law, then it should directly seek
to address a wrong: You may not sell hardware or software on the consumer
market lest you have established that it will not harm anyone. Having
source freely available is a means to an end, not an end in and of itself,
and not a means which does not have any negative consequences.

>Access to the source is not a universal right,
>like freedom to use, copy and modify information;
>but it _might_ possibly be a valid regulatory condition
>concerning open sale of software on the consumer market.

Yes, it might. But in this case it should be done to demonstrably benefit
consumers, and not out of moral principle.

>Note that a side effect is that is increases the value of programmers
>as opposed to resellers, since resellers must convince (with money)
>the programmer to release source
>before they can get the dough by distributing the software.

I think you have a definite misunderstanding of the role of money and
marginal benefits.

In a world where all source code must be free the marginal benefit of an
additional line of source code approaches zero. Or, rather, it approaches
zero to a particular person or reseller, but is quite valuable to the
community in general.

Thus there is no incentive for anyone to pay for any particular line of
code for the purpose of reselling it. There is an incentive for someone to
pay for a particular line of code to USE it, but not to resell it.

>As for the extremist libertarian point of view according to which such rule
>would be out of the acceptable attributions of state, let me disagree.
>If you re-read classical liberal texts (US people: "classical" means
>before the word "liberal" was stolen in the US by socialists),
>you'll see that what classical liberalism opposes
>is governmental _management_, not governmental _regulation_.
>On the contrary, the very role of the government is to enforce the rules
>necessary so there be as free and unbiased a market as possible,
>in ways such that the fundamental rights of citizens are guaranteed.
>It is unacceptable that the government tell me to use any given software;
>it is acceptable that it demand guarantees as to what software I may use
>in a variety of purposes that also concern other people.

I agree with all the previous.

>For instance, games and other end-user software isn't used during public
>transactions, and it is not necessary to demand source, whereas accounting
>and otherwise business management software is critical and publication
>of its source might be demanded by governmental regulations, to ensure
>that the fundamental rights and personal security of employees and consumers
>be respected.

I disagree with the non sequitur that you draw. Recall that in classical
principles (especially J.S. Mill) the duties of the state are to protect
the rights of individuals, but NOT to protect them from themselves. This
can be done in many ways that are far more useful to consumers than simply
forcing source code to be released.

1) Require that the software seller enter into a contract where the results
of the program ARE waranted (i.e. if the program breaks and you lose all
your data, the manufacturer is to blame and is liable).
2) Require that software which is NOT waranted print that in huge letters
on the cover (less useful as more software is distributed online without
boxes), to allow the consumer to decide which to use.

Note that these types of options preserve:
a) The rights of the individual to choose what type of software to use,
b) The rights of the individual to get open source software if it is
available,
c) The rights of the individual to choose to injure himself if he so chooses,
d) The rights of the individual to choose amongst various licensing
strategies.

Note that the individual gets the chance to buy software which works well,
but also to buy cheaper/worse software which doesn't. The only times in
classical thought where it is permissible to restrict the rights of
individuals is when their exercising of those rights would infringe on the
rights of others. I find it ironic that in one part of your argument you
would argue that the consumer must be protected from himself and/or the
rapacious software companies, yet in another you evoke liberal thought to
argue that a free market must be maintained.

Explain to me how my purchase of a copy of Microsoft Office infringes on
the rights of others.

Kirk Wylie
Kirk Wylie                         | kwylie@weblogic.com
Software Engineer                  | +1 415 364 4539
BEA WebXpress, nee WebLogic        | http://weblogic.beasys.com