[ Note: This document is available on the web at
http://www.**-**.com/ ]

Introduction and Thesis

I have read a lot recently about the effort some net activists are
making to replace the term "free software" with "open-source
software".  (I will use "non-proprietary software" in an effort to be
somewhat non-partisan). Supporters of this include such people as Eric
Raymond (who wrote a manifesto on the subject (See
http://www.**-**.com/ ~esr/open-source.htm), Linus Torvalds, and Bruce
Perens.  This move has been opposed by some people, most notably
Richard Stallman of the GNU Project (who has posted a rebuttal
(see http://www.**-**.com/ )

The supporters of "open source" give two main reasons for their
proposed change:

-- It eliminates confusion about the term "free software",
   which can be taken to mean either software received at no cost, or
   software that is free under some other definition, such as the
   one promulgated by the Free Software Foundation (FSF). (See
    http://www.**-**.com/ ).

-- It will make this type of software more acceptable to corporate
   purchasers who might be scared off by the term "free".

These two reasons have largely framed the terms of the debate.
Most postings I have seen on the Net debate the point in terms of
which term is clearer or which one will appeal most to business.
There has been a lot of discussion, but no real resolution to this
dispute.

But I believe the real reason there is a division is not because of
a difference of opinion over which term best communicates what
non-proprietary software is.  I believe there are fundamental
philosophical differences about the nature of software ownership
between many people on both sides of the debate, and that those
differences play a part (perhaps unconsciously) in how people feel
about a switch.

My belief is that the "open source" advocates do not believe
software should intrinsically be free.  They do want to see
non-proprietary software thrive, but it is because they believe it is
technically superior and because they prefer it as a matter of
personal taste.  So they define the terms of discourse in a manner
compatible with that philosophy, and in a manner they hope will bring
about their vision of world domination by non-proprietary software.

I think the "free software" advocates do believe in the intrinsic
freedom of software and, having successfully promoted a term
consistent with their philosophy, do not want to see it changed. They
are more likely to care more about the purity or ethics of
non-proprietary software as opposed to its market acceptance.

Many people fall in the middle, of course, including myself.  But
after thinking a lot about this, I have come to the conclusion that
"free software" should remain the preferred term.  I also believe
that, at a bare minimum, everyone should look at this terminology
debate not strictly as a disagreement about definitions, but as an
important philosophical decision point.

I will try to explain each of these points in detail, hopefully
without getting too boring or offending everyone on the planet.

Disclaimers

Some of my analysis is based on my conclusions about the beliefs of
the principals of the different camps.  Obviously there are a lot of
people out there and judging a movement by its leaders provides only a
limited insight.  In particular, I should point out that most people's
personal views probably fall somewhere in the middle.  This posting
should be read with that limitation in mind.

I do not personally know any of the principals, and therefore I
have had to rely on reading their published works.  This limits me to
talking about people who have published a great deal on the net.
Those people are Richard Stallman, Eric Raymond, and Linus Torvalds.
Obviously my interpretation of their writings or quotes is just that -
my opinion.  It should not be viewed as authoritative, particularly in
light of the fact that all these people are alive to explain
themselves.

Philosophical Differerences

It may seem controversial to suggest that the "open source" camp
does not believe in the freedom of software.  Indeed, people like
Linus Torvalds and Eric Raymond have produced some of the finest and
most widely used software available today - much of it distributed
under the exact same terms as Richard Stallman's.  No one (certainly not
me) questions their personal contributions to non-proprietary software
and computing in general.

However, neither Eric nor Linus believe software needs to be free.
In fact, both of them are very comfortable with the idea of
proprietary software.  Linus is upfront about stating he does not mind
proprietary software.  He was quoted in an interview (see
http://www.**-**.com/ ~tlug/linus.html) as saying:

<QUOTE>
Then there is software that is commercial but doesn't come
with sources (the "traditional" commercial software as opposed to a
Red Hat Linux distribution). And I don't try to preach against that
either: I hate the fact that I (and others) can't fix bugs in them,
but sometimes that kind of software is the way to go.
</QUOTE>

So he personally prefers free software (and says so in the
interview), but does not think that software particularly needs to be
free or that there is anything wrong with proprietary software (except
perhaps not being able to fix bugs).

Eric Raymond is even stronger in his support of proprietary
software.  Here is a quote from a paper he wrote (see
http://www.**-**.com/ ~esr/writings/open-economics.html):

<QUOTE>
I believe in strong intellectual-property rights. I believe
any programmer has the same fundamental right as any other producer to
`hoard' and sell the fruits of his/her labor on whatever terms the
market will bear. While I choose to publish nearly all of my software
as open-source, I will never, ever condemn others for choosing
differently.
</QUOTE>

So there are at least two principal advocates of the term "open
source" that personally do not believe in the intrinsic freedom of
software, and in fact have no problems with proprietary software.  For
them, the choice of non-proprietary versus proprietary is one
primarily of personal preference and/or choosing the right tool for
the job.

On the other side of the fence, Richard Stallman's views on
software are perhaps the best known on the net.  He wrote a paper
called "Why Software Should Not Have Owners".  (See
http://www.**-**.com/ ).  The title alone
pretty much sums up his views.  Richard Stallman believes that it
is morally wrong to produce software that is not free.  He
personally refuses to even use software that is not free (except
for the purpose of creating a free replacement).

So in this debate, we have leadership with diametrically opposed
views on the nature of ownership in software.  Because of this, I do
not find it surprising that these people have a major difference over
terminology.

The absolutism of Richard Stallman's free software stance has often
been troublesome to people who feel that what they produce is theirs
to do with as they please.  I suspect this is particularly true of
Americans, who tend to be believers in strong private property rights.
For a long time however, not too many people challenged this
philosophy.  My personal speculation is that this is because during
much of the 1980's and early 1990's, the FSF was the {*filter*} producer
of major non-proprietary software packages.  People were willing to
play along with the hard line philosophy in return for the great
dividends in software that it paid (Emacs, gcc, gdb, etc).

In the mid-1990's though, the FSF ceased to be the driving force in
non-proprietary software.  Most of their replacements for Unix
utilities were complete and had reached a point of reasonable
maturity.  The major uncompleted component of the GNU system - the
kernel - was under development, but the design was ambitious (perhaps
overly ambitious) and the pace of development slow.  Meanwhile, the
non-proprietary Linux kernel, released under the GPL and combined with
the rest of the GNU programs, exploded in popularity as the OS of
choice among those who preferred non-proprietary software.  Other
programs like perl, under a different but similar license to the GPL,
also rose in popularity.  So with most of the developments in the
non-proprietary software world coming from non-FSF sources, the
incentive to ignore basic philosophical differences with the FSF began
to disappear.

"Why" Is As Important As "What"

So we come to the "open source" versus "free software" debate.
First let me state clearly that I do not believe the backers of the
term "open source", including Eric Raymond and Linus Torvalds, are
engaged in a sinister plot against the FSF, non-proprietary software,
or anything else for that matter.  I am not alleging a {*filter*}.
But I do believe that switching to the term "open source" and focusing
on penetrating corporations represents more than just a change in
terms.  It is a fundamental shift in thinking about what
non-proprietary software is all about.  A shift away from thinking
that non-proprietary software should be developed because it is the
right thing to do, towards thinking non-proprietary software should be
developed because it provides greater user benefits than proprietary
software.  Or, as Eric Raymond said, getting rid of the "losing
attitude" associated with term "free software" and concentrating on
how to increase the number of people using non-proprietary software.

This effort has a great many guns in its arsenal, because there are
a great many tangible benefits to non-proprietary software.  For
example:

-- Non-proprietary software tends to be of higher quality
   than proprietary software.

-- You can get good support for non-proprietary software through the
   user community - often times better than from a proprietary software
   company's help desk.

-- ...

read more »

"Aaron M. Renn" <ar...@urbanophile.com> writes:

You forgot the important one: "open source" is trademarked. Any
proprietary company can write ad copy saying "Our product is free" but
they can't say it's "open source" or use an eventual "open source"
logo. So "open source" helps a lot there, I think. Legal enforcement
of concept purity.

Otherwise I like your argument. Perhaps you should mention that there
are many practical benefits which are also ethically important, having
to do with centralization of power in the hands of software companies,
for example, or keeping technology in the hands of the people, or
making technology genuinely useful and subordinate to our needs rather
than a cycle of perpetual obsolescence for the benefit of the
corporate world. These things sound important enough to interest
people in the cause and make the value-laden word "free" seem
warranted.

Personally I don't find it convincing to say that there's anything
ethically inherent about software, any more than there's anything
ethical to say about a lump of coal. But once you put the software in
a context, and talk about its practical implications, you have an
ethical question, just as the lump of coal becomes ethically important
when you bring in how it was mined, what effect burning it might have,
and so on.

So I sympathize with those who would like to focus on the practical
benefits of free software, but since those practical benefits include
much more than simply the "greater user benefits" you mention, I think
you're right to emphasize ethics. The practical benefits are in many
ways ethically important ones.

Havoc Pennington
http://pobox.com/~hp

Dear Mr. Renn,

    You are right in saying that this debate is a debate as much  about
philosophy and ethics as it is about terminology. The philosophy that "free
software" advocates have is that ideas should be free and not subject to
restriction. However, the philosophy that "open source" advocates have is
that a person's work should not be free. How about this idea: suppose there
is a website where a user, who does not know how to make modifications to
software himself, has an idea and posts a description of his idea, plus an
amount he is willing to pay for someone to do the work.. Another user might
see the contract and say, "That's a great idea, I will add X dollars to this
contract." The contract amount can then grow or decrease until some
programmer (or group of programmers) decides to take the contract. The
contract then becomes closed, and subject to the normal operatings of the
law. The point I am trying to make is that he who does the work, or supports
it, should be the first to enjoy its benefits. If he is feeling generous,
then he may share.

Sincerely,
Howard Hong

: How about this idea: suppose there
: is a website where a user, who does not know how to make modifications to
: software himself, has an idea and posts a description of his idea, plus an
: amount he is willing to pay for someone to do the work.. Another user might
: see the contract and say, "That's a great idea, I will add X dollars to this
: contract." The contract amount can then grow or decrease until some
: programmer (or group of programmers) decides to take the contract. The
: contract then becomes closed, and subject to the normal operatings of the
: law.

It's a *great* idea, except how are you going to enforce the contract?  If
someone decides not to pay, who's going to pay the legal, etc.?

: The point I am trying to make is that he who does the work, or supports
: it, should be the first to enjoy its benefits. If he is feeling generous,
: then he may share.

These days, "he who does the work" is not alone...  There's 100 people out
there writing an identical package (in concept) to what you're working on.
The reality is that the software market is bottoming out (towards zero)...
Someone else will give it away for free...

Save the dollar transactions to something where value is added.  Documentation,
support, and similar areas are viable businesses these days.  Software
development and distribution is not.  The Internet replaces both of these.
It'll only be time before Microsoft wakes up to this, if they're not already.

Mark

This may be true in some cases, but not nearly all. I used to work at
Papyrus, a leading auto-sim maker... Sure, driving games are a dime a
dozen, and some are free with source, but Papyrus continues to roll in
dough, because they do it irrefutably better than all but one or two
other developers. So, the price for "any software that allows me to
drive a virtual car" might head to 0, but the price for "the best such
software" is still around $50.

Could someone topple Papyrus? Sure, but you'd need to be damn
dedicated, good engineers, good coders, live and breath racing, and
code for maybe 10-20 man-years to catch up. That's a tough nut to
crack, and I'm sure the Papyrus folks didn't lose sleep over it last
night.

Suppose a business needs a customized inventory purchasing and
tracking database with integrations to their factory floor, inventory
room, and suppliers? Are there 100 people out there working on that
software? If so, why is the business hiring a contractor to build the
system in the low 6-figures for them? Are they just too lazy to turn
over a rock, under which there is no doubt someone working on exactly
what they need?

I haven't seen an RFC on an "Internet computerized software
developer". The internet can replace distribution, and thus can change
the landscape for development, but saying the internet obviates the
need for developers or turns development into an non-viable business
is fallacy.

If anything, I think it stretches the difference from the best to the
worst programmers. The best programmers are now free to work (for free
or for whatever they and their employer/client/investor agree they're
worth) anywhere in the world without uprooting themselves and family.

Someone in India could now work for someone in California if they
wanted. This has an upward bias on the income potential on the best
programmers, and probably has a downard bias on the worst. (After all,
I don't *have* to hire some local joker just because he's the only
game in town. Now, I can hire from around the world, giving me more
choice.)

Sure, some jobs will still be local, but the more jobs that are
wide-open to telecommuting, the better for the industry as a whole,
but it's bad news for programmers who live in expensive areas and
suck. :-)

---Jim

I was thinking about making it a formal legal contract, just like buying a
house, or drawing up a rental lease or buying a car. While it is true that
other people are working on packages similar to your idea, a business might
want to see a schedule of results, to know that someone is definitely
working on the problem, and that they can expect a result in the near
future. Business is about certainties, and on certainties they can draw a
profit. Perhaps middle arrangements could be made, like a deposit on the
contract, or the object code could be delivered with a timed expiration on
it. If one or more of the parties reneges on the deal, the Internet will
know about it. Perhaps somebody could draw a profit from running a
matchmaking website, and pooling a deposit from contractees.

Howard

On 10 Mar 1998 07:25:32 GMT, Mark Jackson <mar...@shell.rmi.net> wrote:

Use the "Open Lawsuit Interface(tm)" like this:
"I'll give you 10% of whatever damages we get."

--
   O<      ( (      [ Josh Fishman      GPL DNA NOW! ]
 _NH >=O    ) )     [  <mailto:jmf9...@is4.nyu.edu>  ]
<_>-<_   + :::::-.  [ Linux:                         ]
 HCl<O>     :::`-'  [  hex, bugs, flock() and poll() ]

:
: >: How about this idea: suppose there
: >: is a website where a user, who does not know how to make modifications to
: >: software himself, has an idea and posts a description of his idea, plus
: an....
: >
: >It's a *great* idea, except how are you going to enforce the contract?  If
: >someone decides not to pay, who's going to pay the legal, etc.?
It sounds like you are talking about an software escrow service. How do
like real estate people handle all those issues and how might that apply
to your model?
: future. Business is about certainties, and on certainties they can draw a
: profit. Perhaps middle arrangements could be made, like a deposit on the
: contract, or the object code could be delivered with a timed expiration on
: it. If one or more of the parties reneges on the deal, the Internet will
: know about it. Perhaps somebody could draw a profit from running a
: matchmaking website, and pooling a deposit from contractees.

M> Save the dollar transactions to something where value is added.
M> Documentation, support, and similar areas are viable businesses
M> these days.  Software development and distribution is not.  The
M> Internet replaces both of these.

That's completely untrue.

For example, the author of shostscript is being paid to work on DGS.
I don't see GM using a freeware CAD/CAM system, they use Unigraphics,
who pays me to develop for them.  Lots and lots of people all over are
purchasing software.

The Internet does not replace this.  It makes it easier to
collaborate, which makes it easier to tackle large projects.  But how
many freeware programmers are going to work on a telecomm switching
system?  Hey, even in the realm of word processors, where we have had
dramatic improvements in free software, commercial apps dominate.
Yes, Star Office is free... if you're an individual running Linux.  If
you are a business, or run any other OS, yous till have to pay.  And
then there's WordPerfect and Applix, both of which are commercial.
All of those applications have numerous features which haven't been
duplicated by free developers.

I'm not saying that given time, the free community can't duplicate or
best these apps.  Look at the Gimp, after all.  But for people who
need something today, not next year, commercial software will be
there, because it's easier for people to program when they can eat.

--
Alan Shutko <a...@acm.org> - By consent of the corrupted
You would if you could but you can't so you won't.

In the US at least, real estate earnest money is usually held in escrow by
the office of the listing agent for the property.  This escrow money earns
interest which is payable (and taxable) to the purchaser on closing.  The
amount of money, the interest terms, and the escrowee are all part of the
real estate sales contract.

--
****************************************************
* Aaron M. Renn                                    *
* Email: ar...@urbanophile.com                     *
* Homepage: <URL:http://www.urbanophile.com/arenn> *
****************************************************