[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnu-arch-users] OT: Slavery???

From: Tom Lord
Subject: Re: [Gnu-arch-users] OT: Slavery???
Date: Wed, 12 Nov 2003 14:37:55 -0800 (PST)

    > From: "Stephen J. Turnbull" <address@hidden>

    Stephen> Thank you for this opportunity to clarify my own
    Stephen> thoughts.   

    Tom> I encourage you to try harder.

    Stephen> Et tu, Brute.

Well, I have.   So, thank you too.   Now, love-fest aside:

Let's review things and see if that clears anything up.  I think you
simply missed some of the structure of my argument that I did not make
explicit enough.  I'll also answer some of your specific concerns at
the end.

You asked:

Stephen> What justifies "outrage" at the offer of a private contract
Stephen> that one is not party to, and one is free to accept or refuse
Stephen> as he chooses, and one is free to advocate publically that
Stephen> others not accept, either?

Putting the question that way probes the heart of the free software
movement in a careful way _and_ summarizes one of the most common
criticsms of the movement elegantly.   It's a good question.

The question also contains some implicit assumptions that perhaps need
to be made more clear.  I think that your question, as you've put it,
can be fairly restated:

        The offer of a private contract that one is not party to, that
        one is free to accept or decline, and that one is free to
        advocate against is presumptively an exercise of the
        legitimate rights of the offering party and impinges on the
        rights of no one who declines the offer.

        A claim that such an offer is ethically and/or morally
        "outrageous" therefore requires a positive defense
        to overcome our presumption that the offer is simply an
        exercise of the offering party's legitimate rights.

        I have seen no such positive defense for your expressed 
        outrage over the BK public license.   What is your defense
        for that outrage?

As evidence that that is a fair restatement, I'll also quote something
else you said later when you thought I was making a "circular

Stephen> You're outraged because _your_ right to software freedom is
Stephen> somehow abridged when some third party offers _me_
Stephen> proprietary software, but the only evidence I can see for
Stephen> your right that _overrides my rights_ to privacy and contract
Stephen> is your outrage.
         ^^^^^^^^^^^^^^^                (emphasis added)

So that is what I set out to reply to and I will restate my arguments
below, hopefully more clearly this time.

(You also, separately and late the discussion, argued that BK was a
kind of necessary, temporary, patch to a broken process -- on balance
more beneficial than harmful -- and I'll speak to that too.)

* examining the implicit assumptions

  Since you are asking for a postive defense to overcome presumptively
  legitmate rights to "privacy and contract", I started by examining
  the nature of that presumptive legitimacy.

  One _can_not_ provide the positive defense you asked for without
  first examining the nature of the presumption.  Where does the
  "right to contract" end, in particular?  How does the "right to
  privacy" apply here?

  I answered that the boundaries around the "right to contract" are
  drawn by the requirements of _regulation_, _ajudication_, and
  _enforcement_.  While the entering into and performance of a
  contract is the conduct of the parties to the contract, it in fact
  involves all members of society for the purposes of (actualized or
  merely potential) regulation, ajudication, and enforcement.

  In short, the "right to contract" is not legitimate because an offer
  has the _form_ of a contract: it is only legitimate if the _content_
  of the contract is consistent with conduct that the whole society
  agrees to regulate and enforce.  Thus, for example, certain
  employment contracts in California are void because they would
  deprive workers of essential liberties and our society has chosen to
  refuse to be the enforcer of such deprivation (or even to tolerate
  it within our borders).  The good to society of certain workers'
  rights trumps the good of an unfettered right to contract.

  So we will have to examine whether the conduct of proprietary
  software licensing agreements in general, the BK free use license in
  particular, is one that we ought to be enforcing as a society.

  I answered that the boundaries around "private [business-like]
  conduct" end at the right to contract.  You're free (within limits,
  of course) to give Joe gold coins and take from him some "secret,
  binary-only, software" in privacy, but if you regard that as an
  enforcable contract then you have invited society to examine that
  conduct and given up your right to privacy in the matter.

  Since your question pertains to contracts, not private,
  non-contractual conduct, the right to privacy that you brought up is
  pretty much a red herring.  We should just ignore it as irrelevent
  to what we're talking about.

  To summarize the first 2/3 of my message, the 2/3 that you replied
  to with "yadda, yadda, yadda :-(", the positive defense you are
  looking for, of my outrage at the BK license and proprietary
  licenses generally, must be an examination that compares the social
  costs and benefits of endorsing vs voiding such contracts.  The
  general question is really: do we as a society benefit by permitting
  such contracts to exist at all?  The BK-specific question is, if we
  as a society do not generally benefit by tolerating such contracts,
  why is BK a particularly egregious example of the harm to society --
  an offensive "taking advantage" of a fault in our legal system by an
  exploitative business?

* defense of outrage

** For Software Freedoms in General

  As I said, one would have to write a book.

  I think it is a particular strength of the free software movement
  that there are so many arguments, quite divergent in nature,
  starting from quite different premsises -- all affirming the
  importance to society of protecting software freedoms.

  I generally like to categorize these arguments as two (intersecting)

        A. those that argue for the rights to use, modify, copy, and

        B. those that argue for the rights to obtain source code

  I did "include by reference" a large but unspecified subset of
  arguments we've all seen before: those put forth by the FSF.  I'm
  not terribly interested in just repeating those here to add length.
  I will mention that some of my favorites are primarily in set (A)
  and say, roughly, that the nature of software permits one who is in
  possession of it to "share and share alike" -- a moral option whose
  benefits are of great value to society as a whole.  So simple and
  basic is that right to be neighborly that it is a travesty of
  justice if we, as a society, create artificial means to take that
  right away in exchange for -- for what?  certain business models?
  Surely there are ways to organize the production of software as a
  business that do not require us to make that offensive sacrifice of
  our rights to behave helpfully.

  If the "right to be neighborly" is a postive incentive to void
  proprietary software contracts, the "natural rights argument" gives
  us a negative incentive: what, realistically, happens if we agree as
  a society to enforce these contracts?  Many of us with the technical
  perspective to appreciate the matter realize that we _can_not_
  enforce such contracts reliably except by extremely intrusive,
  destructive, unjust means.  Realistically, it would require a kind
  of de facto facism: an orientation of all computing hardware towards
  copy protection; a government complicit in creating and sustaining
  that condition; a surveillance-oriented enforcement activity aimed
  at hunting down violators.  Nightmarish as that description sounds,
  have we not seen over the past 25 or so years a steady ratcheting up
  of _exactly_that_?  Are there not now, on the table, seriously
  intended and difficult to defeat, corporate led proposals to keep
  going in those directions?

  For the most part, those arguments are _primarily_ in the (A) set:
  they speak to the freedoms to use, modify, copy, and distribute
  software.  What of the (B) set: the right of access to software?

  The FSF has, as I recall, provided "positive incentive" arguments
  for source code access rights: that it is of great utility,
  essentially.  It would be difficult, though, to argue from that
  utility to a general prohibition of distribution that does not
  provide access to source.  For example, while the freedoms to use,
  copy, modify, and distribute _affirm_and_protect_ a basic right to
  free expression, a requirement of access to source is a _limitation_
  on free expression in the form of a compulsion to speak.

  Compulsions to speak (or not speak, for that matter) are not
  unprecedented, however.  In fact they are quite common in areas that
  relate to public safety and well-being.  Witnesses may be compelled
  to give (some kinds of) testimony.  Drug manufacturers are compelled
  to disclose active ingredients and possible side effects.  Food
  manufacturers are compelled to disclose ingredients and nutritional
  information.  Construction contractors are compelled to disclose
  building plans.  The latter three are interesting in contrast to the
  first: whereas a witness may be compelled to speak of a harm that
  occured, in the latter three cases, the compulsion is to make
  disclosures which help to preempt future harm that might or might
  not take place.  On balance, we have recognized, the burdon of this
  compulsion outweighs even its challenges to one of our most
  cherished rights (that "of expression").

  I hope I do not need to persuade you that the public safety and
  wellbeing concerns surrounding software are profound.  I mentioned
  some particular "in the headlines" examples: software vulnerable to
  Internet viruses, voting machines, and privacy issues.  But I don't
  think I needed those examples to make a case which is a foregone
  conclusion but only to make clear what conclusion I was referring
  to: software and computing systems, in spite of their benefits, are
  profoundly dangerous to society.

  Disclosure of source code (with terms that permit use, copying,
  modification, and distribution) is a partial but signficant
  mediation of that danger, and one which is without peer.  A severing
  of source access from any of the rights to use, copy, modify, and
  distribute would significantly undermine the mediation of the danger
  of software (do you really need a lengthy explanation of why?).  By
  no other means than access to source code is a human-readable,
  pragamatically useful, and above all reliable account of the
  behavior of the software available.  Software is and mostly likely
  will continue to be produced in such volume and deployed in so many
  situations that limited disclosure of source code to, for example, a
  regulatory or certification organization, is impractical and hence

  The strong public interest is squarely in the box that says:
  disclosure of source code to all recipients of a program is
  compulsary, even though not in and of itself a complete mediation of
  the risks:

        "Don't trust any software you get from me."
                        Ken Thompson -- "On Trusting Trust"

  In summary: 

  Regarding my outrage at proprietary software generally, and why it
  overcomes your "right to contract", I have been persuaded by
  overwhelming and multifaceted arguments that:

        (a) enforcement of your proprietary contract is, on balance,
            more detrimental to society than beneficial.  Such
            proposed contracts are among those that we as a society 
            have interest in declaring void on the basis of their

        (b) beyond just that, because of the unbounded and signficant
            nature of the dangers presented to society by software,
            disclosure of source code with the right to use, modify,
            copy, and distribute that source should be made compulsary
            for those who distribute software

   It is true that our society does not currently enforce those
   principles.  And it is true that some businesses conduct themselves
   in a manner that takes advantage of that lack of enforcement.  This
   is a case of "two wrongs don't make a right": that a business
   activity is legally permitted does not spare it from reactions of
   moral outrage, and in the case of proprietary software, we have
   plenty of reasons for such outrage.

* BK in Particular

  The fight for software freedoms is "a long, hard slog".   There are 
  many proprietary programs that might be singled out for attention
  as worthy of becoming the focus of outrage.

  Why, then, BK?  Why not pick on, say, "Super Space Invadors 9.5" or
  the operating system in my phone, or the latest version of Solaris?

  I see two (intersecting) sets of reasons:

        (a) those that pertain to the GNU Arch community of 
            contributors specifically

        (b) those that should concern all who are concerned about
            protecting and promoting software freedoms

  The former category, (a), is pretty easy and obvious: BK has
  redirected resources that might otherwise have been spent to create
  a free software revision control system for the kernel project.  We
  have in the public record some interesting claims and countering
  demonstrations.  Larry has remarked (my paraphrase) that "there is
  _no_way_ BK could have been developed as a GPL program" and that the
  costs of its development equate to something around $12M in
  spending.  While that $12M-equivalent in development effort, we are
  informed, was provided in significant part by unsalaried workers
  performing in anticipation of future returns, the _only_way_, Larry
  would have us believe, to secure those future returns was to make BK

  Larry's claims are bunk on at least two grounds:

  First, "$12M" may indeed be the ballpark of the cost-equivalent of
  writing BK -- this does not mean that it was the necessary
  cost-equivalent of producing a comparable or superior system.  The
  existence, status, and history of arch, while not necessarily
  conclusively to casual observers, strongly suggests that "$12M" is
  the cost of developing a system which takes an inferior approach to

  Second, the idea that BK had to be proprietary to secure future 
  returns relies on a premise:  that businesses _can_not_ be expected
  or persuaded to pay for the production of a GPLed revision control 
  system.   My personal experience, for many months now, has been
  that many _individuals_ find it in their interest to help fund
  the development of a GPL revision control system -- where _exactly_ 
  is the difference in a corporation that means it should reach a 
  different conclusion?  Larry has made public record of his access
  to various "luminaries" in the software industry and at least one 
  of those luminaries has publicly confirmed his access and influence.
  Larry has made public record that he noted a flaw in the kernel 
  project's engineering processes (that "Linus doesn't scale").
  Has he not, therefore, squandered an opportunity to use his access
  and influence to persuade various corporations to fund the
  development of GPLed revision control system?

  But instead of promoting software freedom, Larry has done nearly the
  opposite: from the podium of his privelege, he argues to the public
  against software freedoms, against the ethics of them, against the
  public's right to demand them.  (Which brings us towards (b), which
  I'll get to below.)

  Now if Larry had merely "won in fair competition" -- that'd be no
  grounds for outrage.  He did not "merely win" -- (1) he won by using
  a proprietary software business model which I've already argued is,
  in general, harmful to society; (2) having secured victory, he has
  used his advantage to actively undermine the chances of the GNU arch
  succeeding as a business model.

  Now (b), pertaining to those not specifically interested in GNU

  The kernel project is, indeed, a major success in spreading the
  understanding of, promotion of, and protection of software freedoms.
  It is also a major beneficiary of preceding and ongoing work to 
  promote and protect software freedoms.

  But just a couple of years ago, the perception arose that there was
  a serious flaw in the engineering processes of the kernel project.
  To the extent that perception was accurate, that flaw was a threat
  to the fight for software freedoms.

  Larry and BitMover stepped forward to "fix" the flaw, but have they,
  and at what cost?

  Is the free software movement strengthened by Larry's unbalanced
  right to engage in political speech against software freedoms on

  Is the free software movement strengthened by Linus "devil's
  bargain" to give up the right to help develop a free software
  revision control system in exchange for using BK?

  Is the free software movement strengthened by my having to choose
  between purging the common english phrase "line of development" from
  the arch documentation or defend myself in court against BitMover?

  Is the free software movement strengthened by a deal between 
  IBM and BitMover that resulted from kernel project development?

  Is the free software movement strengthened by a partition 
  of the community into "those who agree to use BK" and "those
  unrestricted in what projects they may contribute to"?

  And on an on.

  I have earlier argued that BitMovers legal right to offer the free
  use license is not a legitimate right and that its illegitimacy 
  is, in fact, of deep concern to our society.   Larry and BitMover's
  conduct reveals a propensity to exploit that legal right to
  actively undermine efforts to address that deep concern.

  As I said, damn right I'm outraged.

* miscellaneous replies to Stephen

    > Your argument from social responsibility is also indirect, and it
    > assumes that you have correct (objective!) knowledge about what social
    > responsibility (and social harm) is.  I've never seen you try to
    > explain what you mean by those concepts, let alone show that your
    > ideas are any good at all.  So I can't evaluate it, except by its use.

  I understand that you are not a software engineer and that we who
  are have an uphill battle to convey to you the danger to society 
  present in computing systems and software.    I am not sure I can do
  the job fully, for you personally, in a single message by admonish
  you to wake the hell up.

    > Which mostly pure defamation ("poster child for socially irresponsible
    > engineering"); I don't think I've ever seen you draw attention to "an
    > excellent example of socially responsible engineering."  It's not like
    > you don't know how to compliment, either; I think you praise others
    > more often than any other programmer I know.  Gotta wonder about this
    > contrast.

  I have become to see Linus as a destructive opportunist.  He has 
  attained massive personal benefit from the free software movement.
  He has, de facto, become a public spokesman for it.   From his
  podium, especially in relation to the BK issues, he mostly begs off
  of any personal responsibility towards the movement.

    > They are extremist in an objective (but limited) sense: you wish to
    > prohibit _all_ free individuals from engaging in _any_ of an entire
    > class of behaviors.  

  I think that there are strong and not even slightly unprecedented 
  arguments for that, which I have hopefully at least begun to
  elaborate above.

    > You do not demonstrate any harm from the behaviors themselves.
    > Your arguments depend entirely on the harm that _might_ proceed
    > from various enforcement mechanisms.  That makes my task rather
    > easy: all I have to do is either (1) present a non-harmful
    > enforcement method or (2) allow the contract to go unenforced,
    > and your argument is completely refuted.  On to the
    > counterexamples!

  (2) would be an endorsement of my position: the voiding of the

  (1): well, good luck.   Let's hear your proposal.

    > Example: You say that you have a direct interest in my private
    > contract because it will be enforced by the state, which you support
    > with your taxes, etc.  Nonsense.  My contract is under Japanese law;
    > do you pay taxes here?

  No, but I do believe in international law and human rights.

    > Example: You say enforcement leads to "fascism."  Nonsense.  True, the
    > "nyaah, nyaah, nyaah" argument combined with the right to privacy
    > makes it impossible (at first glance, anyway) to prevent copying and
    > modification without "fascism", but there's nothing "fascist" about
    > enforcing a voluntary contract that says "I won't advertise copies for
    > sale."  Or, if we accept rms as an authority on the boundary of
    > privacy, then a contract which imposes conditions on redistribution
    > between individuals is legitimate; the GPL does that.

  Please see the elaboration of my views on socially responsible
  engineering and society's interests in it above, and then get back
  to me on what parts need to be further elaborated.   I fully admit
  that I glossed over some matters in that section with, in essense, 
  "you don't _really_ need this part explained to you, do you?"
  But maybe you do, and that's fine.

    > It's even possible to contract to allow restricted audits, which would
    > be "fascist" if arbitrarily imposed from outside, of course.

  This might be a topic worth exploring further.   You're idea here is
  a little vague and there are some possible interpretations of it
  that think might be consistent with software freedoms;  there are
  also interpretations that would be clearly inconsistent.

    >     Tom> Should I be reading "yadda yadda yadda" as if it said "Yes,
    >     Tom> you are right, mine was not a valid criticism"?

    > No, it means "same old same old, nothing new here, no need to respond."

  I think that, with replies like this one, I have extended you
  considerable respect.   Reciprocation (which you have not _entirely_
  failed to supply :-) is appreciated.   In the above I did poke fun
  at your use of "yadda yadda yadda" once more but.... overall, let's
  move on, shall we?

    >     > By the time you got to "outrage," you were once again so
    >     > emotional as to forget to explain why.  I still don't get it.

    >     Tom> I forgot to explain?

    > Just guessing.  Deliberate evasion is not your way, and more permanent
    > mental incapacity seems unlikely.  So, why didn't you explain directly
    > why there is a right to software freedom, which is inalienable and
    > therefore universally overrides the right to contract over software?

  I believe I did but I have hopefully more clearly stated the case
  above.  Part of the "problem" is that there are _so_many_ good
  arguments for software freedom, so well rehearsed in the public
  sphere already, that it is impossible to sum them up in a single

    >   * In the rest, the "evidence" adduced is phrased in terms like "this
    >     [voluntary contract] is a relationship of subjugation" with no
    >     explanation whatsoever for the obvious contradiction between the
    >     dictionary definition of subjugation and the voluntariness of the
    >     contract.  "Two movements divided by a common language."

  Hopefully I have begun to make some things more explicit for you.

    >     Tom> endorsed Andrew's "natural rights" argument,

    > I don't think you did, actually, because (1) Andrew doesn't claim that
    > argument (he explicitly says "rights don't exist"), and (2) Andrew is
    > citing anarchists, and you are definitely not an anarchist.  

  It was, indeed, a mistake to attribute that argument to Andrew.  I
  have corrected that mistake above.   It's not unrelated to what
  Andrew was saying, but I shouldn't put words in his mouth.

    > In any case, your explicit explanation of what that argument is, is
    > completely different from a natural rights argument.  Your argument
    > proceeds from another right (privacy == "no fascism"), rather than
    > directly from the nature of human beings.  It's indirect, and false.

  I have a vague sense that you are using the phrases "natural rights
  argument" and "nature of human beings" in some technical sense that
  I am unfamiliar with.

  Let's try to avoid getting too hung up on vocabulary.   If you want
  to relable my "natural rights argument" as a "no facism argument",
  that's probably just fine.   I have no shame from making a "no
  facism argument".

    >     Tom> and outlined some arguments proceeding from the idea that
    >     Tom> there is objectively such a thing as "socially responsible
    >     Tom> engineering".

    > The objective existence of "Tom's opinion as to what is socially
    > responsible engineering" is not disputed.  But I and more illustrious
    > folks like Bill Gates, Larry McVoy, Linus Torvalds, Eric S. Raymond,
    > Russell Nelson, Michael Tiemann, and Bruce Perens hold varied opinions
    > that range from diametrically opposed to nearly coincident.

  You left out "regional zoning boards" from those who have an opinion
  about the nature of "socially responsible engineering."

    > I've seen you make several statements that a given person is "socially
    > irresponsible".  I've seen you make a few statements that a particular
    > practice (sometimes simply "distributing software under proprietary
    > terms") is "socially irresponsible."  I've never seen you describe
    > what body of practices you consider "socially responsible
    > engineering."

  In the above, I have begun to elaborate.

    > So if there's a division in the community, it's not Linus's
    > doing.

  He is most certainly one of the responsible parties because his
  decisions take place in the context of his role and influence over
  the community.

    > A prediction of _fascism_ is "not emotional"?  Has Godwin's Law been
    > repealed?

  Reconsider your proposed censorship against discussion about facism,

    > Calling a respected leader in the community a "poster child for
    > irresponsibility" with _no discussion_ whatsoever is "not emotional"?

  I did that coldly and with calculation specifically to invite
  discussion, which has now commenced.

    > Accusing others of being responsible for division in the community is
    > "not emotional"?

  Nope.  It's an invitation to discussion which has now commenced.

    > What _would_ you call "emotional"?

  "Linus makes me very angry and I don't like him.   I hope others
   will join me in not liking him because he makes me angry.
   His making me angry is a good reason for others to not like him."


reply via email to

[Prev in Thread] Current Thread [Next in Thread]