Changeset 834


Ignore:
Timestamp:
Oct 8, 2006, 3:30:48 PM (19 years ago)
Author:
Александър Шопов
Message:

anarchism: още един абзац. Потъвам бавно в този превод.

File:
1 edited

Legend:

Unmodified
Added
Removed

  • non-gtk/emoglen/anarchism.bg.xml

    r833 r834  
    33]>
    44<article lang="bg">
    5 <title>Триумфиращият анархизъм</title>
    6 <articleinfo>
    7 <releaseinfo>$Id$</releaseinfo>
    8 </articleinfo>
    9 
    10 <!-- <html><head>  -->
    11 <!-- base
    12      href="http://firstmonday.org/issues/issue4_8/moglen/index.html"
    13 -->
    14 
    15 <!--
    16 <meta name="Description" content="This paper shows why free software, far from
    17 being a marginal participant in the commercial software market, is the
    18 first step in the withering away of the intellectual property system.">
    19 <meta name="Keywords" content="anarchism triumphant, free software, death of copyright, Linux operating system kernel, software as property, article">
    20 <meta name="DC.Title" content="Anarchism triumphant">
    21 <meta name="DC.Title" content="Free software and the death of copyright">
    22 <meta name="DC.Creator" content="Moglen, Eben">
    23 <meta name="DC.Subject" content="anarchism triumphant, free software, death of copyright, Linux operating system kernel, software as property, article">
    24 <meta name="DC.Description" content="This paper shows why free software, far from being a marginal participant in the commercial software market, is the
    25 first step in the withering away of the intellectual property system.">
    26 <meta name="DC.Publisher" content="Valauskas, Edward J.">
    27 <meta name="DC.Publisher" content="Dyson, Esther">
    28 <meta name="DC.Publisher" content="Ghosh, Rishab Aiyer">
    29 <meta name="DC.Date" content="1999-08-02">
    30 <meta name="DC.Type" content="text">
    31 <meta name="DC.Format" content="text/html">
    32 <meta name="DC.Identifier" content="http://firstmonday.org/issues/issue4_8/moglen/index.html">
    33 <meta name="DC.Language" content="en">
    34 <meta name="DC.Relation" content="IsPartOf First Monday, vol 4, no. 8"></head><body alink="#ffee99" bgcolor="#ffffff" link="#bb7777" text="#000000" vlink="#7777bb">
     5  <title>Триумфиращият анархизъм</title>
     6  <articleinfo>
     7    <releaseinfo>$Id$</releaseinfo>
     8  </articleinfo>
     9
     10  <!-- <html><head>  -->
     11  <!-- base
     12       href="http://firstmonday.org/issues/issue4_8/moglen/index.html"
     13  -->
     14
     15  <!--
     16       <meta name="Description" content="This paper shows why free software, far from
     17       being a marginal participant in the commercial software market, is the
     18       first step in the withering away of the intellectual property system.">
     19       <meta name="Keywords" content="anarchism triumphant, free software, death of copyright, Linux operating system kernel, software as property, article">
     20       <meta name="DC.Title" content="Anarchism triumphant">
     21       <meta name="DC.Title" content="Free software and the death of copyright">
     22       <meta name="DC.Creator" content="Moglen, Eben">
     23       <meta name="DC.Subject" content="anarchism triumphant, free software, death of copyright, Linux operating system kernel, software as property, article">
     24       <meta name="DC.Description" content="This paper shows why free software, far from being a marginal participant in the commercial software market, is the
     25       first step in the withering away of the intellectual property system.">
     26       <meta name="DC.Publisher" content="Valauskas, Edward J.">
     27       <meta name="DC.Publisher" content="Dyson, Esther">
     28       <meta name="DC.Publisher" content="Ghosh, Rishab Aiyer">
     29       <meta name="DC.Date" content="1999-08-02">
     30       <meta name="DC.Type" content="text">
     31       <meta name="DC.Format" content="text/html">
     32       <meta name="DC.Identifier" content="http://firstmonday.org/issues/issue4_8/moglen/index.html">
     33       <meta name="DC.Language" content="en">
     34       <meta name="DC.Relation" content="IsPartOf First Monday, vol 4, no. 8"></head><body alink="#ffee99" bgcolor="#ffffff" link="#bb7777" text="#000000" vlink="#7777bb">
    3535
    3636<blockquote><img src="anarchism_files/logo.gif" alt="First Monday" align="bottom" border="0" height="40" width="256"><br>
    3737
    3838</blockquote>
    39 -->
    40 
    41 
    42 <para><ulink url="http://firstmonday.org/issues/issue4_8/moglen/index.html#author"><!-- <img src="anarchism_files/moglen.gif" alt="Anarchism Triumphant: Free Software and the Death of Copyright" border="0">--> </ulink></para>
    43 
    44 <blockquote><para>Разпространението на ядрото за операционни системи
    45 Линукс насочи вниманието към движението за свободен софтуер.  Това есе
    46 показва защо свободният софтуер, който далеч не е нищожен участник в
    47 пазара на комерсиален софтуер, е важната първа стъпка в премахването
    48 на системата на интелектуална собственост.</para></blockquote>
    49 
    50 <section>
    51 <title>Софтуерът като собственост: Теоретичният парадокс</title>
    52 
    53 <para><emphasis>Софтуер</emphasis>: никоя друга дума не въплъщава
    54 толкова пълно рактическите и социалните ефекти на цифровата революция.
    55 Първоначално терминът е бил чисто технически и е означавал частите на
    56 една компютърна система, която за разлика от "хардуера" -- направен
    57 непроменим от производителя си в електрониката на системата, е можел
    58 свободно да бъде променян.  Първият софтуер е представлявал начина на
    59 включване на кабели и прекъсвачи на външните панели на електронни
    60 устройства, но още с появата на езикови средства за промяната на
    61 поведението на компютъра, "софтуер" започнал да обозначава предимно
    62 изразяванията в повече или по-малко понятех за хората език, който
    63 както описвал, така и контролирал поведението на машината<footnote>
    64 <para>1. Тази отлика е била само приблизителна в първоначалния
    65 контекст.  В края на 60-те определена част от основните операции на
    66 хардуера са контролирани от програми, които са цифрово кодирани в
    67 електрониката на компютърното оборудване, които не могат да бъдат
    68 променяни веднъж след като продукцията е излязла от фабриката.  Такива
    69 символни, но непроменими компоненти, са били известни като "микрокод"
    70 на жаргона на индустрията, но стана обичайно те да се наричат
    71 "фърмуеър".  Изменчивостта, както бе показано от термина
    72 "фърмуеър"<!-- БЕЛЕЖКА ЗА ЗНАЧЕНИЕТО НА КОРЕНИТЕ НА ДУМИТЕ СОФТУЕР,
    73 ХАРДУЕР, ФЪРМУЕР -->,се отнася главно към възможността на
    74 потребителите да изменят символите, които определят поведението на
    75 машината.  Понеже цифровата революция доведе до широката употреба на
    76 компютрите от технически некомпетентни лица, повечето от традиционния
    77 софтуер -- приложни програми, операционни системи, инструкции за
    78 числово управление и т. н. -- е, за повечето от потребителите си,
    79 фърмуер.  Може да е символен, а не електронен в начина, по който е
    80 направен, но те не могат да го променят, дори и да искат, нещо което
    81 те често, но безсилно и с негодуванние правят.  Това "затвърдяване на
    82 софтуера" е основното условие на собственическия подход към законовата
    83 организация на цифровото обществео, което е темата на този
    84 доклад.</para></footnote>.</para>
    85 
    86 <para>That was then and this is now. Technology based on the
    87 manipulation of digitally-encoded information is now socially dominant
    88 in most aspects of human culture in the "developed" societies
    89 <footnote><para>2. Within the present generation, the very conception
    90 of social "development" is shifting away from possession of heavy
    91 industry based on the internal-combustion engine to "post-industry"
    92 based on digital communications and the related "knowledge-based"
    93 forms of economic activity.</para></footnote>.  The movement from
    94 analog to digital representation - in video, music, printing,
    95 telecommunications, and even choreography, religious worship, and
    96 sexual gratification - potentially turns all forms of human symbolic
    97 activity into software, that is, modifiable instructions for
    98 describing and controlling the behavior of machines. By a conceptual
    99 back-formation characteristic of Western scientistic thinking, the
    100 division between hardware and software is now being observed in the
    101 natural or social world, and has become a new way to express the
    102 conflict between ideas of determinism and free will, nature and
    103 nurture, or genes and culture.  Our "hardware," genetically wired, is
    104 our nature, and determines us. Our nurture is "software," establishes
    105 our cultural programming, which is our comparative freedom. And so on,
    106 for those reckless of blather.<footnote><para>3. Actually, a moment's
    107 thought will reveal, our genes are firmware. Evolution made the
    108 transition from analog to digital before the fossil record begins. But
    109 we haven't possessed the power of controlled direct
    110 modification. Until the day before yesterday. In the next century the
    111 genes too will become software, and while I don't discuss the issue
    112 further in this paper, the political consequences of unfreedom of
    113 software in this context are even more disturbing than they are with
    114 respect to cultural artifacts.</para></footnote> Thus "software"
    115 becomes a viable metaphor for all symbolic activity, apparently
    116 divorced from the technical context of the word's origin, despite the
    117 unease raised in the technically competent when the term is thus
    118 bandied about, eliding the conceptual significance of its
    119 derivation.<footnote><para>4. <emphasis>See, e.g.,</emphasis>
    120 J. M. Balkin, 1998. <emphasis>Cultural Software: a Theory of
    121 Ideology.</emphasis> New Haven: Yale University
    122 Press.</para></footnote></para>
    123 
    124 <para>But the widespread adoption of digital technology for use by
    125 those who do not understand the principles of its operation, while it
    126 apparently licenses the broad metaphoric employment of "software,"
    127 does not in fact permit us to ignore the computers that are now
    128 everywhere underneath our social skin. The movement from analog to
    129 digital is more important for the structure of social and legal
    130 relations than the more famous if less certain movement from status to
    131 contract <footnote><para>5. <emphasis>See</emphasis> Henry Sumner
    132 Maine, 1861. <emphasis>Ancient Law: Its Connection with the Early
    133 History of Society, and Its Relation to Modern Idea.</emphasis> First
    134 edition. London: J. Murray.</para></footnote>.  This is bad news for
    135 those legal thinkers who do not understand it, which is why so much
    136 pretending to understand now goes so floridly on.  Potentially,
    137 however, our great transition is very good news for those who can turn
    138 this new-found land into property for themselves. Which is why the
    139 current "owners" of software so strongly support and encourage the
    140 ignorance of everyone else. Unfortunately for them - for reasons
    141 familiar to legal theorists who haven't yet understood how to apply
    142 their traditional logic in this area - the trick won't work. This
    143 paper explains why<footnote><para>6. In general I dislike the
    144 intrusion of autobiography into scholarship. But because it is here my
    145 sad duty and great pleasure to challenge the qualifications or
    146 <emphasis>bona fides</emphasis> of just about everyone, I must enable
    147 the assessment of my own. I was first exposed to the craft of computer
    148 programming in 1971. I began earning wages as a commercial programmer
    149 in 1973 - at the age of thirteen - and did so, in a variety of
    150 computer services, engineering, and multinational technology
    151 enterprises, until 1985. In 1975 I helped write one of the first
    152 networked e-mail systems in the United States; from 1979 I was engaged
    153 in research and development of advanced computer programming languages
    154 at IBM. These activities made it economically possible for me to study
    155 the arts of historical scholarship and legal cunning. My wages were
    156 sufficient to pay my tuitions, but not - to anticipate an argument
    157 that will be made by the econodwarves further along - because my
    158 programs were the intellectual property of my employer, but rather
    159 because they made the hardware my employer sold work better. Most of
    160 what I wrote was effectively free software, as we shall see. Although
    161 I subsequently made some inconsiderable technical contributions to the
    162 actual free software movement this paper describes, my primary
    163 activities on its behalf have been legal: I have served for the past
    164 five years (without pay, naturally) as general counsel of the Free
    165 Software Foundation.</para></footnote>.</para>
    166 
    167 <para>We need to begin by considering the technical essence of the
    168 familiar devices that surround us in the era of "cultural software." A
    169 CD player is a good example. Its primary input is a bitstream read
    170 from an optical storage disk. The bitstream describes music in terms
    171 of measurements, taken 44,000 times per second, of frequency and
    172 amplitude in each of two audio channels. The player's primary output
    173 is analog audio signals <footnote><para>7. The player, of course, has
    174 secondary inputs and outputs in control channels: buttons or infrared
    175 remote control are input, and time and track display are
    176 output.</para></footnote>. Like everything else in the digital world,
    177 music as seen by a CD player is mere numeric information; a particular
    178 recording of Beethoven's Ninth Symphony recorded by Arturo Toscanini
    179 and the NBC Symphony Orchestra and Chorale is (to drop a few
    180 insignificant digits) 1276749873424, while Glenn Gould's peculiarly
    181 perverse last recording of the Goldberg Variations is (similarly
    182 rather truncated) 767459083268.</para>
    183 
    184 <para>Oddly enough, these two numbers are "copyrighted." This means,
    185 supposedly, that you can't possess another copy of these numbers, once
    186 fixed in any physical form, unless you have licensed them. And you
    187 can't turn 767459083268 into 2347895697 for your friends (thus
    188 correcting Gould's ridiculous judgment about tempi) without making a
    189 "derivative work," for which a license is necessary.</para>
    190 
    191 <para>At the same time, a similar optical storage disk contains
    192 another number, let us call it 7537489532. This one is an algorithm
    193 for linear programming of large systems with multiple constraints,
    194 useful for example if you want to make optimal use of your rolling
    195 stock in running a freight railroad. This number (in the U.S.) is
    196 "patented," which means you cannot derive 7537489532 for yourself, or
    197 otherwise "practice the art" of the patent with respect to solving
    198 linear programming problems no matter how you came by the idea,
    199 including finding it out for yourself, unless you have a license from
    200 the number's owner.</para>
    201 
    202 <para>Then there's 9892454959483. This one is the source code for
    203 Microsoft Word. In addition to being "copyrighted," this one is a
    204 trade secret.  That means if you take this number from Microsoft and
    205 give it to anyone else you can be punished.</para>
    206 
    207 <para>Lastly, there's 588832161316. It doesn't do anything, it's just
    208 the square of 767354. As far as I know, it isn't owned by anybody
    209 under any of these rubrics. Yet.</para>
    210 
    211 <para>At this point we must deal with our first objection from the
    212 learned. It comes from a creature known as the IPdroid. The droid has
    213 a sophisticated mind and a cultured life. It appreciates very much the
    214 elegant dinners at academic and ministerial conferences about the
    215 TRIPs, not to mention the privilege of frequent appearances on MSNBC.
    216 It wants you to know that I'm committing the mistake of confusing the
    217 embodiment with the intellectual property itself. It's not the number
    218 that's patented, stupid, just the Kamarkar algorithm. The number
    219 <emphasis>can</emphasis> be copyrighted, because copyright covers the
    220 expressive qualities of a particular tangible embodiment of an idea
    221 (in which some functional properties may be mysteriously merged,
    222 provided that they're not too merged), but not the algorithm. Whereas
    223 the number isn't patentable, just the "teaching" of the number with
    224 respect to making railroads run on time. And the number representing
    225 the source code of Microsoft Word can be a trade secret, but if you
    226 find it out for yourself (by performing arithmetic manipulation of
    227 other numbers issued by Microsoft, for example, which is known as
    228 "reverse engineering"), you're not going to be punished, at least if
    229 you live in some parts of the United States.</para>
    230 
    231 <para>This droid, like other droids, is often right. The condition of
    232 being a droid is to know everything about something and nothing about
    233 anything else. By its timely and urgent intervention the droid has
    234 established that the current intellectual property system contains
    235 many intricate and ingenious features. The complexities combine to
    236 allow professors to be erudite, Congressmen to get campaign
    237 contributions, lawyers to wear nice suits and tassel loafers, and
    238 Murdoch to be rich.  The complexities mostly evolved in an age of
    239 industrial information distribution, when information was inscribed in
    240 analog forms on physical objects that cost something significant to
    241 make, move, and sell. When applied to digital information that moves
    242 frictionlessly through the network and has zero marginal cost per
    243 copy, everything still works, mostly, as long as you don't stop
    244 squinting.</para>
    245 
    246 <para>But that wasn't what I was arguing about. I wanted to point out
    247 something else: that our world consists increasingly of nothing but
    248 large numbers (also known as bitstreams), and that - for reasons
    249 having nothing to do with emergent properties of the numbers
    250 themselves - the legal system is presently committed to treating
    251 similar numbers radically differently. No one can tell, simply by
    252 looking at a number that is 100 million digits long, whether that
    253 number is subject to patent, copyright, or trade secret protection, or
    254 indeed whether it is "owned" by anyone at all. So the legal system we
    255 have - blessed as we are by its consequences if we are copyright
    256 teachers, Congressmen, Gucci-gulchers or Big Rupert himself - is
    257 compelled to treat indistinguishable things in unlike ways.</para>
    258 
    259 <para>Now, in my role as a legal historian concerned with the secular
    260 (that is, very long term) development of legal thought, I claim that
    261 legal regimes based on sharp but unpredictable distinctions among
    262 similar objects are radically unstable. They fall apart over time
    263 because every instance of the rules' application is an invitation to
    264 at least one side to claim that instead of fitting in ideal category A
    265 the particular object in dispute should be deemed to fit instead in
    266 category B, where the rules will be more favorable to the party making
    267 the claim. This game - about whether a typewriter should be deemed a
    268 musical instrument for purposes of railway rate regulation, or whether
    269 a steam shovel is a motor vehicle - is the frequent stuff of legal
    270 ingenuity. But when the conventionally-approved legal categories
    271 require judges to distinguish among the identical, the game is
    272 infinitely lengthy, infinitely costly, and almost infinitely offensive
    273 to the unbiased bystander <footnote><para>8. This is not an insight
    274 unique to our present enterprise. A closely-related idea forms one of
    275 the most important principles in the history of Anglo-American law,
    276 perfectly put by Toby Milsom in the following terms:</para>
    277 <blockquote><para>The life of the common law has been in the abuse of
    278 its elementary ideas. If the rules of property give what now seems an
    279 unjust answer, try obligation; and equity has proved that from the
    280 materials of obligation you can counterfeit the phenomena of
    281 property. If the rules of contract give what now seems an unjust
    282 answer, try tort. ... If the rules of one tort, say deceit, give what
    283 now seems an unjust answer, try another, try negligence. And so the
    284 legal world goes round.</para></blockquote><para>S.F.C. Milsom,
    285 1981. <emphasis>Historical Foundations of the Common Law.</emphasis>
    286 Second edition. London: Butterworths, p. 6.</para> </footnote>.</para>
    287 
    288 <para>Thus parties can spend all the money they want on all the
    289 legislators and judges they can afford - which for the new "owners" of
    290 the digital world is quite a few - but the rules they buy aren't going
    291 to work in the end. Sooner or later, the paradigms are going to
    292 collapse. Of course, if later means two generations from now, the
    293 distribution of wealth and power sanctified in the meantime may not be
    294 reversible by any course less drastic than a <emphasis>bellum
    295 servile</emphasis> of couch potatoes against media magnates. So
    296 knowing that history isn't on Bill Gates' side isn't enough. We are
    297 predicting the future in a very limited sense: we know that the
    298 existing rules, which have yet the fervor of conventional belief
    299 solidly enlisted behind them, are no longer meaningful. Parties will
    300 use and abuse them freely until the mainstream of "respectable"
    301 conservative opinion acknowledges their death, with uncertain
    302 results. But realistic scholarship should already be turning its
    303 attention to the clear need for new thoughtways.</para>
    304 
    305 <para>When we reach this point in the argument, we find ourselves
    306 contending with the other primary protagonist of educated idiocy: the
    307 econodwarf. Like the IPdroid, the econodwarf is a species of hedgehog,
    308 <footnote><para>9. <emphasis>See</emphasis> Isaiah Berlin,
    309 1953. <emphasis>The Hedgehog and the Fox: An Essay on Tolstoy's View
    310 of History.</emphasis> New York: Simon and Schuster.</para>
    311 </footnote> but where the droid is committed to logic over experience,
    312 the econodwarf specializes in an energetic and well-focused but
    313 entirely erroneous view of human nature. According to the econodwarf's
    314 vision, each human being is an individual possessing "incentives,"
    315 which can be retrospectively unearthed by imagining the state of the
    316 bank account at various times.  So in this instance the econodwarf
    317 feels compelled to object that without the rules I am lampooning,
    318 there would be no incentive to create the things the rules treat as
    319 property: without the ability to exclude others from music there would
    320 be no music, because no one could be sure of getting paid for creating
    321 it.</para>
    322 
    323 <para>Music is not really our subject; the software I am considering
    324 at the moment is the old kind: computer programs. But as he is
    325 determined to deal at least cursorily with the subject, and because,
    326 as we have seen, it is no longer really possible to distinguish
    327 computer programs from music performances, a word or two should be
    328 said. At least we can have the satisfaction of indulging in an
    329 argument <emphasis>ad pygmeam</emphasis>.  When the econodwarf grows
    330 rich, in my experience, he attends the opera.  But no matter how often
    331 he hears <emphasis>Don Giovanni</emphasis> it never occurs to him that
    332 Mozart's fate should, on his logic, have entirely discouraged
    333 Beethoven, or that we have <emphasis>The Magic Flute</emphasis> even
    334 though Mozart knew very well he wouldn't be paid. In fact,
    335 <emphasis>The Magic Flute</emphasis>, <emphasis>St. Matthew's
    336 Passion</emphasis>, and the motets of the wife-murderer Carlo Gesualdo
    337 are all part of the centuries-long tradition of free software, in the
    338 more general sense, which the econodwarf never quite
    339 acknowledges.</para> <!--<center><img
    340 src="anarchism_files/mog1.gif"></center> --> <para> The dwarf's basic
    341 problem is that "incentives" is merely a metaphor, and as a metaphor
    342 to describe human creative activity it's pretty crummy. I have said
    343 this before, <footnote> <para>10. <emphasis>See</emphasis> <ulink
    344 url="http://emoglen.law.columbia.edu/my_pubs/nospeech.html">The
    345 Virtual Scholar and Network Liberation.</ulink></para> </footnote> but
    346 the better metaphor arose on the day Michael Faraday first noticed
    347 what happened when he wrapped a coil of wire around a magnet and spun
    348 the magnet. Current flows in such a wire, but we don't ask what the
    349 incentive is for the electrons to leave home. We say that the current
    350 results from an emergent property of the system, which we call
    351 induction. The question we ask is "what's the resistance of the wire?"
    352 So Moglen's Metaphorical Corollary to Faraday's Law says that if you
    353 wrap the Internet around every person on the planet and spin the
    354 planet, software flows in the network. It's an emergent property of
    355 connected human minds that they create things for one another's
    356 pleasure and to conquer their uneasy sense of being too alone. The
    357 only question to ask is, what's the resistance of the network?
    358 Moglen's Metaphorical Corollary to Ohm's Law states that the
    359 resistance of the network is directly proportional to the field
    360 strength of the "intellectual property" system. So the right answer to
    361 the econodwarf is, resist the resistance.</para>
    362 
    363 <para>Of course, this is all very well in theory. "Resist the
    364 resistance" sounds good, but we'd have a serious problem, theory
    365 notwithstanding, if the dwarf were right and we found ourselves
    366 under-producing good software because we didn't let people own it. But
    367 dwarves and droids are formalists of different kinds, and the
    368 advantage of realism is that if you start from the facts the facts are
    369 always on your side. It turns out that treating software as property
    370 makes bad software.</para>
    371 
    372 </section>
    373 <section>
    374 <title>II. Software as Property: The Practical Problem</title>
    375 
    376 <para>In order to understand why turning software into property
    377 produces bad software, we need an introduction to the history of the
    378 art. In fact, we'd better start with the word "art" itself. The
    379 programming of computers combines determinate reasoning with literary
    380 invention.</para>
    381 
    382 <para>At first glance, to be sure, source code appears to be a
    383 non-literary form of composition <footnote><para>11. Some basic
    384 vocabulary is essential. Digital computers actually execute numerical
    385 instructions: bitstrings that contain information in the "native"
    386 language created by the machine's designers. This is usually referred
    387 to as "machine language." The machine languages of hardware are
    388 designed for speed of execution at the hardware level, and are not
    389 suitable for direct use by human beings. So among the central
    390 components of a computer system are "programming languages," which
    391 translate expressions convenient for humans into machine language. The
    392 most common and relevant, but by no means the only, form of computer
    393 language is a "compiler." The compiler performs static translation, so
    394 that a file containing human-readable instructions, known as "source
    395 code" results in the generation of one or more files of executable
    396 machine language, known as "object code."</para> </footnote>.  The
    397 primary desideratum in a computer program is that it works, that is to
    398 say, performs according to specifications formally describing its
    399 outputs in terms of its inputs. At this level of generality, the
    400 functional content of programs is all that can be seen.</para>
    401 
    402 <para>But working computer programs exist as parts of computer
    403 systems, which are interacting collections of hardware, software, and
    404 human beings. The human components of a computer system include not
    405 only the users, but also the (potentially different) persons who
    406 maintain and improve the system. Source code not only communicates
    407 with the computer that executes the program, through the intermediary
    408 of the compiler that produces machine-language object code, but also
    409 with other programmers.</para>
    410 
    411 <para>The function of source code in relation to other human beings is
    412 not widely grasped by non-programmers, who tend to think of computer
    413 programs as incomprehensible. They would be surprised to learn that
    414 the bulk of information contained in most programs is, from the point
    415 of view of the compiler or other language processor, "comment," that
    416 is, non-functional material. The comments, of course, are addressed to
    417 others who may need to fix a problem or to alter or enhance the
    418 program's operation. In most programming languages, far more space is
    419 spent in telling people what the program does than in telling the
    420 computer how to do it.</para>
    421 
    422 <para>The design of programming languages has always proceeded under
    423 the dual requirements of complete specification for machine execution
    424 and informative description for human readers. One might identify
    425 three basic strategies in language design for approaching this dual
    426 purpose.  The first, pursued initially with respect to the design of
    427 languages specific to particular hardware products and collectively
    428 known as "assemblers," essentially separated the human- and
    429 machine-communication portions of the program. Assembler instructions
    430 are very close relatives of machine-language instructions: in general,
    431 one line of an assembler program corresponds to one instruction in the
    432 native language of the machine. The programmer controls machine
    433 operation at the most specific possible level, and (if
    434 well-disciplined) engages in running commentary alongside the machine
    435 instructions, pausing every few hundred instructions to create "block
    436 comments," which provide a summary of the strategy of the program, or
    437 document the major data structures the program manipulates.</para>
    438 
    439 <para>A second approach, characteristically depicted by the language
    440 COBOL (which stood for "Common Business-Oriented Language"), was to
    441 make the program itself look like a set of natural language
    442 directions, written in a crabbed but theoretically human-readable
    443 style. A line of COBOL code might say, for example "MULTIPLY PRICE
    444 TIMES QUANTITY GIVING EXPANSION." At first, when the Pentagon and
    445 industry experts began the joint design of COBOL in the early 1960's,
    446 this seemed a promising approach. COBOL programs appeared largely
    447 self-documenting, allowing both the development of work teams able to
    448 collaborate on the creation of large programs, and the training of
    449 programmers who, while specialized workers, would not need to
    450 understand the machine as intimately as assembler programs had to. But
    451 the level of generality at which such programs documented themselves
    452 was wrongly selected. A more formulaic and compressed expression of
    453 operational detail "expansion = price x quantity," for example, was
    454 better suited even to business and financial applications where the
    455 readers and writers of programs were accustomed to mathematical
    456 expression, while the processes of describing both data structures and
    457 the larger operational context of the program were not rendered
    458 unnecessary by the wordiness of the language in which the details of
    459 execution were specified.</para>
    460 
    461 <para>Accordingly, language designers by the late 1960s began
    462 experimenting with forms of expression in which the blending of
    463 operational details and non-functional information necessary for
    464 modification or repair was more subtle. Some designers chose the path
    465 of highly symbolic and compressed languages, in which the programmer
    466 manipulated data abstractly, so that "A x B" might mean the
    467 multiplication of two integers, two complex numbers, two vast arrays,
    468 or any other data type capable of some process called
    469 "multiplication," to be undertaken by the computer on the basis of the
    470 context for the variables "A" and "B" at the moment of execution
    471 <footnote> <para>12. This, I should say, was the path that most of my
    472 research and development followed, largely in connection with a
    473 language called APL ("A Programming Language") and its successors. It
    474 was not, however, the ultimately-dominant approach, for reasons that
    475 will be suggested below.</para> </footnote> .  Because this approach
    476 resulted in extremely concise programs, it was thought, the problem of
    477 making code comprehensible to those who would later seek to modify or
    478 repair it was simplified. By hiding the technical detail of computer
    479 operation and emphasizing the algorithm, languages could be devised
    480 that were better than English or other natural languages for the
    481 expression of stepwise processes. Commentary would be not only
    482 unnecessary but distracting, just as the metaphors used to convey
    483 mathematical concepts in English do more to confuse than to
    484 enlighten.</para>
    485 
    486 <section>
    487 <title>How We Created the Microbrain Mess</title>
    488 
    489 <para>Thus the history of programming languages directly reflected the
    490 need to find forms of human-machine communication that were also
    491 effective in conveying complex ideas to human readers. "Expressivity"
    492 became a property of programming languages, not because it facilitated
    493 computation, but because it facilitated the collaborative creation and
    494 maintenance of increasingly complex software systems.</para>
    495 
    496 <para>At first impression, this seems to justify the application of
    497 traditional copyright thinking to the resulting works. Though
    498 substantially involving "functional" elements, computer programs
    499 contained "expressive" features of paramount importance. Copyright
    500 doctrine recognized the merger of function and expression as
    501 characteristic of many kinds of copyrighted works. "Source code,"
    502 containing both the machine instructions necessary for functional
    503 operation and the expressive "commentary" intended for human readers,
    504 was an appropriate candidate for copyright treatment.</para>
    505 
    506 <para>True, so long as it is understood that the expressive component
    507 of software was present solely in order to facilitate the making of
    508 "derivative works." Were it not for the intention to facilitate
    509 alteration, the expressive elements of programs would be entirely
    510 supererogatory, and source code would be no more copyrightable than
    511 object code, the output of the language processor, purged of all but
    512 the program's functional characteristics.</para>
    513 
    514 <para>The state of the computer industry throughout the 1960's and
    515 1970's, when the grundnorms of sophisticated computer programming were
    516 established, concealed the tension implicit in this situation. In that
    517 period, hardware was expensive. Computers were increasingly large and
    518 complex collections of machines, and the business of designing and
    519 building such an array of machines for general use was dominated, not
    520 to say monopolized, by one firm. IBM gave away its software. To be
    521 sure, it owned the programs its employees wrote, and it copyrighted
    522 the source code. But it also distributed the programs - including the
    523 source code - to its customers at no additional charge, and encouraged
    524 them to make and share improvements or adaptations of the programs
    525 thus distributed. For a dominant hardware manufacturer, this strategy
    526 made sense: better programs sold more computers, which is where the
    527 profitability of the business rested.</para>
    528 
    529 <para>Computers, in this period, tended to aggregate within particular
    530 organizations, but not to communicate broadly with one another. The
    531 software needed to operate was distributed not through a network, but
    532 on spools of magnetic tape. This distribution system tended to
    533 centralize software development, so that while IBM customers were free
    534 to make modifications and improvements to programs, those
    535 modifications were shared in the first instance with IBM, which then
    536 considered whether and in what way to incorporate those changes in the
    537 centrally-developed and distributed version of the software. Thus in
    538 two important senses the best computer software in the world was free:
    539 it cost nothing to acquire, and the terms on which it was furnished
    540 both allowed and encouraged experimentation, change, and improvement
    541 <footnote><para>13. This description elides some details. By the
    542 mid-1970's IBM had acquired meaningful competition in the mainframe
    543 computer business, while the large-scale antitrust action brought
    544 against it by the U.S. government prompted the decision to "unbundle,"
    545 or charge separately, for software. In this less important sense,
    546 software ceased to be free. But - without entering into the now-dead
    547 but once-heated controversy over IBM's software pricing policies - the
    548 unbundling revolution had less effect on the social practices of
    549 software manufacture than might be supposed. As a fellow responsible
    550 for technical improvement of one programming language product at IBM
    551 from 1979 to 1984, for example, I was able to treat the product as
    552 "almost free," that is, to discuss with users the changes they had
    553 proposed or made in the programs, and to engage with them in
    554 cooperative development of the product for the benefit of all
    555 users.</para> </footnote>.  That the software in question was IBM's
    556 property under prevailing copyright law certainly established some
    557 theoretical limits on users' ability to distribute their improvements
    558 or adaptations to others, but in practice mainframe software was
    559 cooperatively developed by the dominant hardware manufacturer and its
    560 technically-sophisticated users, employing the manufacturer's
    561 distribution resources to propagate the resulting improvements through
    562 the user community. The right to exclude others, one of the most
    563 important "sticks in the bundle" of property rights (in an image
    564 beloved of the United States Supreme Court), was practically
    565 unimportant, or even undesirable, at the heart of the software
    566 business <footnote> <para>14. This description is highly compressed,
    567 and will seem both overly simplified and unduly rosy to those who also
    568 worked in the industry during this period of its
    569 development. Copyright protection of computer software was a
    570 controversial subject in the 1970's, leading to the famous CONTU
    571 commission and its mildly pro-copyright recommendations of 1979. And
    572 IBM seemed far less cooperative to its users at the time than this
    573 sketch makes out. But the most important element is the contrast with
    574 the world created by the PC, the Internet, and the dominance of
    575 Microsoft, with the resulting impetus for the free software movement,
    576 and I am here concentrating on the features that express that
    577 contrast.</para></footnote>.</para>
    578 
    579 <para>After 1980, everything was different. The world of mainframe
    580 hardware gave way within ten years to the world of the commodity PC.
    581 And, as a contingency of the industry's development, the single most
    582 important element of the software running on that commodity PC, the
    583 operating system, became the sole significant product of a company
    584 that made no hardware. High-quality basic software ceased to be part
    585 of the product-differentiation strategy of hardware
    586 manufacturers. Instead, a firm with an overwhelming share of the
    587 market, and with the near-monopolist's ordinary absence of interest in
    588 fostering diversity, set the practices of the software industry. In
    589 such a context, the right to exclude others from participation in the
    590 product's formation became profoundly important. Microsoft's power in
    591 the market rested entirely on its ownership of the Windows source
    592 code.</para>
    593 
    594 <para>To Microsoft, others' making of "derivative works," otherwise
    595 known as repairs and improvements, threatened the central asset of the
    596 business. Indeed, as subsequent judicial proceedings have tended to
    597 establish, Microsoft's strategy as a business was to find innovative
    598 ideas elsewhere in the software marketplace, buy them up and either
    599 suppress them or incorporate them in its proprietary product. The
    600 maintenance of control over the basic operation of computers
    601 manufactured, sold, possessed, and used by others represented profound
    602 and profitable leverage over the development of the culture <footnote>
    603 <para>15. I discuss the importance of PC software in this context, the
    604 evolution of "the market for eyeballs" and "the sponsored life" in
    605 other chapters of my forthcoming book, <emphasis>The Invisible
    606 Barbecue</emphasis>, of which this essay forms a part.</para>
    607 </footnote>.; the right to exclude returned to center stage in the
    608 concept of software as property.</para>
    609 
    610 <para>The result, so far as the quality of software was concerned, was
    611 disastrous. The monopoly was a wealthy and powerful corporation that
    612 employed a large number of programmers, but it could not possibly
    613 afford the number of testers, designers, and developers required to
    614 produce flexible, robust and technically-innovative software
    615 appropriate to the vast array of conditions under which increasingly
    616 ubiquitous personal computers operated. Its fundamental marketing
    617 strategy involved designing its product for the least
    618 technically-sophisticated users, and using "fear, uncertainty, and
    619 doubt" (known within Microsoft as "FUD") to drive sophisticated users
    620 away from potential competitors, whose long-term survivability in the
    621 face of Microsoft's market power was always in question.</para>
    622 
    623 <para>Without the constant interaction between users able to repair
    624 and improve and the operating system's manufacturer, the inevitable
    625 deterioration of quality could not be arrested. But because the
    626 personal computer revolution expanded the number of users
    627 exponentially, almost everyone who came in contact with the resulting
    628 systems had nothing against which to compare them. Unaware of the
    629 standards of stability, reliability, maintainability and effectiveness
    630 that had previously been established in the mainframe world, users of
    631 personal computers could hardly be expected to understand how badly,
    632 in relative terms, the monopoly's software functioned. As the power
    633 and capacity of personal computers expanded rapidly, the defects of
    634 the software were rendered less obvious amidst the general increase of
    635 productivity. Ordinary users, more than half afraid of the technology
    636 they almost completely did not understand, actually welcomed the
    637 defectiveness of the software. In an economy undergoing mysterious
    638 transformations, with the concomitant destabilization of millions of
    639 careers, it was tranquilizing, in a perverse way, that no personal
    640 computer seemed to be able to run for more than a few consecutive
    641 hours without crashing. Although it was frustrating to lose work in
    642 progress each time an unnecessary failure occurred, the evident
    643 fallibility of computers was intrinsically reassuring <footnote>
    644 <para>16. This same pattern of ambivalence, in which bad programming
    645 leading to widespread instability in the new technology is
    646 simultaneously frightening and reassuring to technical incompetents,
    647 can be seen also in the primarily-American phenomenon of Y2K
    648 hysteria.</para> </footnote> .</para>
    649 
    650 <para>None of this was necessary. The low quality of personal computer
    651 software could have been reversed by including users directly in the
    652 inherently evolutionary process of software design and implementation.
    653 A Lamarckian mode, in which improvements could be made anywhere, by
    654 anyone, and inherited by everyone else, would have wiped out the
    655 deficit, restoring to the world of the PC the stability and
    656 reliability of the software made in the quasi-propertarian environment
    657 of the mainframe era. But the Microsoft business model precluded
    658 Lamarckian inheritance of software improvements. Copyright doctrine,
    659 in general and as it applies to software in particular, biases the
    660 world towards creationism; in this instance, the problem is that BillG
    661 the Creator was far from infallible, and in fact he wasn't even
    662 trying.</para> <!--<center><img src="anarchism_files/mog2.gif"
    663 hspace="0" vspace="0"></center>--> <para>To make the irony more
    664 severe, the growth of the network rendered the non-propertarian
    665 alternative even more practical. What scholarly and popular writing
    666 alike denominate as a thing ("the Internet") is actually the name of a
    667 social condition: the fact that everyone in the network society is
    668 connected directly, without intermediation, to everyone else
    669 <footnote> <para>17. The critical implications of this simple
    670 observation about our metaphors are worked out in "How Not to Think
    671 about 'The Internet'," in <emphasis>The Invisible Barbecue</emphasis>,
    672 forthcoming.</para> </footnote>. The global interconnection of
    673 networks eliminated the bottleneck that had required a centralized
    674 software manufacturer to rationalize and distribute the outcome of
    675 individual innovation in the era of the mainframe.</para>
    676 
    677 <para>And so, in one of history's little ironies, the global triumph
    678 of bad software in the age of the PC was reversed by a surprising
    679 combination of forces: the social transformation initiated by the
    680 network, a long-discarded European theory of political economy, and a
    681 small band of programmers throughout the world mobilized by a single
    682 simple idea.</para>
    683 
    684 </section>
    685 <section>
    686 
    687 <title>Software Wants to Be Free; or, How We Stopped Worrying and
    688 Learned to Love the Bomb</title>
    689 
    690 <para>Long before the network of networks was a practical reality,
    691 even before it was an aspiration, there was a desire for computers to
    692 operate on the basis of software freely available to everyone. This
    693 began as a reaction against propertarian software in the mainframe
    694 era, and requires another brief historical digression.</para>
    695 
    696 <para>Even though IBM was the largest seller of general purpose
    697 computers in the mainframe era, it was not the largest designer and
    698 builder of such hardware. The telephone monopoly, American Telephone
    699 &amp; Telegraph, was in fact larger than IBM, but it consumed its
    700 products internally. And at the famous Bell Labs research arm of the
    701 telephone monopoly, in the late 1960's, the developments in computer
    702 languages previously described gave birth to an operating system
    703 called Unix.</para>
    704 
    705 <para>The idea of Unix was to create a single, scalable operating
    706 system to exist on all the computers, from small to large, that the
    707 telephone monopoly made for itself. To achieve this goal meant writing
    708 an operating system not in machine language, nor in an assembler whose
    709 linguistic form was integral to a particular hardware design, but in a
    710 more expressive and generalized language. The one chosen was also a
    711 Bell Labs invention, called "C" <footnote> <para>18. Technical readers
    712 will again observe that this compresses developments occurring from
    713 1969 through 1973.</para> </footnote>. The C language became common,
    714 even dominant, for many kinds of programming tasks, and by the late
    715 1970's the Unix operating system written in that language had been
    716 transferred (or "ported," in professional jargon) to computers made by
    717 many manufacturers and of many designs.</para>
    718 
    719 <para>AT&amp;T distributed Unix widely, and because of the very design
    720 of the operating system, it had to make that distribution in C source
    721 code.  But AT&amp;T retained ownership of the source code and
    722 compelled users to purchase licenses that prohibited redistribution
    723 and the making of derivative works. Large computing centers, whether
    724 industrial or academic, could afford to purchase such licenses, but
    725 individuals could not, while the license restrictions prevented the
    726 community of programmers who used Unix from improving it in an
    727 evolutionary rather than episodic fashion. And as programmers
    728 throughout the world began to aspire to and even expect a personal
    729 computer revolution, the "unfree" status of Unix became a source of
    730 concern.</para>
    731 
    732 <para>Between 1981 and 1984, one man envisioned a crusade to change
    733 the situation. Richard M. Stallman, then an employee of MIT's
    734 Artificial Intelligence Laboratory, conceived the project of
    735 independent, collaborative redesign and implementation of an operating
    736 system that would be true free software. In Stallman's phrase, free
    737 software would be a matter of freedom, not of price. Anyone could
    738 freely modify and redistribute such software, or sell it, subject only
    739 to the restriction that he not try to reduce the rights of others to
    740 whom he passed it along. In this way free software could become a
    741 self-organizing project, in which no innovation would be lost through
    742 proprietary exercises of rights. The system, Stallman decided, would
    743 be called GNU, which stood (in an initial example of a taste for
    744 recursive acronyms that has characterized free software ever since),
    745 for "GNU's Not Unix."  Despite misgivings about the fundamental design
    746 of Unix, as well as its terms of distribution, GNU was intended to
    747 benefit from the wide if unfree source distribution of Unix. Stallman
    748 began Project GNU by writing components of the eventual system that
    749 were also designed to work without modification on existing Unix
    750 systems. Development of the GNU tools could thus proceed directly in
    751 the environment of university and other advanced computing centers
    752 around the world.</para>
    753 
    754 <para>The scale of such a project was immense. Somehow, volunteer
    755 programmers had to be found, organized, and set to work building all
    756 the tools that would be necessary for the ultimate construction.
    757 Stallman himself was the primary author of several fundamental tools.
    758 Others were contributed by small or large teams of programmers
    759 elsewhere, and assigned to Stallman's project or distributed
    760 directly. A few locations around the developing network became
    761 archives for the source code of these GNU components, and throughout
    762 the 1980's the GNU tools gained recognition and acceptance by Unix
    763 users throughout the world. The stability, reliability, and
    764 maintainability of the GNU tools became a by-word, while Stallman's
    765 profound abilities as a designer continued to outpace, and provide
    766 goals for, the evolving process. The award to Stallman of a MacArthur
    767 Fellowship in 1990 was an appropriate recognition of his conceptual
    768 and technical innovations and their social consequences.</para>
    769 
    770 <para>Project GNU, and the Free Software Foundation to which it gave
    771 birth in 1985, were not the only source of free software
    772 ideas. Several forms of copyright license designed to foster free or
    773 partially free software began to develop in the academic community,
    774 mostly around the Unix environment. The University of California at
    775 Berkeley began the design and implementation of another version of
    776 Unix for free distribution in the academic community. BSD Unix, as it
    777 came to be known, also treated AT&amp;T's Unix as a design
    778 standard. The code was broadly released and constituted a reservoir of
    779 tools and techniques, but its license terms limited the range of its
    780 application, while the elimination of hardware-specific proprietary
    781 code from the distribution meant that no one could actually build a
    782 working operating system for any particular computer from BSD. Other
    783 university-based work also eventuated in quasi-free software; the
    784 graphical user interface (or GUI) for Unix systems called X Windows,
    785 for example, was created at MIT and distributed with source code on
    786 terms permitting free modification. And in 1989-1990, an undergraduate
    787 computer science student at the University of Helsinki, Linus
    788 Torvalds, began the project that completed the circuit and fully
    789 energized the free software vision.</para>
    790 
    791 <para>What Torvalds did was to begin adapting a computer science
    792 teaching tool for real life use. Andrew Tannenbaum's MINIX kernel
    793 <footnote> <para>19. Operating systems, even Windows (which hides the
    794 fact from its users as thoroughly as possible), are actually
    795 collections of components, rather than undivided unities. Most of what
    796 an operating system does (manage file systems, control process
    797 execution, etc.) can be abstracted from the actual details of the
    798 computer hardware on which the operating system runs. Only a small
    799 inner core of the system must actually deal with the eccentric
    800 peculiarities of particular hardware.  Once the operating system is
    801 written in a general language such as C, only that inner core, known
    802 in the trade as the kernel, will be highly specific to a particular
    803 computer architecture.</para> </footnote> , was a staple of Operating
    804 Systems courses, providing an example of basic solutions to basic
    805 problems. Slowly, and at first without recognizing the intention,
    806 Linus began turning the MINIX kernel into an actual kernel for Unix on
    807 the Intel x86 processors, the engines that run the world's commodity
    808 PCs. As Linus began developing this kernel, which he named Linux, he
    809 realized that the best way to make his project work would be to adjust
    810 his design decisions so that the existing GNU components would be
    811 compatible with his kernel.</para>
    812 
    813 <para>The result of Torvalds' work was the release on the net in 1991
    814 of a sketchy working model of a free software kernel for a Unix-like
    815 operating system for PCs, fully compatible with and designed
    816 convergently with the large and high-quality suite of system
    817 components created by Stallman's Project GNU and distributed by the
    818 Free Software Foundation. Because Torvalds chose to release the Linux
    819 kernel under the Free Software Foundation's General Public License, of
    820 which more below, the hundreds and eventually thousands of programmers
    821 around the world who chose to contribute their effort towards the
    822 further development of the kernel could be sure that their efforts
    823 would result in permanently free software that no one could turn into
    824 a proprietary product. Everyone knew that everyone else would be able
    825 to test, improve, and redistribute their improvements. Torvalds
    826 accepted contributions freely, and with a genially effective style
    827 maintained overall direction without dampening enthusiasm. The
    828 development of the Linux kernel proved that the Internet made it
    829 possible to aggregate collections of programmers far larger than any
    830 commercial manufacturer could afford, joined almost non-hierarchically
    831 in a development project ultimately involving more than one million
    832 lines of computer code - a scale of collaboration among geographically
    833 dispersed unpaid volunteers previously unimaginable in human history
    834 <footnote> <para>20. A careful and creative analysis of how Torvalds
    835 made this process work, and what it implies for the social practices
    836 of creating software, was provided by Eric S. Raymond in his seminal
    837 1997 paper, <ulink
    838 url="http://firstmonday.org/issues/issue3_3/raymond/index.html">The
    839 Cathedral and the Bazaar,</ulink> which itself played a significant
    840 role in the expansion of the free software idea.</para>
    841 </footnote>.</para>
    842 
    843 <para>By 1994, Linux had reached version 1.0, representing a usable
    844 production kernel. Level 2.0 was reached in 1996, and by 1998, with
    845 the kernel at 2.2.0 and available not only for x86 machines but for a
    846 variety of other machine architectures, GNU/Linux - the combination of
    847 the Linux kernel and the much larger body of Project GNU components -
    848 and Windows NT were the only two operating systems in the world
    849 gaining market share. A Microsoft internal assessment of the situation
    850 leaked in October 1998 and subsequently acknowledged by the company as
    851 genuine concluded that "Linux represents a best-of-breed UNIX, that is
    852 trusted in mission critical applications, and - due to it's [sic] open
    853 source code - has a long term credibility which exceeds many other
    854 competitive OS's." <footnote> <para>21. This is a quotation from what
    855 is known in the trade as the "Halloween memo," which can be found, as
    856 annotated by Eric Raymond, to whom it was leaked, at <ulink
    857 url="http://www.opensource.org/halloween/halloween1.html">
    858 http://www.opensource.org/halloween/halloween1.html</ulink>.</para></footnote>
    859 GNU/Linux systems are now used throughout the world, operating
    860 everything from Web servers at major electronic commerce sites to
    861 "ad-hoc supercomputer" clusters to the network infrastructure of
    862 money-center banks. GNU/Linux is found on the space shuttle, and
    863 running behind-the-scenes computers at (yes) Microsoft. Industry
    864 evaluations of the comparative reliability of Unix systems have
    865 repeatedly shown that Linux is far and away the most stable and
    866 reliable Unix kernel, with a reliability exceeded only by the GNU
    867 tools themselves. GNU/Linux not only out-performs commercial
    868 proprietary Unix versions for PCs in benchmarks, but is renowned for
    869 its ability to run, undisturbed and uncomplaining, for months on end
    870 in high-volume high-stress environments without crashing.</para>
    871 
    872 <para>Other components of the free software movement have been equally
    873 successful. Apache, far and away the world's leading Web server
    874 program, is free software, as is Perl, the programming language which
    875 is the lingua franca for the programmers who build sophisticated Web
    876 sites. Netscape Communications now distributes its Netscape
    877 Communicator 5.0 browser as free software, under a close variant of
    878 the Free Software Foundation's General Public License. Major PC
    879 manufacturers, including IBM, have announced plans or are already
    880 distributing GNU/Linux as a customer option on their top-of-the-line
    881 PCs intended for use as Web- and file servers. Samba, a program that
    882 allows GNU/Linux computers to act as Windows NT file servers, is used
    883 worldwide as an alternative to Windows NT Server, and provides
    884 effective low-end competition to Microsoft in its own home market. By
    885 the standards of software quality that have been recognized in the
    886 industry for decades - and whose continuing relevance will be clear to
    887 you the next time your Windows PC crashes - the news at century's end
    888 is unambiguous. The world's most profitable and powerful corporation
    889 comes in a distant second, having excluded all but the real victor
    890 from the race. Propertarianism joined to capitalist vigor destroyed
    891 meaningful commercial competition, but when it came to making good
    892 software, anarchism won.</para>
    893 
    894 
    895 </section>
    896 </section>
    897 <!--<para><img src="anarchism_files/quad.gif"></para><a name="m3"></a>-->
    898 <section>
    899 <title>III. Anarchism as a Mode of Production</title>
    900  
    901 <para>It's a pretty story, and if only the IPdroid and the econodwarf
    902 hadn't been blinded by theory, they'd have seen it coming. But though
    903 some of us had been working for it and predicting it for years, the
    904 theoretical consequences are so subversive for the thoughtways that
    905 maintain our dwarves and droids in comfort that they can hardly be
    906 blamed for refusing to see. The facts proved that something was wrong
    907 with the "incentives" metaphor that underprops conventional
    908 intellectual property reasoning <footnote> <para>22. As recently as
    909 early 1994 a talented and technically competent (though Windows-using)
    910 law and economics scholar at a major U.S. law school confidently
    911 informed me that free software couldn't possibly exist, because no one
    912 would have any incentive to make really sophisticated programs
    913 requiring substantial investment of effort only to give them
    914 away.</para> </footnote> . But they did more. They provided an initial
    915 glimpse into the future of human creativity in a world of global
    916 interconnection, and it's not a world made for dwarves and
    917 droids.</para>
    918 
    919 <para>My argument, before we paused for refreshment in the real world,
    920 can be summarized this way: Software - whether executable programs,
    921 music, visual art, liturgy, weaponry, or what have you - consists of
    922 bitstreams, which although essentially indistinguishable are treated
    923 by a confusing multiplicity of legal categories. This multiplicity is
    924 unstable in the long term for reasons integral to the legal process.
    925 The unstable diversity of rules is caused by the need to distinguish
    926 among kinds of property interests in bitstreams. This need is
    927 primarily felt by those who stand to profit from the socially
    928 acceptable forms of monopoly created by treating ideas as
    929 property. Those of us who are worried about the social inequity and
    930 cultural hegemony created by this intellectually unsatisfying and
    931 morally repugnant regime are shouted down. Those doing the shouting,
    932 the dwarves and the droids, believe that these property rules are
    933 necessary not from any overt yearning for life in Murdochworld -
    934 though a little luxurious co-optation is always welcome - but because
    935 the metaphor of incentives, which they take to be not just an image
    936 but an argument, proves that these rules - despite their lamentable
    937 consequences - are necessary if we are to make good software. The only
    938 way to continue to believe this is to ignore the facts. At the center
    939 of the digital revolution, with the executable bitstreams that make
    940 everything else possible, propertarian regimes not only do not make
    941 things better, they can make things radically worse. Property
    942 concepts, whatever else may be wrong with them, do not enable and have
    943 in fact retarded progress.</para>
    944 
    945 <para>
    946 But what is this mysterious alternative? Free software exists, but
    947 what are its mechanisms, and how does it generalize towards a
    948 non-propertarian theory of the digital society?</para>
    949 
    950 </section>
    951 <section>
    952 
    953 <title>The Legal Theory of Free Software</title>
    954 
    955 <para>There is a myth, like most myths partially founded on reality,
    956 that computer programmers are all libertarians. Right-wing ones are
    957 capitalists, cleave to their stock options, and disdain taxes, unions,
    958 and civil rights laws; left-wing ones hate the market and all
    959 government, believe in strong encryption no matter how much nuclear
    960 terrorism it may cause, <footnote> <para>23. This question too
    961 deserves special scrutiny, encrusted as it is with special pleading on
    962 the state-power side. See my brief essay <ulink
    963 url="http://emoglen.law.columbia.edu/my_pubs/yu-encrypt.html">"<emphasis>So
    964 Much for Savages</emphasis>: Navajo 1, Government 0 in Final Moments of
    965 Play."</ulink></para> </footnote> and dislike Bill Gates because he's
    966 rich. There is doubtless a foundation for this belief. But the most
    967 significant difference between political thought inside the digirati
    968 and outside it is that in the network society, anarchism (or more
    969 properly, anti-possessive individualism) is a viable political
    970 philosophy.</para>
    971 
    972 <para>The center of the free software movement's success, and the
    973 greatest achievement of Richard Stallman, is not a piece of computer
    974 code. The success of free software, including the overwhelming success
    975 of GNU/Linux, results from the ability to harness extraordinary
    976 quantities of high-quality effort for projects of immense size and
    977 profound complexity. And this ability in turn results from the legal
    978 context in which the labor is mobilized. As a visionary designer
    979 Richard Stallman created more than Emacs, GDB, or GNU. He created the
    980 General Public License.</para>
    981 
    982 <!-- <center><img src="anarchism_files/mog3.gif" hspace="0"
    983 vspace="0"></center> --> <para>The GPL, <footnote>
    984 <para>24. <emphasis>See</emphasis> <ulink
    985 url="http://www.fsf.org/copyleft/gpl.txt">GNU General Public License,
    986 Version 2, June 1991.</ulink></para> </footnote> also known as the
    987 copyleft, uses copyright, to paraphrase Toby Milsom, to counterfeit
    988 the phenomena of anarchism. As the license preamble expresses
    989 it:</para>
    990 
    991 <blockquote><para>When we speak of free software, we are referring to
    992 freedom, not price. Our General Public Licenses are designed to make
    993 sure that you have the freedom to distribute copies of free software
    994 (and charge for this service if you wish), that you receive source
    995 code or can get it if you want it, that you can change the software or
    996 use pieces of it in new free programs; and that you know you can do
    997 these things.</para>
    998 
    999 <para>To protect your rights, we need to make restrictions that
    1000 forbid anyone to deny you these rights or to ask you to surrender the
    1001 rights.  These restrictions translate to certain responsibilities for
    1002 you if you distribute copies of the software, or if you modify
    1003 it.</para>
    1004 
    1005 <para>For example, if you distribute copies of such a program,
    1006 whether gratis or for a fee, you must give the recipients all the
    1007 rights that you have. You must make sure that they, too, receive or
    1008 can get the source code. And you must show them these terms so they
    1009 know their rights.</para>
    1010 
    1011 <para>Many variants of this basic free software idea have been
    1012 expressed in licenses of various kinds, as I have already
    1013 indicated. The GPL is different from the other ways of expressing
    1014 these values in one crucial respect. Section 2 of the license provides
    1015 in pertinent part:</para>
    1016 
    1017 <para>You may modify your copy or copies of the Program or any
    1018 portion of it, thus forming a work based on the Program, and copy and
    1019 distribute such modifications or work ..., provided that you also meet
    1020 all of these conditions: </para>
    1021 
    1022 <para>...</para>
    1023 
    1024 <para>b) You must cause any work that you distribute or publish,
    1025 that in whole or in part contains or is derived from the Program or
    1026 any part thereof, to be licensed as a whole at no charge to all third
    1027 parties under the terms of this License.</para></blockquote>
    1028 
    1029 <para>Section 2(b) of the GPL is sometimes called "restrictive," but
    1030 its intention is liberating. It creates a commons, to which anyone may
    1031 add but from which no one may subtract. Because of §2(b), each
    1032 contributor to a GPL'd project is assured that she, and all other
    1033 users, will be able to run, modify and redistribute the program
    1034 indefinitely, that source code will always be available, and that,
    1035 unlike commercial software, its longevity cannot be limited by the
    1036 contingencies of the marketplace or the decisions of future
    1037 developers. This "inheritance" of the GPL has sometimes been
    1038 criticized as an example of the free software movement's
    1039 anti-commercial bias.  Nothing could be further from the truth. The
    1040 effect of §2(b) is to make commercial distributors of free software
    1041 better competitors against proprietary software businesses. For
    1042 confirmation of this point, one can do no better than to ask the
    1043 proprietary competitors. As the author of the Microsoft "Halloween"
    1044 memorandum, Vinod Vallopillil, put it:</para>
    1045 
    1046 <blockquote><para>The GPL and its aversion to code forking reassures
    1047 customers that they aren't riding an evolutionary `dead-end' by
    1048 subscribing to a particular commercial version of Linux.</para>
    1049 
    1050 <para>The "evolutionary dead-end" is the core of the software
    1051 FUD argument <footnote> <para>25. <ulink
    1052 url="http://www.opensource.org/halloween/halloween1.html">V. Vallopillil,
    1053 Open Source Software: A (New?) Development Methodology.</ulink></para>
    1054 </footnote> .</para></blockquote>
    1055 
    1056 <para>Translated out of Microspeak, this means that the strategy by
    1057 which the dominant proprietary manufacturer drives customers away from
    1058 competitors - by sowing fear, uncertainty and doubt about other
    1059 software's long-term viability - is ineffective with respect to GPL'd
    1060 programs. Users of GPL'd code, including those who purchase software
    1061 and systems from a commercial reseller, know that future improvements
    1062 and repairs will be accessible from the commons, and need not fear
    1063 either the disappearance of their supplier or that someone will use a
    1064 particularly attractive improvement or a desperately necessary repair
    1065 as leverage for "taking the program private."</para>
    1066 
    1067 <para>This use of intellectual property rules to create a commons in
    1068 cyberspace is the central institutional structure enabling the
    1069 anarchist triumph. Ensuring free access and enabling modification at
    1070 each stage in the process means that the evolution of software occurs
    1071 in the fast Lamarckian mode: each favorable acquired characteristic of
    1072 others' work can be directly inherited. Hence the speed with which the
    1073 Linux kernel, for example, outgrew all of its proprietary
    1074 predecessors. Because defection is impossible, free riders are
    1075 welcome, which resolves one of the central puzzles of collective
    1076 action in a propertarian social system.</para>
    1077 
    1078 <para>Non-propertarian production is also directly responsible for the
    1079 famous stability and reliability of free software, which arises from
    1080 what Eric Raymond calls "Linus' law": With enough eyeballs, all bugs
    1081 are shallow. In practical terms, access to source code means that if I
    1082 have a problem I can fix it. Because I can fix it, I almost never have
    1083 to, because someone else has almost always seen it and fixed it
    1084 first.</para>
    1085 
    1086 <para>For the free software community, commitment to anarchist
    1087 production may be a moral imperative; as Richard Stallman wrote, it's
    1088 about freedom, not about price. Or it may be a matter of utility,
    1089 seeking to produce better software than propertarian modes of work
    1090 will allow.  From the droid point of view, the copyleft represents the
    1091 perversion of theory, but better than any other proposal over the past
    1092 decades it resolves the problems of applying copyright to the
    1093 inextricably merged functional and expressive features of computer
    1094 programs. That it produces better software than the alternative does
    1095 not imply that traditional copyright principles should now be
    1096 prohibited to those who want to own and market inferior software
    1097 products, or (more charitably) whose products are too narrow in appeal
    1098 for communal production. But our story should serve as a warning to
    1099 droids: The world of the future will bear little relation to the world
    1100 of the past. The rules are now being bent in two directions. The
    1101 corporate owners of "cultural icons" and other assets who seek
    1102 ever-longer terms for corporate authors, converting the "limited Time"
    1103 of Article I, §8 into a freehold have naturally been whistling music
    1104 to the android ear <footnote> <para>26. The looming expiration of
    1105 Mickey Mouse's ownership by Disney requires, from the point of view of
    1106 that wealthy "campaign contributor," for example, an alteration of the
    1107 general copyright law of the United States. See "Not Making it Any
    1108 More?  Vaporizing the Public Domain," in <emphasis>The Invisible
    1109 Barbecue</emphasis>, forthcoming.</para> </footnote> .  After all, who bought
    1110 the droids their concert tickets? But as the propertarian position
    1111 seeks to embed itself ever more strongly, in a conception of copyright
    1112 liberated from the minor annoyances of limited terms and fair use, at
    1113 the very center of our "cultural software" system, the anarchist
    1114 counter-strike has begun. Worse is yet to befall the droids, as we
    1115 shall see. But first, we must pay our final devoirs to the
    1116 dwarves.</para>
    1117 
    1118 </section>
    1119 <section>
    1120 <title>Because It's There: Faraday's Magnet and Human Creativity</title>
    1121 
    1122 <para>After all, they deserve an answer. Why do people make free
    1123 software if they don't get to profit? Two answers have usually been
    1124 given. One is half-right and the other is wrong, but both are
    1125 insufficiently simple.</para>
    1126 
    1127 <para>The wrong answer is embedded in numerous references to "the
    1128 hacker gift-exchange culture." This use of ethnographic jargon
    1129 wandered into the field some years ago and became rapidly, if
    1130 misleadingly, ubiquitous. It reminds us only that the
    1131 economeretricians have so corrupted our thought processes that any
    1132 form of non-market economic behavior seems equal to every other
    1133 kind. But gift-exchange, like market barter, is a propertarian
    1134 institution. Reciprocity is central to these symbolic enactments of
    1135 mutual dependence, and if either the yams or the fish are
    1136 short-weighted, trouble results. Free software, at the risk of
    1137 repetition, is a commons: no reciprocity ritual is enacted there. A
    1138 few people give away code that others sell, use, change, or borrow
    1139 wholesale to lift out parts for something else. Notwithstanding the
    1140 very large number of people (tens of thousands, at most) who have
    1141 contributed to GNU/Linux, this is orders of magnitude less than the
    1142 number of users who make no contribution whatever <footnote>
    1143 <para>27. A recent industry estimate puts the number of Linux systems
    1144 worldwide at 7.5 million. <emphasis>See</emphasis> Josh McHugh, 1998. <ulink
    1145 url="http://www.forbes.com/forbes/98/0810/6203094s1.htm">"Linux: The
    1146 Making of a Global Hack,"</ulink> <emphasis>Forbes</emphasis> (August 10). Because the
    1147 software is freely obtainable throughout the Net, there is no simple
    1148 way to assess actual usage.</para> </footnote>.</para>
    1149 
    1150 <para>A part of the right answer is suggested by the claim that free
    1151 software is made by those who seek reputational compensation for their
    1152 activity. Famous Linux hackers, the theory is, are known all over the
    1153 planet as programming deities. From this they derive either enhanced
    1154 self-esteem or indirect material advancement <footnote> <para>28. Eric
    1155 Raymond is a partisan of the "ego boost" theory, to which he adds
    1156 another faux-ethnographic comparison, of free software composition to
    1157 the Kwakiutl potlatch. <emphasis>See</emphasis> Eric S. Raymond, 1998. <ulink
    1158 url="http://firstmonday.org/issues/issue3_10/raymond/index.html">Homesteading
    1159 the Noosphere.</ulink>.  But the potlatch, certainly a form of status
    1160 competition, is unlike free software for two fundamental reasons: it
    1161 is essentially hierarchical, which free software is not, and, as we
    1162 have known since Thorstein Veblen first called attention to its
    1163 significance, it is a form of conspicuous waste. <emphasis>See</emphasis> Thorstein
    1164 Veblen, 1967. <emphasis>The Theory of the Leisure Class.</emphasis> New York:
    1165 Viking, p. 75. These are precisely the grounds which distinguish the
    1166 anti-hierarchical and utilitiarian free software culture from its
    1167 propertarian counterparts.</para></footnote>.  But the programming
    1168 deities, much as they have contributed to free software, have not done
    1169 the bulk of the work. Reputations, as Linus Torvalds himself has often
    1170 pointed out, are made by willingly acknowledging that it was all done
    1171 by someone else. And, as many observers have noted, the free software
    1172 movement has also produced superlative
    1173 documentation. Documentation-writing is not what hackers do to attain
    1174 cool, and much of the documentation has been written by people who
    1175 didn't write the code. Nor must we limit the indirect material
    1176 advantages of authorship to increases in reputational capital.  Most
    1177 free software authors I know have day jobs in the technology
    1178 industries, and the skills they hone in the more creative work they do
    1179 outside the market no doubt sometimes measurably enhance their value
    1180 within it. And as the free software products gained critical mass and
    1181 became the basis of a whole new set of business models built around
    1182 commercial distribution of that which people can also get for nothing,
    1183 an increasing number of people are specifically employed to write free
    1184 software. But in order to be employable in the field, they must
    1185 already have established themselves there. Plainly, then, this motive
    1186 is present, but it isn't the whole explanation.</para>
    1187 
    1188 <para>Indeed, the rest of the answer is just too simple to have
    1189 received its due. The best way to understand is to follow the brief
    1190 and otherwise unsung career of an initially-grudging free software
    1191 author.  Microsoft's Vinod Vallopillil, in the course of writing the
    1192 competitive analysis of Linux that was leaked as the second of the
    1193 famous "Halloween memoranda," bought and installed a Linux system on
    1194 one of his office computers. He had trouble because the (commercial)
    1195 Linux distribution he installed did not contain a daemon to handle the
    1196 DHCP protocol for assignment of dynamic IP addresses. The result was
    1197 important enough for us to risk another prolonged exposure to the
    1198 Microsoft Writing Style:</para>
    1199 
    1200 <blockquote><para>A small number of Web sites and FAQs later, I found an FTP
    1201 site with a Linux DHCP client. The DHCP client was developed by an
    1202 engineer employed by Fore Systems (as evidenced by his e-mail address;
    1203 I believe, however, that it was developed in his own free time). A
    1204 second set of documentation/manuals was written for the DHCP client by
    1205 a hacker in <emphasis>Hungary</emphasis> which provided relatively simple
    1206 instructions on how to install/load the client.</para>
    1207 
    1208 <para>I downloaded &amp; uncompressed the client and typed two
    1209 simple commands:</para>
    1210 
    1211 <para>Make - compiles the client binaries</para>
    1212 
    1213 <para>Make Install -installed the binaries as a Linux Daemon</para>
    1214 
    1215 <para>Typing "DHCPCD" (for DHCP Client Daemon) on the command
    1216 line triggered the DHCP discovery process and voila, I had IP
    1217 networking running.  </para>
    1218 
    1219 <para>Since I had just downloaded the DHCP client code, on an
    1220 impulse I played around a bit. Although the client wasn't as
    1221 extensible as the DHCP client we are shipping in NT5 (for example, it
    1222 won't query for arbitrary options &amp; store results), it was obvious
    1223 how I could write the additional code to implement this functionality.
    1224 The full client consisted of about 2,600 lines of code.</para>
    1225 
    1226 <para>One example of esoteric, extended functionality that was
    1227 clearly patched in by a third party was a set of routines to that
    1228 would pad the DHCP request with host-specific strings required by
    1229 Cable Modem / ADSL sites.</para>
    1230 
    1231 <para>A few other steps were required to configure the DHCP
    1232 client to auto-start and auto-configure my Ethernet interface on boot
    1233 but these were documented in the client code and in the DHCP
    1234 documentation from the Hungarian developer.</para>
    1235 
    1236 <para>I'm a poorly skilled UNIX programmer but it was
    1237 immediately obvious to me how to incrementally extend the DHCP client
    1238 code (the feeling was exhilarating and addictive).</para>
    1239 
    1240 <para>Additionally, due directly to GPL + having the full development
    1241 environment in front of me, I was in a position where I could write up
    1242 my changes and e-mail them out within a couple of hours (in contrast
    1243 to how things like this would get done in NT). Engaging in that
    1244 process would have prepared me for a larger, more ambitious Linux
    1245 project in the future <footnote><para>29. Vinod Vallopillil, <ulink
    1246 url="http://www.opensource.org/halloween/halloween2.html">Linux OS
    1247 Competitive Analysis (Halloween II).</ulink> Note Vallopillil's
    1248 surprise that a program written in California had been subsequently
    1249 documented by a programmer in Hungary.</para>
    1250 </footnote>.</para></blockquote>
    1251 
    1252 <para>"The feeling was exhilarating and addictive." Stop the presses:
    1253 Microsoft experimentally verifies Moglen's Metaphorical Corollary to
    1254 Faraday's Law. Wrap the Internet around every brain on the planet and
    1255 spin the planet. Software flows in the wires. It's an emergent
    1256 property of human minds to create. "Due directly to the GPL," as
    1257 Vallopillil rightly pointed out, free software made available to him
    1258 an exhilarating increase in his own creativity, of a kind not
    1259 achievable in his day job working for the Greatest Programming Company
    1260 on Earth. If only he had e-mailed that first addictive fix, who knows
    1261 where he'd be now?</para>
    1262 
    1263 <para>So, in the end, my dwarvish friends, it's just a human thing.
    1264 Rather like why Figaro sings, why Mozart wrote the music for him to
    1265 sing to, and why we all make up new words: Because we can. Homo
    1266 ludens, meet Homo faber. The social condition of global
    1267 interconnection that we call the Internet makes it possible for all of
    1268 us to be creative in new and previously undreamed-of ways. Unless we
    1269 allow "ownership" to interfere. Repeat after me, ye dwarves and men:
    1270 Resist the resistance!</para>
    1271 
    1272 </section>
    1273 <!--<para><img src="anarchism_files/quad.gif"></para><a name="m4"></a>-->
    1274 
    1275 <section>
    1276 <title>IV. Their Lordships Die in the Dark?</title>
    1277 
    1278 <para>For the IPdroid, fresh off the plane from a week at Bellagio
    1279 paid for by Dreamworks SKG, it's enough to cause indigestion.</para>
    1280 
    1281 <para>Unlock the possibilities of human creativity by connecting
    1282 everyone to everyone else? Get the ownership system out of the way so
    1283 that we can all add our voices to the choir, even if that means
    1284 pasting our singing on top of the Mormon Tabernacle and sending the
    1285 output to a friend? No one sitting slack-jawed in front of a televised
    1286 mixture of violence and imminent copulation carefully devised to
    1287 heighten the young male eyeball's interest in a beer commercial? What
    1288 will become of civilization? Or at least of copyright teachers?</para>
    1289 
    1290 <para>But perhaps this is premature. I've only been talking about
    1291 software. Real software, the old kind, that runs computers. Not like
    1292 the software that runs DVD players, or the kind made by the Grateful
    1293 Dead. "Oh yes, the Grateful Dead. Something strange about them, wasn't
    1294 there? Didn't prohibit recording at their concerts. Didn't mind if
    1295 their fans rather riled the recording industry. Seem to have done all
    1296 right, though, you gotta admit. Senator Patrick Leahy, isn't he a
    1297 former Deadhead? I wonder if he'll vote to extend corporate authorship
    1298 terms to 125 years, so that Disney doesn't lose The Mouse in 2004. And
    1299 those DVD players - they're computers, aren't they?"</para>
    1300 
    1301 <para>In the digital society, it's all connected. We can't depend for
    1302 the long run on distinguishing one bitstream from another in order to
    1303 figure out which rules apply. What happened to software is already
    1304 happening to music. Their recording industry lordships are now
    1305 scrambling wildly to retain control over distribution, as both
    1306 musicians and listeners realize that the middlepeople are no longer
    1307 necessary. The Great Potemkin Village of 1999, the so-called Secure
    1308 Digital Music Initiative, will have collapsed long before the first
    1309 Internet President gets inaugurated, for simple technical reasons as
    1310 obvious to those who know as the ones that dictated the triumph of
    1311 free software <footnote> <para>30. See "They're Playing Our Song: The
    1312 Day the Music Industry Died," in <emphasis>The Invisible Barbecue</emphasis>,
    1313 forthcoming.</para> </footnote> . The anarchist revolution in music is
    1314 different from the one in software <emphasis>tout court</emphasis>, but here too -
    1315 as any teenager with an MP3 collection of self-released music from
    1316 unsigned artists can tell you - theory has been killed off by the
    1317 facts. Whether you are Mick Jagger, or a great national artist from
    1318 the third world looking for a global audience, or a garret-dweller
    1319 reinventing music, the recording industry will soon have nothing to
    1320 offer you that you can't get better for free.  And music doesn't sound
    1321 worse when distributed for free, pay what you want directly to the
    1322 artist, and don't pay anything if you don't want to. Give it to your
    1323 friends; they might like it.</para>
    1324 
    1325 <para>
    1326 What happened to music is also happening to news. The wire services,
    1327 as any U.S. law student learns even before taking the near-obligatory
    1328 course in Copyright for Droids, have a protectible property interest
    1329 in their expression of the news, even if not in the facts the news
    1330 reports <footnote><para>31. International News Service v. Associated
    1331 Press, 248 U.S. 215 (1918). With regard to the actual terse, purely
    1332 functional expressions of breaking news actually at stake in the
    1333 jostling among wire services, this was always a distinction only a
    1334 droid could love.</para></footnote>.  So why are they now giving all
    1335 their output away? Because in the world of the Net, most news is
    1336 commodity news. And the original advantage of the news gatherers, that
    1337 they were internally connected in ways others were not when
    1338 communications were expensive, is gone. Now what matters is collecting
    1339 eyeballs to deliver to advertisers. It isn't the wire services that
    1340 have the advantage in covering Kosovo, that's for sure. Much less
    1341 those paragons of "intellectual" property, their television
    1342 lordships. They, with their overpaid pretty people and their massive
    1343 technical infrastructure, are about the only organizations in the
    1344 world that can't afford to be everywhere all the time. And then they
    1345 have to limit themselves to ninety seconds a story, or the eyeball
    1346 hunters will go somewhere else. So who makes better news, the
    1347 propertarians or the anarchists?  We shall soon see.</para>
    1348 
    1349 <para>Oscar Wilde says somewhere that the problem with socialism is
    1350 that it takes up too many evenings. The problems with anarchism as a
    1351 social system are also about transaction costs. But the digital
    1352 revolution alters two aspects of political economy that have been
    1353 otherwise invariant throughout human history. All software has zero
    1354 marginal cost in the world of the Net, while the costs of social
    1355 coordination have been so far reduced as to permit the rapid formation
    1356 and dissolution of large-scale and highly diverse social groupings
    1357 entirely without geographic limitation <footnote> <para>32. See "No
    1358 Prodigal Son: The Political Theory of Universal Interconnection," in
    1359 <emphasis>The Invisible Barbecue</emphasis>, forthcoming.</para> </footnote> . Such
    1360 fundamental change in the material circumstances of life necessarily
    1361 produces equally fundamental changes in culture. Think not? Tell it to
    1362 the Iroquois. And of course such profound shifts in culture are
    1363 threats to existing power relations. Think not? Ask the Chinese
    1364 Communist Party.  Or wait 25 years and see if you can find them for
    1365 purposes of making the inquiry.</para>
    1366 
    1367 <para>In this context, the obsolescence of the IPdroid is neither
    1368 unforseeable nor tragic. Indeed it may find itself clanking off into
    1369 the desert, still lucidly explaining to an imaginary room the
    1370 profitably complicated rules for a world that no longer exists. But at
    1371 least it will have familiar company, recognizable from all those
    1372 glittering parties in Davos, Hollywood, and Brussels. Our Media Lords
    1373 are now at handigrips with fate, however much they may feel that the
    1374 Force is with them. The rules about bitstreams are now of dubious
    1375 utility for maintaining power by co-opting human creativity. Seen
    1376 clearly in the light of fact, these Emperors have even fewer clothes
    1377 than the models they use to grab our eyeballs. Unless supported by
    1378 user-disabling technology, a culture of pervasive surveillance that
    1379 permits every reader of every "property" to be logged and charged, and
    1380 a smokescreen of droid-breath assuring each and every young person
    1381 that human creativity would vanish without the benevolent aristocracy
    1382 of BillG the Creator, Lord Murdoch of Everywhere, the Spielmeister and
    1383 the Lord High Mouse, their reign is nearly done. But what's at stake
    1384 is the control of the scarcest resource of all: our
    1385 attention. Conscripting that makes all the money in the world in the
    1386 digital economy, and the current lords of the earth will fight for
    1387 it. Leagued against them are only the anarchists: nobodies, hippies,
    1388 hobbyists, lovers, and artists. The resulting unequal contest is the
    1389 great political and legal issue of our time.  Aristocracy looks hard
    1390 to beat, but that's how it looked in 1788 and 1913 too. It is, as Chou
    1391 En-Lai said about the meaning of the French Revolution, too soon to
    1392 tell.</para>
    1393 
    1394 </section>
    1395 <section>
    1396 <title>About the Author</title>
    1397 
    1398 <para>Eben Moglen is Professor of Law &amp; Legal History, Columbia Law School.
    1399 E-mail: <ulink url="mailto:moglen@columbia.edu">Mail: moglen@columbia.edu</ulink></para>
    1400 
    1401 <para>Acknowledgments</para>
    1402 
    1403 <para>This paper was prepared for delivery at the Buchmann
    1404 International Conference on Law, Technology and Information, at Tel
    1405 Aviv University, May 1999; my thanks to the organizers for their kind
    1406 invitation. I owe much as always to Pamela Karlan for her insight and
    1407 encouragement. I especially wish to thank the programmers throughout
    1408 the world who made free software possible.</para>
    1409 
    1410 
    1411 <blockquote>
    1412 <para>
    1413 <ulink url="http://firstmonday.org/issues/issue4_8/index.html"><!--<img src="anarchism_files/contents.gif" alt="Contents" align="bottom" border="0">--></ulink> </para>
    1414 <para>
    1415 <ulink url="http://firstmonday.org/issues/index.html"><!--<img src="anarchism_files/index.gif" alt="Index" border="0">--></ulink>
    1416 </para>
    1417 <para>Copyright <ulink url="http://firstmonday.org/copy.html">©</ulink> 1999, First Monday</para></blockquote>
    1418 
    1419 
    1420 </section>
     39  -->
     40
     41
     42  <para><ulink url="http://firstmonday.org/issues/issue4_8/moglen/index.html#author"><!-- <img src="anarchism_files/moglen.gif" alt="Anarchism Triumphant: Free Software and the Death of Copyright" border="0">--> </ulink></para>
     43
     44  <blockquote><para>Разпространението на ядрото за операционни системи
     45  Линукс насочи вниманието към движението за свободен софтуер.  Това есе
     46  показва защо свободният софтуер, който далеч не е нищожен участник в
     47  пазара на комерсиален софтуер, е важната първа стъпка в премахването
     48  на системата на интелектуална собственост.</para></blockquote>
     49
     50  <section>
     51    <title>Софтуерът като собственост: Теоретичният парадокс</title>
     52
     53    <para><emphasis>Софтуер</emphasis>: никоя друга дума не въплъщава
     54    толкова пълно рактическите и социалните ефекти на цифровата революция.
     55    Първоначално терминът е бил чисто технически и е означавал частите на
     56    една компютърна система, която за разлика от "хардуера" -- направен
     57    непроменим от производителя си в електрониката на системата, е можел
     58    свободно да бъде променян.  Първият софтуер е представлявал начина на
     59    включване на кабели и прекъсвачи на външните панели на електронни
     60    устройства, но още с появата на езикови средства за промяната на
     61    поведението на компютъра, "софтуер" започнал да обозначава предимно
     62    изразяванията в повече или по-малко понятех за хората език, който
     63    както описвал, така и контролирал поведението на машината<footnote>
     64    <para>1. Тази отлика е била само приблизителна в първоначалния
     65    контекст.  В края на 60-те определена част от основните операции на
     66    хардуера са контролирани от програми, които са цифрово кодирани в
     67    електрониката на компютърното оборудване, които не могат да бъдат
     68    променяни веднъж след като продукцията е излязла от фабриката.  Такива
     69    символни, но непроменими компоненти, са били известни като "микрокод"
     70    на жаргона на индустрията, но стана обичайно те да се наричат
     71    "фърмуеър".  Изменчивостта, както бе показано от термина
     72    "фърмуеър"<!-- БЕЛЕЖКА ЗА ЗНАЧЕНИЕТО НА КОРЕНИТЕ НА ДУМИТЕ СОФТУЕР,
     73    ХАРДУЕР, ФЪРМУЕР -->,се отнася главно към възможността на
     74    потребителите да изменят символите, които определят поведението на
     75    машината.  Понеже цифровата революция доведе до широката употреба на
     76    компютрите от технически некомпетентни лица, повечето от традиционния
     77    софтуер -- приложни програми, операционни системи, инструкции за
     78    числово управление и т. н. -- е, за повечето от потребителите си,
     79    фърмуер.  Може да е символен, а не електронен в начина, по който е
     80    направен, но те не могат да го променят, дори и да искат, нещо което
     81    те често, но безсилно и с негодуванние правят.  Това "затвърдяване на
     82    софтуера" е основното условие на собственическия подход към законовата
     83    организация на цифровото обществео, което е темата на този
     84    доклад.</para></footnote>.</para>
     85
     86    <para>Така е било тогава, а сега е така: технологиите базирани на
     87    обработката на информация кодирана в цифров вид сега е социално
     88    доминираща в повечето аспекти на човешката култура в "развитите"
     89    общества.  <footnote><para>2. В рамките на сегашното поколение,
     90    самата концепция за социално "равитие" се измества от притежанието
     91    на индустрия основана на двигател с вътрешно горене към
     92    "пост-индустрия" базирана на цифровите комуникации и свързаните с
     93    тях форми на икономическа дейност, основани на
     94    "знания".</para></footnote>.  Преминаването от аналогово към
     95    цифрово представяне -- във видеото, музиката, печатането,
     96    телекомуникациите и дори хореографията, религиозните култове и
     97    сексуалното задоволяване <!-- religious worship, sexual
     98    gratification --> -- потенциално превръща всички форми на
     99    човешката символна дейност във софтуер, то ест -- променими
     100    инструкции за описание и управление на поведението на машините.
     101    Чрез концептуално постформиране, характено за западното научно
     102    мислене, разделението между хардуера и софтуера се наблюдава в
     103    природния или социалния свят и е станал нов начин за изразяване на
     104    конфликта между идеите на детерминизъм и свободата на волята
     105    (действие?), природата и човека, или гените и културата.  <!--
     106    Какво е backformation? Аналог на transformation ли? Nature <->
     107    Nurture, как е free will на български.  By a conceptual
     108    back-formation characteristic of Western scientistic thinking, the
     109    division between hardware and software is now being observed in
     110    the natural or social world, and has become a new way to express
     111    the conflict between ideas of determinism and free will, nature
     112    and nurture, or genes and culture.  --> Нашият "хардуер", който е
     113    генетично зададен е нашата природа и ни определя.  Нашето
     114    възпитание е "софтуера", който задава културното ни прграмиране,
     115    което е нашата относителна свобода.  И така нататък, за неразумно
     116    дърдорещите. <!-- And so on, for those reckless of blather
     117    -->.<footnote><para>3. Всъщност, едно бързо замисляне ще разкрие,
     118    че нашите гени са фърмуеър.  Еволюцията направи прехода от
     119    аналогово към цифрово още преди периода на първите вкаменелости.
     120    Но ние не притежавахме властта за управлявани, преки промени.  До
     121    завчера.  През следващото столетие гените също ще се превърнат в
     122    софтуер и въпреки че не разглеждам проблема по нататък в това есе,
     123    политиеските последствия на несвободността на софтуера в този
     124    контекст са още по-плашещи в сравнение с културните
     125    артефакти.</para></footnote> Този "софтуер" се превръща в
     126    жизнеспособна метафора за цялата символна активност, която
     127    очевидно е разведена (еманципирана) от техническия контекст на
     128    произхода на думата, въпреки неудобството, което се появява в
     129    технически компетентните, когато термина влиза в устите на хората,
     130    като се изпуска концептуалното значение на неговия
     131    произход.<footnote><para>4. <emphasis>Виж напр.:</emphasis>
     132    J. M. Balkin, 1998. <emphasis>Cultural Software: a Theory of
     133    Ideology.</emphasis> New Haven: Yale University
     134    Press.</para></footnote></para>
     135
     136
     137    <para>But the widespread adoption of digital technology for use by
     138    those who do not understand the principles of its operation, while it
     139    apparently licenses the broad metaphoric employment of "software,"
     140    does not in fact permit us to ignore the computers that are now
     141    everywhere underneath our social skin. The movement from analog to
     142    digital is more important for the structure of social and legal
     143    relations than the more famous if less certain movement from status to
     144    contract <footnote><para>5. <emphasis>See</emphasis> Henry Sumner
     145    Maine, 1861. <emphasis>Ancient Law: Its Connection with the Early
     146    History of Society, and Its Relation to Modern Idea.</emphasis> First
     147    edition. London: J. Murray.</para></footnote>.  This is bad news for
     148    those legal thinkers who do not understand it, which is why so much
     149    pretending to understand now goes so floridly on.  Potentially,
     150    however, our great transition is very good news for those who can turn
     151    this new-found land into property for themselves. Which is why the
     152    current "owners" of software so strongly support and encourage the
     153    ignorance of everyone else. Unfortunately for them - for reasons
     154    familiar to legal theorists who haven't yet understood how to apply
     155    their traditional logic in this area - the trick won't work. This
     156    paper explains why<footnote><para>6. In general I dislike the
     157    intrusion of autobiography into scholarship. But because it is here my
     158    sad duty and great pleasure to challenge the qualifications or
     159    <emphasis>bona fides</emphasis> of just about everyone, I must enable
     160    the assessment of my own. I was first exposed to the craft of computer
     161    programming in 1971. I began earning wages as a commercial programmer
     162    in 1973 - at the age of thirteen - and did so, in a variety of
     163    computer services, engineering, and multinational technology
     164    enterprises, until 1985. In 1975 I helped write one of the first
     165    networked e-mail systems in the United States; from 1979 I was engaged
     166    in research and development of advanced computer programming languages
     167    at IBM. These activities made it economically possible for me to study
     168    the arts of historical scholarship and legal cunning. My wages were
     169    sufficient to pay my tuitions, but not - to anticipate an argument
     170    that will be made by the econodwarves further along - because my
     171    programs were the intellectual property of my employer, but rather
     172    because they made the hardware my employer sold work better. Most of
     173    what I wrote was effectively free software, as we shall see. Although
     174    I subsequently made some inconsiderable technical contributions to the
     175    actual free software movement this paper describes, my primary
     176    activities on its behalf have been legal: I have served for the past
     177    five years (without pay, naturally) as general counsel of the Free
     178    Software Foundation.</para></footnote>.</para>
     179
     180    <para>We need to begin by considering the technical essence of the
     181    familiar devices that surround us in the era of "cultural software." A
     182    CD player is a good example. Its primary input is a bitstream read
     183    from an optical storage disk. The bitstream describes music in terms
     184    of measurements, taken 44,000 times per second, of frequency and
     185    amplitude in each of two audio channels. The player's primary output
     186    is analog audio signals <footnote><para>7. The player, of course, has
     187    secondary inputs and outputs in control channels: buttons or infrared
     188    remote control are input, and time and track display are
     189    output.</para></footnote>. Like everything else in the digital world,
     190    music as seen by a CD player is mere numeric information; a particular
     191    recording of Beethoven's Ninth Symphony recorded by Arturo Toscanini
     192    and the NBC Symphony Orchestra and Chorale is (to drop a few
     193    insignificant digits) 1276749873424, while Glenn Gould's peculiarly
     194    perverse last recording of the Goldberg Variations is (similarly
     195    rather truncated) 767459083268.</para>
     196
     197    <para>Oddly enough, these two numbers are "copyrighted." This means,
     198    supposedly, that you can't possess another copy of these numbers, once
     199    fixed in any physical form, unless you have licensed them. And you
     200    can't turn 767459083268 into 2347895697 for your friends (thus
     201    correcting Gould's ridiculous judgment about tempi) without making a
     202    "derivative work," for which a license is necessary.</para>
     203
     204    <para>At the same time, a similar optical storage disk contains
     205    another number, let us call it 7537489532. This one is an algorithm
     206    for linear programming of large systems with multiple constraints,
     207    useful for example if you want to make optimal use of your rolling
     208    stock in running a freight railroad. This number (in the U.S.) is
     209    "patented," which means you cannot derive 7537489532 for yourself, or
     210    otherwise "practice the art" of the patent with respect to solving
     211    linear programming problems no matter how you came by the idea,
     212    including finding it out for yourself, unless you have a license from
     213    the number's owner.</para>
     214
     215    <para>Then there's 9892454959483. This one is the source code for
     216    Microsoft Word. In addition to being "copyrighted," this one is a
     217    trade secret.  That means if you take this number from Microsoft and
     218    give it to anyone else you can be punished.</para>
     219
     220    <para>Lastly, there's 588832161316. It doesn't do anything, it's just
     221    the square of 767354. As far as I know, it isn't owned by anybody
     222    under any of these rubrics. Yet.</para>
     223
     224    <para>At this point we must deal with our first objection from the
     225    learned. It comes from a creature known as the IPdroid. The droid has
     226    a sophisticated mind and a cultured life. It appreciates very much the
     227    elegant dinners at academic and ministerial conferences about the
     228    TRIPs, not to mention the privilege of frequent appearances on MSNBC.
     229    It wants you to know that I'm committing the mistake of confusing the
     230    embodiment with the intellectual property itself. It's not the number
     231    that's patented, stupid, just the Kamarkar algorithm. The number
     232    <emphasis>can</emphasis> be copyrighted, because copyright covers the
     233    expressive qualities of a particular tangible embodiment of an idea
     234    (in which some functional properties may be mysteriously merged,
     235    provided that they're not too merged), but not the algorithm. Whereas
     236    the number isn't patentable, just the "teaching" of the number with
     237    respect to making railroads run on time. And the number representing
     238    the source code of Microsoft Word can be a trade secret, but if you
     239    find it out for yourself (by performing arithmetic manipulation of
     240    other numbers issued by Microsoft, for example, which is known as
     241    "reverse engineering"), you're not going to be punished, at least if
     242    you live in some parts of the United States.</para>
     243
     244    <para>This droid, like other droids, is often right. The condition of
     245    being a droid is to know everything about something and nothing about
     246    anything else. By its timely and urgent intervention the droid has
     247    established that the current intellectual property system contains
     248    many intricate and ingenious features. The complexities combine to
     249    allow professors to be erudite, Congressmen to get campaign
     250    contributions, lawyers to wear nice suits and tassel loafers, and
     251    Murdoch to be rich.  The complexities mostly evolved in an age of
     252    industrial information distribution, when information was inscribed in
     253    analog forms on physical objects that cost something significant to
     254    make, move, and sell. When applied to digital information that moves
     255    frictionlessly through the network and has zero marginal cost per
     256    copy, everything still works, mostly, as long as you don't stop
     257    squinting.</para>
     258
     259    <para>But that wasn't what I was arguing about. I wanted to point out
     260    something else: that our world consists increasingly of nothing but
     261    large numbers (also known as bitstreams), and that - for reasons
     262    having nothing to do with emergent properties of the numbers
     263    themselves - the legal system is presently committed to treating
     264    similar numbers radically differently. No one can tell, simply by
     265    looking at a number that is 100 million digits long, whether that
     266    number is subject to patent, copyright, or trade secret protection, or
     267    indeed whether it is "owned" by anyone at all. So the legal system we
     268    have - blessed as we are by its consequences if we are copyright
     269    teachers, Congressmen, Gucci-gulchers or Big Rupert himself - is
     270    compelled to treat indistinguishable things in unlike ways.</para>
     271
     272    <para>Now, in my role as a legal historian concerned with the secular
     273    (that is, very long term) development of legal thought, I claim that
     274    legal regimes based on sharp but unpredictable distinctions among
     275    similar objects are radically unstable. They fall apart over time
     276    because every instance of the rules' application is an invitation to
     277    at least one side to claim that instead of fitting in ideal category A
     278    the particular object in dispute should be deemed to fit instead in
     279    category B, where the rules will be more favorable to the party making
     280    the claim. This game - about whether a typewriter should be deemed a
     281    musical instrument for purposes of railway rate regulation, or whether
     282    a steam shovel is a motor vehicle - is the frequent stuff of legal
     283    ingenuity. But when the conventionally-approved legal categories
     284    require judges to distinguish among the identical, the game is
     285    infinitely lengthy, infinitely costly, and almost infinitely offensive
     286    to the unbiased bystander <footnote><para>8. This is not an insight
     287    unique to our present enterprise. A closely-related idea forms one of
     288    the most important principles in the history of Anglo-American law,
     289    perfectly put by Toby Milsom in the following terms:</para>
     290    <blockquote><para>The life of the common law has been in the abuse of
     291    its elementary ideas. If the rules of property give what now seems an
     292    unjust answer, try obligation; and equity has proved that from the
     293    materials of obligation you can counterfeit the phenomena of
     294    property. If the rules of contract give what now seems an unjust
     295    answer, try tort. ... If the rules of one tort, say deceit, give what
     296    now seems an unjust answer, try another, try negligence. And so the
     297    legal world goes round.</para></blockquote><para>S.F.C. Milsom,
     298    1981. <emphasis>Historical Foundations of the Common Law.</emphasis>
     299    Second edition. London: Butterworths, p. 6.</para> </footnote>.</para>
     300
     301    <para>Thus parties can spend all the money they want on all the
     302    legislators and judges they can afford - which for the new "owners" of
     303    the digital world is quite a few - but the rules they buy aren't going
     304    to work in the end. Sooner or later, the paradigms are going to
     305    collapse. Of course, if later means two generations from now, the
     306    distribution of wealth and power sanctified in the meantime may not be
     307    reversible by any course less drastic than a <emphasis>bellum
     308    servile</emphasis> of couch potatoes against media magnates. So
     309    knowing that history isn't on Bill Gates' side isn't enough. We are
     310    predicting the future in a very limited sense: we know that the
     311    existing rules, which have yet the fervor of conventional belief
     312    solidly enlisted behind them, are no longer meaningful. Parties will
     313    use and abuse them freely until the mainstream of "respectable"
     314    conservative opinion acknowledges their death, with uncertain
     315    results. But realistic scholarship should already be turning its
     316    attention to the clear need for new thoughtways.</para>
     317
     318    <para>When we reach this point in the argument, we find ourselves
     319    contending with the other primary protagonist of educated idiocy: the
     320    econodwarf. Like the IPdroid, the econodwarf is a species of hedgehog,
     321    <footnote><para>9. <emphasis>See</emphasis> Isaiah Berlin,
     322    1953. <emphasis>The Hedgehog and the Fox: An Essay on Tolstoy's View
     323    of History.</emphasis> New York: Simon and Schuster.</para>
     324    </footnote> but where the droid is committed to logic over experience,
     325    the econodwarf specializes in an energetic and well-focused but
     326    entirely erroneous view of human nature. According to the econodwarf's
     327    vision, each human being is an individual possessing "incentives,"
     328    which can be retrospectively unearthed by imagining the state of the
     329    bank account at various times.  So in this instance the econodwarf
     330    feels compelled to object that without the rules I am lampooning,
     331    there would be no incentive to create the things the rules treat as
     332    property: without the ability to exclude others from music there would
     333    be no music, because no one could be sure of getting paid for creating
     334    it.</para>
     335
     336    <para>Music is not really our subject; the software I am considering
     337    at the moment is the old kind: computer programs. But as he is
     338    determined to deal at least cursorily with the subject, and because,
     339    as we have seen, it is no longer really possible to distinguish
     340    computer programs from music performances, a word or two should be
     341    said. At least we can have the satisfaction of indulging in an
     342    argument <emphasis>ad pygmeam</emphasis>.  When the econodwarf grows
     343    rich, in my experience, he attends the opera.  But no matter how often
     344    he hears <emphasis>Don Giovanni</emphasis> it never occurs to him that
     345    Mozart's fate should, on his logic, have entirely discouraged
     346    Beethoven, or that we have <emphasis>The Magic Flute</emphasis> even
     347    though Mozart knew very well he wouldn't be paid. In fact,
     348    <emphasis>The Magic Flute</emphasis>, <emphasis>St. Matthew's
     349    Passion</emphasis>, and the motets of the wife-murderer Carlo Gesualdo
     350    are all part of the centuries-long tradition of free software, in the
     351    more general sense, which the econodwarf never quite
     352    acknowledges.</para> <!--<center><img
     353    src="anarchism_files/mog1.gif"></center> --> <para> The dwarf's basic
     354    problem is that "incentives" is merely a metaphor, and as a metaphor
     355    to describe human creative activity it's pretty crummy. I have said
     356    this before, <footnote> <para>10. <emphasis>See</emphasis> <ulink
     357    url="http://emoglen.law.columbia.edu/my_pubs/nospeech.html">The
     358    Virtual Scholar and Network Liberation.</ulink></para> </footnote> but
     359    the better metaphor arose on the day Michael Faraday first noticed
     360    what happened when he wrapped a coil of wire around a magnet and spun
     361    the magnet. Current flows in such a wire, but we don't ask what the
     362    incentive is for the electrons to leave home. We say that the current
     363    results from an emergent property of the system, which we call
     364    induction. The question we ask is "what's the resistance of the wire?"
     365    So Moglen's Metaphorical Corollary to Faraday's Law says that if you
     366    wrap the Internet around every person on the planet and spin the
     367    planet, software flows in the network. It's an emergent property of
     368    connected human minds that they create things for one another's
     369    pleasure and to conquer their uneasy sense of being too alone. The
     370    only question to ask is, what's the resistance of the network?
     371    Moglen's Metaphorical Corollary to Ohm's Law states that the
     372    resistance of the network is directly proportional to the field
     373    strength of the "intellectual property" system. So the right answer to
     374    the econodwarf is, resist the resistance.</para>
     375
     376    <para>Of course, this is all very well in theory. "Resist the
     377    resistance" sounds good, but we'd have a serious problem, theory
     378    notwithstanding, if the dwarf were right and we found ourselves
     379    under-producing good software because we didn't let people own it. But
     380    dwarves and droids are formalists of different kinds, and the
     381    advantage of realism is that if you start from the facts the facts are
     382    always on your side. It turns out that treating software as property
     383    makes bad software.</para>
     384
     385  </section>
     386  <section>
     387    <title>II. Software as Property: The Practical Problem</title>
     388
     389    <para>In order to understand why turning software into property
     390    produces bad software, we need an introduction to the history of the
     391    art. In fact, we'd better start with the word "art" itself. The
     392    programming of computers combines determinate reasoning with literary
     393    invention.</para>
     394
     395    <para>At first glance, to be sure, source code appears to be a
     396    non-literary form of composition <footnote><para>11. Some basic
     397    vocabulary is essential. Digital computers actually execute numerical
     398    instructions: bitstrings that contain information in the "native"
     399    language created by the machine's designers. This is usually referred
     400    to as "machine language." The machine languages of hardware are
     401    designed for speed of execution at the hardware level, and are not
     402    suitable for direct use by human beings. So among the central
     403    components of a computer system are "programming languages," which
     404    translate expressions convenient for humans into machine language. The
     405    most common and relevant, but by no means the only, form of computer
     406    language is a "compiler." The compiler performs static translation, so
     407    that a file containing human-readable instructions, known as "source
     408    code" results in the generation of one or more files of executable
     409    machine language, known as "object code."</para> </footnote>.  The
     410    primary desideratum in a computer program is that it works, that is to
     411    say, performs according to specifications formally describing its
     412    outputs in terms of its inputs. At this level of generality, the
     413    functional content of programs is all that can be seen.</para>
     414
     415    <para>But working computer programs exist as parts of computer
     416    systems, which are interacting collections of hardware, software, and
     417    human beings. The human components of a computer system include not
     418    only the users, but also the (potentially different) persons who
     419    maintain and improve the system. Source code not only communicates
     420    with the computer that executes the program, through the intermediary
     421    of the compiler that produces machine-language object code, but also
     422    with other programmers.</para>
     423
     424    <para>The function of source code in relation to other human beings is
     425    not widely grasped by non-programmers, who tend to think of computer
     426    programs as incomprehensible. They would be surprised to learn that
     427    the bulk of information contained in most programs is, from the point
     428    of view of the compiler or other language processor, "comment," that
     429    is, non-functional material. The comments, of course, are addressed to
     430    others who may need to fix a problem or to alter or enhance the
     431    program's operation. In most programming languages, far more space is
     432    spent in telling people what the program does than in telling the
     433    computer how to do it.</para>
     434
     435    <para>The design of programming languages has always proceeded under
     436    the dual requirements of complete specification for machine execution
     437    and informative description for human readers. One might identify
     438    three basic strategies in language design for approaching this dual
     439    purpose.  The first, pursued initially with respect to the design of
     440    languages specific to particular hardware products and collectively
     441    known as "assemblers," essentially separated the human- and
     442    machine-communication portions of the program. Assembler instructions
     443    are very close relatives of machine-language instructions: in general,
     444    one line of an assembler program corresponds to one instruction in the
     445    native language of the machine. The programmer controls machine
     446    operation at the most specific possible level, and (if
     447    well-disciplined) engages in running commentary alongside the machine
     448    instructions, pausing every few hundred instructions to create "block
     449    comments," which provide a summary of the strategy of the program, or
     450    document the major data structures the program manipulates.</para>
     451
     452    <para>A second approach, characteristically depicted by the language
     453    COBOL (which stood for "Common Business-Oriented Language"), was to
     454    make the program itself look like a set of natural language
     455    directions, written in a crabbed but theoretically human-readable
     456    style. A line of COBOL code might say, for example "MULTIPLY PRICE
     457    TIMES QUANTITY GIVING EXPANSION." At first, when the Pentagon and
     458    industry experts began the joint design of COBOL in the early 1960's,
     459    this seemed a promising approach. COBOL programs appeared largely
     460    self-documenting, allowing both the development of work teams able to
     461    collaborate on the creation of large programs, and the training of
     462    programmers who, while specialized workers, would not need to
     463    understand the machine as intimately as assembler programs had to. But
     464    the level of generality at which such programs documented themselves
     465    was wrongly selected. A more formulaic and compressed expression of
     466    operational detail "expansion = price x quantity," for example, was
     467    better suited even to business and financial applications where the
     468    readers and writers of programs were accustomed to mathematical
     469    expression, while the processes of describing both data structures and
     470    the larger operational context of the program were not rendered
     471    unnecessary by the wordiness of the language in which the details of
     472    execution were specified.</para>
     473
     474    <para>Accordingly, language designers by the late 1960s began
     475    experimenting with forms of expression in which the blending of
     476    operational details and non-functional information necessary for
     477    modification or repair was more subtle. Some designers chose the path
     478    of highly symbolic and compressed languages, in which the programmer
     479    manipulated data abstractly, so that "A x B" might mean the
     480    multiplication of two integers, two complex numbers, two vast arrays,
     481    or any other data type capable of some process called
     482    "multiplication," to be undertaken by the computer on the basis of the
     483    context for the variables "A" and "B" at the moment of execution
     484    <footnote> <para>12. This, I should say, was the path that most of my
     485    research and development followed, largely in connection with a
     486    language called APL ("A Programming Language") and its successors. It
     487    was not, however, the ultimately-dominant approach, for reasons that
     488    will be suggested below.</para> </footnote> .  Because this approach
     489    resulted in extremely concise programs, it was thought, the problem of
     490    making code comprehensible to those who would later seek to modify or
     491    repair it was simplified. By hiding the technical detail of computer
     492    operation and emphasizing the algorithm, languages could be devised
     493    that were better than English or other natural languages for the
     494    expression of stepwise processes. Commentary would be not only
     495    unnecessary but distracting, just as the metaphors used to convey
     496    mathematical concepts in English do more to confuse than to
     497    enlighten.</para>
     498
     499    <section>
     500      <title>How We Created the Microbrain Mess</title>
     501
     502      <para>Thus the history of programming languages directly reflected the
     503      need to find forms of human-machine communication that were also
     504      effective in conveying complex ideas to human readers. "Expressivity"
     505      became a property of programming languages, not because it facilitated
     506      computation, but because it facilitated the collaborative creation and
     507      maintenance of increasingly complex software systems.</para>
     508
     509      <para>At first impression, this seems to justify the application of
     510      traditional copyright thinking to the resulting works. Though
     511      substantially involving "functional" elements, computer programs
     512      contained "expressive" features of paramount importance. Copyright
     513      doctrine recognized the merger of function and expression as
     514      characteristic of many kinds of copyrighted works. "Source code,"
     515      containing both the machine instructions necessary for functional
     516      operation and the expressive "commentary" intended for human readers,
     517      was an appropriate candidate for copyright treatment.</para>
     518
     519      <para>True, so long as it is understood that the expressive component
     520      of software was present solely in order to facilitate the making of
     521      "derivative works." Were it not for the intention to facilitate
     522      alteration, the expressive elements of programs would be entirely
     523      supererogatory, and source code would be no more copyrightable than
     524      object code, the output of the language processor, purged of all but
     525      the program's functional characteristics.</para>
     526
     527      <para>The state of the computer industry throughout the 1960's and
     528      1970's, when the grundnorms of sophisticated computer programming were
     529      established, concealed the tension implicit in this situation. In that
     530      period, hardware was expensive. Computers were increasingly large and
     531      complex collections of machines, and the business of designing and
     532      building such an array of machines for general use was dominated, not
     533      to say monopolized, by one firm. IBM gave away its software. To be
     534      sure, it owned the programs its employees wrote, and it copyrighted
     535      the source code. But it also distributed the programs - including the
     536      source code - to its customers at no additional charge, and encouraged
     537      them to make and share improvements or adaptations of the programs
     538      thus distributed. For a dominant hardware manufacturer, this strategy
     539      made sense: better programs sold more computers, which is where the
     540      profitability of the business rested.</para>
     541
     542      <para>Computers, in this period, tended to aggregate within particular
     543      organizations, but not to communicate broadly with one another. The
     544      software needed to operate was distributed not through a network, but
     545      on spools of magnetic tape. This distribution system tended to
     546      centralize software development, so that while IBM customers were free
     547      to make modifications and improvements to programs, those
     548      modifications were shared in the first instance with IBM, which then
     549      considered whether and in what way to incorporate those changes in the
     550      centrally-developed and distributed version of the software. Thus in
     551      two important senses the best computer software in the world was free:
     552      it cost nothing to acquire, and the terms on which it was furnished
     553      both allowed and encouraged experimentation, change, and improvement
     554      <footnote><para>13. This description elides some details. By the
     555      mid-1970's IBM had acquired meaningful competition in the mainframe
     556      computer business, while the large-scale antitrust action brought
     557      against it by the U.S. government prompted the decision to "unbundle,"
     558      or charge separately, for software. In this less important sense,
     559      software ceased to be free. But - without entering into the now-dead
     560      but once-heated controversy over IBM's software pricing policies - the
     561      unbundling revolution had less effect on the social practices of
     562      software manufacture than might be supposed. As a fellow responsible
     563      for technical improvement of one programming language product at IBM
     564      from 1979 to 1984, for example, I was able to treat the product as
     565      "almost free," that is, to discuss with users the changes they had
     566      proposed or made in the programs, and to engage with them in
     567      cooperative development of the product for the benefit of all
     568      users.</para> </footnote>.  That the software in question was IBM's
     569      property under prevailing copyright law certainly established some
     570      theoretical limits on users' ability to distribute their improvements
     571      or adaptations to others, but in practice mainframe software was
     572      cooperatively developed by the dominant hardware manufacturer and its
     573      technically-sophisticated users, employing the manufacturer's
     574      distribution resources to propagate the resulting improvements through
     575      the user community. The right to exclude others, one of the most
     576      important "sticks in the bundle" of property rights (in an image
     577      beloved of the United States Supreme Court), was practically
     578      unimportant, or even undesirable, at the heart of the software
     579      business <footnote> <para>14. This description is highly compressed,
     580      and will seem both overly simplified and unduly rosy to those who also
     581      worked in the industry during this period of its
     582      development. Copyright protection of computer software was a
     583      controversial subject in the 1970's, leading to the famous CONTU
     584      commission and its mildly pro-copyright recommendations of 1979. And
     585      IBM seemed far less cooperative to its users at the time than this
     586      sketch makes out. But the most important element is the contrast with
     587      the world created by the PC, the Internet, and the dominance of
     588      Microsoft, with the resulting impetus for the free software movement,
     589      and I am here concentrating on the features that express that
     590      contrast.</para></footnote>.</para>
     591
     592      <para>After 1980, everything was different. The world of mainframe
     593      hardware gave way within ten years to the world of the commodity PC.
     594      And, as a contingency of the industry's development, the single most
     595      important element of the software running on that commodity PC, the
     596      operating system, became the sole significant product of a company
     597      that made no hardware. High-quality basic software ceased to be part
     598      of the product-differentiation strategy of hardware
     599      manufacturers. Instead, a firm with an overwhelming share of the
     600      market, and with the near-monopolist's ordinary absence of interest in
     601      fostering diversity, set the practices of the software industry. In
     602      such a context, the right to exclude others from participation in the
     603      product's formation became profoundly important. Microsoft's power in
     604      the market rested entirely on its ownership of the Windows source
     605      code.</para>
     606
     607      <para>To Microsoft, others' making of "derivative works," otherwise
     608      known as repairs and improvements, threatened the central asset of the
     609      business. Indeed, as subsequent judicial proceedings have tended to
     610      establish, Microsoft's strategy as a business was to find innovative
     611      ideas elsewhere in the software marketplace, buy them up and either
     612      suppress them or incorporate them in its proprietary product. The
     613      maintenance of control over the basic operation of computers
     614      manufactured, sold, possessed, and used by others represented profound
     615      and profitable leverage over the development of the culture <footnote>
     616      <para>15. I discuss the importance of PC software in this context, the
     617      evolution of "the market for eyeballs" and "the sponsored life" in
     618      other chapters of my forthcoming book, <emphasis>The Invisible
     619      Barbecue</emphasis>, of which this essay forms a part.</para>
     620      </footnote>.; the right to exclude returned to center stage in the
     621      concept of software as property.</para>
     622
     623      <para>The result, so far as the quality of software was concerned, was
     624      disastrous. The monopoly was a wealthy and powerful corporation that
     625      employed a large number of programmers, but it could not possibly
     626      afford the number of testers, designers, and developers required to
     627      produce flexible, robust and technically-innovative software
     628      appropriate to the vast array of conditions under which increasingly
     629      ubiquitous personal computers operated. Its fundamental marketing
     630      strategy involved designing its product for the least
     631      technically-sophisticated users, and using "fear, uncertainty, and
     632      doubt" (known within Microsoft as "FUD") to drive sophisticated users
     633      away from potential competitors, whose long-term survivability in the
     634      face of Microsoft's market power was always in question.</para>
     635
     636      <para>Without the constant interaction between users able to repair
     637      and improve and the operating system's manufacturer, the inevitable
     638      deterioration of quality could not be arrested. But because the
     639      personal computer revolution expanded the number of users
     640      exponentially, almost everyone who came in contact with the resulting
     641      systems had nothing against which to compare them. Unaware of the
     642      standards of stability, reliability, maintainability and effectiveness
     643      that had previously been established in the mainframe world, users of
     644      personal computers could hardly be expected to understand how badly,
     645      in relative terms, the monopoly's software functioned. As the power
     646      and capacity of personal computers expanded rapidly, the defects of
     647      the software were rendered less obvious amidst the general increase of
     648      productivity. Ordinary users, more than half afraid of the technology
     649      they almost completely did not understand, actually welcomed the
     650      defectiveness of the software. In an economy undergoing mysterious
     651      transformations, with the concomitant destabilization of millions of
     652      careers, it was tranquilizing, in a perverse way, that no personal
     653      computer seemed to be able to run for more than a few consecutive
     654      hours without crashing. Although it was frustrating to lose work in
     655      progress each time an unnecessary failure occurred, the evident
     656      fallibility of computers was intrinsically reassuring <footnote>
     657      <para>16. This same pattern of ambivalence, in which bad programming
     658      leading to widespread instability in the new technology is
     659      simultaneously frightening and reassuring to technical incompetents,
     660      can be seen also in the primarily-American phenomenon of Y2K
     661      hysteria.</para> </footnote> .</para>
     662
     663      <para>None of this was necessary. The low quality of personal computer
     664      software could have been reversed by including users directly in the
     665      inherently evolutionary process of software design and implementation.
     666      A Lamarckian mode, in which improvements could be made anywhere, by
     667      anyone, and inherited by everyone else, would have wiped out the
     668      deficit, restoring to the world of the PC the stability and
     669      reliability of the software made in the quasi-propertarian environment
     670      of the mainframe era. But the Microsoft business model precluded
     671      Lamarckian inheritance of software improvements. Copyright doctrine,
     672      in general and as it applies to software in particular, biases the
     673      world towards creationism; in this instance, the problem is that BillG
     674      the Creator was far from infallible, and in fact he wasn't even
     675      trying.</para> <!--<center><img src="anarchism_files/mog2.gif"
     676      hspace="0" vspace="0"></center>--> <para>To make the irony more
     677      severe, the growth of the network rendered the non-propertarian
     678      alternative even more practical. What scholarly and popular writing
     679      alike denominate as a thing ("the Internet") is actually the name of a
     680      social condition: the fact that everyone in the network society is
     681      connected directly, without intermediation, to everyone else
     682      <footnote> <para>17. The critical implications of this simple
     683      observation about our metaphors are worked out in "How Not to Think
     684      about 'The Internet'," in <emphasis>The Invisible Barbecue</emphasis>,
     685      forthcoming.</para> </footnote>. The global interconnection of
     686      networks eliminated the bottleneck that had required a centralized
     687      software manufacturer to rationalize and distribute the outcome of
     688      individual innovation in the era of the mainframe.</para>
     689
     690      <para>And so, in one of history's little ironies, the global triumph
     691      of bad software in the age of the PC was reversed by a surprising
     692      combination of forces: the social transformation initiated by the
     693      network, a long-discarded European theory of political economy, and a
     694      small band of programmers throughout the world mobilized by a single
     695      simple idea.</para>
     696
     697    </section>
     698    <section>
     699
     700      <title>Software Wants to Be Free; or, How We Stopped Worrying and
     701      Learned to Love the Bomb</title>
     702
     703      <para>Long before the network of networks was a practical reality,
     704      even before it was an aspiration, there was a desire for computers to
     705      operate on the basis of software freely available to everyone. This
     706      began as a reaction against propertarian software in the mainframe
     707      era, and requires another brief historical digression.</para>
     708
     709      <para>Even though IBM was the largest seller of general purpose
     710      computers in the mainframe era, it was not the largest designer and
     711      builder of such hardware. The telephone monopoly, American Telephone
     712      &amp; Telegraph, was in fact larger than IBM, but it consumed its
     713      products internally. And at the famous Bell Labs research arm of the
     714      telephone monopoly, in the late 1960's, the developments in computer
     715      languages previously described gave birth to an operating system
     716      called Unix.</para>
     717
     718      <para>The idea of Unix was to create a single, scalable operating
     719      system to exist on all the computers, from small to large, that the
     720      telephone monopoly made for itself. To achieve this goal meant writing
     721      an operating system not in machine language, nor in an assembler whose
     722      linguistic form was integral to a particular hardware design, but in a
     723      more expressive and generalized language. The one chosen was also a
     724      Bell Labs invention, called "C" <footnote> <para>18. Technical readers
     725      will again observe that this compresses developments occurring from
     726      1969 through 1973.</para> </footnote>. The C language became common,
     727      even dominant, for many kinds of programming tasks, and by the late
     728      1970's the Unix operating system written in that language had been
     729      transferred (or "ported," in professional jargon) to computers made by
     730      many manufacturers and of many designs.</para>
     731
     732      <para>AT&amp;T distributed Unix widely, and because of the very design
     733      of the operating system, it had to make that distribution in C source
     734      code.  But AT&amp;T retained ownership of the source code and
     735      compelled users to purchase licenses that prohibited redistribution
     736      and the making of derivative works. Large computing centers, whether
     737      industrial or academic, could afford to purchase such licenses, but
     738      individuals could not, while the license restrictions prevented the
     739      community of programmers who used Unix from improving it in an
     740      evolutionary rather than episodic fashion. And as programmers
     741      throughout the world began to aspire to and even expect a personal
     742      computer revolution, the "unfree" status of Unix became a source of
     743      concern.</para>
     744
     745      <para>Between 1981 and 1984, one man envisioned a crusade to change
     746      the situation. Richard M. Stallman, then an employee of MIT's
     747      Artificial Intelligence Laboratory, conceived the project of
     748      independent, collaborative redesign and implementation of an operating
     749      system that would be true free software. In Stallman's phrase, free
     750      software would be a matter of freedom, not of price. Anyone could
     751      freely modify and redistribute such software, or sell it, subject only
     752      to the restriction that he not try to reduce the rights of others to
     753      whom he passed it along. In this way free software could become a
     754      self-organizing project, in which no innovation would be lost through
     755      proprietary exercises of rights. The system, Stallman decided, would
     756      be called GNU, which stood (in an initial example of a taste for
     757      recursive acronyms that has characterized free software ever since),
     758      for "GNU's Not Unix."  Despite misgivings about the fundamental design
     759      of Unix, as well as its terms of distribution, GNU was intended to
     760      benefit from the wide if unfree source distribution of Unix. Stallman
     761      began Project GNU by writing components of the eventual system that
     762      were also designed to work without modification on existing Unix
     763      systems. Development of the GNU tools could thus proceed directly in
     764      the environment of university and other advanced computing centers
     765      around the world.</para>
     766
     767      <para>The scale of such a project was immense. Somehow, volunteer
     768      programmers had to be found, organized, and set to work building all
     769      the tools that would be necessary for the ultimate construction.
     770      Stallman himself was the primary author of several fundamental tools.
     771      Others were contributed by small or large teams of programmers
     772      elsewhere, and assigned to Stallman's project or distributed
     773      directly. A few locations around the developing network became
     774      archives for the source code of these GNU components, and throughout
     775      the 1980's the GNU tools gained recognition and acceptance by Unix
     776      users throughout the world. The stability, reliability, and
     777      maintainability of the GNU tools became a by-word, while Stallman's
     778      profound abilities as a designer continued to outpace, and provide
     779      goals for, the evolving process. The award to Stallman of a MacArthur
     780      Fellowship in 1990 was an appropriate recognition of his conceptual
     781      and technical innovations and their social consequences.</para>
     782
     783      <para>Project GNU, and the Free Software Foundation to which it gave
     784      birth in 1985, were not the only source of free software
     785      ideas. Several forms of copyright license designed to foster free or
     786      partially free software began to develop in the academic community,
     787      mostly around the Unix environment. The University of California at
     788      Berkeley began the design and implementation of another version of
     789      Unix for free distribution in the academic community. BSD Unix, as it
     790      came to be known, also treated AT&amp;T's Unix as a design
     791      standard. The code was broadly released and constituted a reservoir of
     792      tools and techniques, but its license terms limited the range of its
     793      application, while the elimination of hardware-specific proprietary
     794      code from the distribution meant that no one could actually build a
     795      working operating system for any particular computer from BSD. Other
     796      university-based work also eventuated in quasi-free software; the
     797      graphical user interface (or GUI) for Unix systems called X Windows,
     798      for example, was created at MIT and distributed with source code on
     799      terms permitting free modification. And in 1989-1990, an undergraduate
     800      computer science student at the University of Helsinki, Linus
     801      Torvalds, began the project that completed the circuit and fully
     802      energized the free software vision.</para>
     803
     804      <para>What Torvalds did was to begin adapting a computer science
     805      teaching tool for real life use. Andrew Tannenbaum's MINIX kernel
     806      <footnote> <para>19. Operating systems, even Windows (which hides the
     807      fact from its users as thoroughly as possible), are actually
     808      collections of components, rather than undivided unities. Most of what
     809      an operating system does (manage file systems, control process
     810      execution, etc.) can be abstracted from the actual details of the
     811      computer hardware on which the operating system runs. Only a small
     812      inner core of the system must actually deal with the eccentric
     813      peculiarities of particular hardware.  Once the operating system is
     814      written in a general language such as C, only that inner core, known
     815      in the trade as the kernel, will be highly specific to a particular
     816      computer architecture.</para> </footnote> , was a staple of Operating
     817      Systems courses, providing an example of basic solutions to basic
     818      problems. Slowly, and at first without recognizing the intention,
     819      Linus began turning the MINIX kernel into an actual kernel for Unix on
     820      the Intel x86 processors, the engines that run the world's commodity
     821      PCs. As Linus began developing this kernel, which he named Linux, he
     822      realized that the best way to make his project work would be to adjust
     823      his design decisions so that the existing GNU components would be
     824      compatible with his kernel.</para>
     825
     826      <para>The result of Torvalds' work was the release on the net in 1991
     827      of a sketchy working model of a free software kernel for a Unix-like
     828      operating system for PCs, fully compatible with and designed
     829      convergently with the large and high-quality suite of system
     830      components created by Stallman's Project GNU and distributed by the
     831      Free Software Foundation. Because Torvalds chose to release the Linux
     832      kernel under the Free Software Foundation's General Public License, of
     833      which more below, the hundreds and eventually thousands of programmers
     834      around the world who chose to contribute their effort towards the
     835      further development of the kernel could be sure that their efforts
     836      would result in permanently free software that no one could turn into
     837      a proprietary product. Everyone knew that everyone else would be able
     838      to test, improve, and redistribute their improvements. Torvalds
     839      accepted contributions freely, and with a genially effective style
     840      maintained overall direction without dampening enthusiasm. The
     841      development of the Linux kernel proved that the Internet made it
     842      possible to aggregate collections of programmers far larger than any
     843      commercial manufacturer could afford, joined almost non-hierarchically
     844      in a development project ultimately involving more than one million
     845      lines of computer code - a scale of collaboration among geographically
     846      dispersed unpaid volunteers previously unimaginable in human history
     847      <footnote> <para>20. A careful and creative analysis of how Torvalds
     848      made this process work, and what it implies for the social practices
     849      of creating software, was provided by Eric S. Raymond in his seminal
     850      1997 paper, <ulink
     851      url="http://firstmonday.org/issues/issue3_3/raymond/index.html">The
     852      Cathedral and the Bazaar,</ulink> which itself played a significant
     853      role in the expansion of the free software idea.</para>
     854      </footnote>.</para>
     855
     856      <para>By 1994, Linux had reached version 1.0, representing a usable
     857      production kernel. Level 2.0 was reached in 1996, and by 1998, with
     858      the kernel at 2.2.0 and available not only for x86 machines but for a
     859      variety of other machine architectures, GNU/Linux - the combination of
     860      the Linux kernel and the much larger body of Project GNU components -
     861      and Windows NT were the only two operating systems in the world
     862      gaining market share. A Microsoft internal assessment of the situation
     863      leaked in October 1998 and subsequently acknowledged by the company as
     864      genuine concluded that "Linux represents a best-of-breed UNIX, that is
     865      trusted in mission critical applications, and - due to it's [sic] open
     866      source code - has a long term credibility which exceeds many other
     867      competitive OS's." <footnote> <para>21. This is a quotation from what
     868      is known in the trade as the "Halloween memo," which can be found, as
     869      annotated by Eric Raymond, to whom it was leaked, at <ulink
     870      url="http://www.opensource.org/halloween/halloween1.html">
     871      http://www.opensource.org/halloween/halloween1.html</ulink>.</para></footnote>
     872      GNU/Linux systems are now used throughout the world, operating
     873      everything from Web servers at major electronic commerce sites to
     874      "ad-hoc supercomputer" clusters to the network infrastructure of
     875      money-center banks. GNU/Linux is found on the space shuttle, and
     876      running behind-the-scenes computers at (yes) Microsoft. Industry
     877      evaluations of the comparative reliability of Unix systems have
     878      repeatedly shown that Linux is far and away the most stable and
     879      reliable Unix kernel, with a reliability exceeded only by the GNU
     880      tools themselves. GNU/Linux not only out-performs commercial
     881      proprietary Unix versions for PCs in benchmarks, but is renowned for
     882      its ability to run, undisturbed and uncomplaining, for months on end
     883      in high-volume high-stress environments without crashing.</para>
     884
     885      <para>Other components of the free software movement have been equally
     886      successful. Apache, far and away the world's leading Web server
     887      program, is free software, as is Perl, the programming language which
     888      is the lingua franca for the programmers who build sophisticated Web
     889      sites. Netscape Communications now distributes its Netscape
     890      Communicator 5.0 browser as free software, under a close variant of
     891      the Free Software Foundation's General Public License. Major PC
     892      manufacturers, including IBM, have announced plans or are already
     893      distributing GNU/Linux as a customer option on their top-of-the-line
     894      PCs intended for use as Web- and file servers. Samba, a program that
     895      allows GNU/Linux computers to act as Windows NT file servers, is used
     896      worldwide as an alternative to Windows NT Server, and provides
     897      effective low-end competition to Microsoft in its own home market. By
     898      the standards of software quality that have been recognized in the
     899      industry for decades - and whose continuing relevance will be clear to
     900      you the next time your Windows PC crashes - the news at century's end
     901      is unambiguous. The world's most profitable and powerful corporation
     902      comes in a distant second, having excluded all but the real victor
     903      from the race. Propertarianism joined to capitalist vigor destroyed
     904      meaningful commercial competition, but when it came to making good
     905      software, anarchism won.</para>
     906
     907
     908    </section>
     909  </section>
     910  <!--<para><img src="anarchism_files/quad.gif"></para><a name="m3"></a>-->
     911  <section>
     912    <title>III. Anarchism as a Mode of Production</title>
     913   
     914    <para>It's a pretty story, and if only the IPdroid and the econodwarf
     915    hadn't been blinded by theory, they'd have seen it coming. But though
     916    some of us had been working for it and predicting it for years, the
     917    theoretical consequences are so subversive for the thoughtways that
     918    maintain our dwarves and droids in comfort that they can hardly be
     919    blamed for refusing to see. The facts proved that something was wrong
     920    with the "incentives" metaphor that underprops conventional
     921    intellectual property reasoning <footnote> <para>22. As recently as
     922    early 1994 a talented and technically competent (though Windows-using)
     923    law and economics scholar at a major U.S. law school confidently
     924    informed me that free software couldn't possibly exist, because no one
     925    would have any incentive to make really sophisticated programs
     926    requiring substantial investment of effort only to give them
     927    away.</para> </footnote> . But they did more. They provided an initial
     928    glimpse into the future of human creativity in a world of global
     929    interconnection, and it's not a world made for dwarves and
     930    droids.</para>
     931
     932    <para>My argument, before we paused for refreshment in the real world,
     933    can be summarized this way: Software - whether executable programs,
     934    music, visual art, liturgy, weaponry, or what have you - consists of
     935    bitstreams, which although essentially indistinguishable are treated
     936    by a confusing multiplicity of legal categories. This multiplicity is
     937    unstable in the long term for reasons integral to the legal process.
     938    The unstable diversity of rules is caused by the need to distinguish
     939    among kinds of property interests in bitstreams. This need is
     940    primarily felt by those who stand to profit from the socially
     941    acceptable forms of monopoly created by treating ideas as
     942    property. Those of us who are worried about the social inequity and
     943    cultural hegemony created by this intellectually unsatisfying and
     944    morally repugnant regime are shouted down. Those doing the shouting,
     945    the dwarves and the droids, believe that these property rules are
     946    necessary not from any overt yearning for life in Murdochworld -
     947    though a little luxurious co-optation is always welcome - but because
     948    the metaphor of incentives, which they take to be not just an image
     949    but an argument, proves that these rules - despite their lamentable
     950    consequences - are necessary if we are to make good software. The only
     951    way to continue to believe this is to ignore the facts. At the center
     952    of the digital revolution, with the executable bitstreams that make
     953    everything else possible, propertarian regimes not only do not make
     954    things better, they can make things radically worse. Property
     955    concepts, whatever else may be wrong with them, do not enable and have
     956    in fact retarded progress.</para>
     957
     958    <para>
     959      But what is this mysterious alternative? Free software exists, but
     960      what are its mechanisms, and how does it generalize towards a
     961    non-propertarian theory of the digital society?</para>
     962
     963  </section>
     964  <section>
     965
     966    <title>The Legal Theory of Free Software</title>
     967
     968    <para>There is a myth, like most myths partially founded on reality,
     969    that computer programmers are all libertarians. Right-wing ones are
     970    capitalists, cleave to their stock options, and disdain taxes, unions,
     971    and civil rights laws; left-wing ones hate the market and all
     972    government, believe in strong encryption no matter how much nuclear
     973    terrorism it may cause, <footnote> <para>23. This question too
     974    deserves special scrutiny, encrusted as it is with special pleading on
     975    the state-power side. See my brief essay <ulink
     976    url="http://emoglen.law.columbia.edu/my_pubs/yu-encrypt.html">"<emphasis>So
     977    Much for Savages</emphasis>: Navajo 1, Government 0 in Final Moments of
     978    Play."</ulink></para> </footnote> and dislike Bill Gates because he's
     979    rich. There is doubtless a foundation for this belief. But the most
     980    significant difference between political thought inside the digirati
     981    and outside it is that in the network society, anarchism (or more
     982    properly, anti-possessive individualism) is a viable political
     983    philosophy.</para>
     984
     985    <para>The center of the free software movement's success, and the
     986    greatest achievement of Richard Stallman, is not a piece of computer
     987    code. The success of free software, including the overwhelming success
     988    of GNU/Linux, results from the ability to harness extraordinary
     989    quantities of high-quality effort for projects of immense size and
     990    profound complexity. And this ability in turn results from the legal
     991    context in which the labor is mobilized. As a visionary designer
     992    Richard Stallman created more than Emacs, GDB, or GNU. He created the
     993    General Public License.</para>
     994
     995    <!-- <center><img src="anarchism_files/mog3.gif" hspace="0"
     996         vspace="0"></center> --> <para>The GPL, <footnote>
     997    <para>24. <emphasis>See</emphasis> <ulink
     998    url="http://www.fsf.org/copyleft/gpl.txt">GNU General Public License,
     999    Version 2, June 1991.</ulink></para> </footnote> also known as the
     1000    copyleft, uses copyright, to paraphrase Toby Milsom, to counterfeit
     1001    the phenomena of anarchism. As the license preamble expresses
     1002    it:</para>
     1003
     1004    <blockquote><para>When we speak of free software, we are referring to
     1005    freedom, not price. Our General Public Licenses are designed to make
     1006    sure that you have the freedom to distribute copies of free software
     1007    (and charge for this service if you wish), that you receive source
     1008    code or can get it if you want it, that you can change the software or
     1009    use pieces of it in new free programs; and that you know you can do
     1010    these things.</para>
     1011
     1012    <para>To protect your rights, we need to make restrictions that
     1013    forbid anyone to deny you these rights or to ask you to surrender the
     1014    rights.  These restrictions translate to certain responsibilities for
     1015    you if you distribute copies of the software, or if you modify
     1016    it.</para>
     1017
     1018    <para>For example, if you distribute copies of such a program,
     1019    whether gratis or for a fee, you must give the recipients all the
     1020    rights that you have. You must make sure that they, too, receive or
     1021    can get the source code. And you must show them these terms so they
     1022    know their rights.</para>
     1023
     1024    <para>Many variants of this basic free software idea have been
     1025    expressed in licenses of various kinds, as I have already
     1026    indicated. The GPL is different from the other ways of expressing
     1027    these values in one crucial respect. Section 2 of the license provides
     1028    in pertinent part:</para>
     1029
     1030    <para>You may modify your copy or copies of the Program or any
     1031    portion of it, thus forming a work based on the Program, and copy and
     1032    distribute such modifications or work ..., provided that you also meet
     1033    all of these conditions: </para>
     1034
     1035    <para>...</para>
     1036
     1037    <para>b) You must cause any work that you distribute or publish,
     1038    that in whole or in part contains or is derived from the Program or
     1039    any part thereof, to be licensed as a whole at no charge to all third
     1040    parties under the terms of this License.</para></blockquote>
     1041
     1042    <para>Section 2(b) of the GPL is sometimes called "restrictive," but
     1043    its intention is liberating. It creates a commons, to which anyone may
     1044    add but from which no one may subtract. Because of §2(b), each
     1045    contributor to a GPL'd project is assured that she, and all other
     1046    users, will be able to run, modify and redistribute the program
     1047    indefinitely, that source code will always be available, and that,
     1048    unlike commercial software, its longevity cannot be limited by the
     1049    contingencies of the marketplace or the decisions of future
     1050    developers. This "inheritance" of the GPL has sometimes been
     1051    criticized as an example of the free software movement's
     1052    anti-commercial bias.  Nothing could be further from the truth. The
     1053    effect of §2(b) is to make commercial distributors of free software
     1054    better competitors against proprietary software businesses. For
     1055    confirmation of this point, one can do no better than to ask the
     1056    proprietary competitors. As the author of the Microsoft "Halloween"
     1057    memorandum, Vinod Vallopillil, put it:</para>
     1058
     1059    <blockquote><para>The GPL and its aversion to code forking reassures
     1060    customers that they aren't riding an evolutionary `dead-end' by
     1061    subscribing to a particular commercial version of Linux.</para>
     1062
     1063    <para>The "evolutionary dead-end" is the core of the software
     1064    FUD argument <footnote> <para>25. <ulink
     1065    url="http://www.opensource.org/halloween/halloween1.html">V. Vallopillil,
     1066    Open Source Software: A (New?) Development Methodology.</ulink></para>
     1067    </footnote> .</para></blockquote>
     1068
     1069    <para>Translated out of Microspeak, this means that the strategy by
     1070    which the dominant proprietary manufacturer drives customers away from
     1071    competitors - by sowing fear, uncertainty and doubt about other
     1072    software's long-term viability - is ineffective with respect to GPL'd
     1073    programs. Users of GPL'd code, including those who purchase software
     1074    and systems from a commercial reseller, know that future improvements
     1075    and repairs will be accessible from the commons, and need not fear
     1076    either the disappearance of their supplier or that someone will use a
     1077    particularly attractive improvement or a desperately necessary repair
     1078    as leverage for "taking the program private."</para>
     1079
     1080    <para>This use of intellectual property rules to create a commons in
     1081    cyberspace is the central institutional structure enabling the
     1082    anarchist triumph. Ensuring free access and enabling modification at
     1083    each stage in the process means that the evolution of software occurs
     1084    in the fast Lamarckian mode: each favorable acquired characteristic of
     1085    others' work can be directly inherited. Hence the speed with which the
     1086    Linux kernel, for example, outgrew all of its proprietary
     1087    predecessors. Because defection is impossible, free riders are
     1088    welcome, which resolves one of the central puzzles of collective
     1089    action in a propertarian social system.</para>
     1090
     1091    <para>Non-propertarian production is also directly responsible for the
     1092    famous stability and reliability of free software, which arises from
     1093    what Eric Raymond calls "Linus' law": With enough eyeballs, all bugs
     1094    are shallow. In practical terms, access to source code means that if I
     1095    have a problem I can fix it. Because I can fix it, I almost never have
     1096    to, because someone else has almost always seen it and fixed it
     1097    first.</para>
     1098
     1099    <para>For the free software community, commitment to anarchist
     1100    production may be a moral imperative; as Richard Stallman wrote, it's
     1101    about freedom, not about price. Or it may be a matter of utility,
     1102    seeking to produce better software than propertarian modes of work
     1103    will allow.  From the droid point of view, the copyleft represents the
     1104    perversion of theory, but better than any other proposal over the past
     1105    decades it resolves the problems of applying copyright to the
     1106    inextricably merged functional and expressive features of computer
     1107    programs. That it produces better software than the alternative does
     1108    not imply that traditional copyright principles should now be
     1109    prohibited to those who want to own and market inferior software
     1110    products, or (more charitably) whose products are too narrow in appeal
     1111    for communal production. But our story should serve as a warning to
     1112    droids: The world of the future will bear little relation to the world
     1113    of the past. The rules are now being bent in two directions. The
     1114    corporate owners of "cultural icons" and other assets who seek
     1115    ever-longer terms for corporate authors, converting the "limited Time"
     1116    of Article I, §8 into a freehold have naturally been whistling music
     1117    to the android ear <footnote> <para>26. The looming expiration of
     1118    Mickey Mouse's ownership by Disney requires, from the point of view of
     1119    that wealthy "campaign contributor," for example, an alteration of the
     1120    general copyright law of the United States. See "Not Making it Any
     1121    More?  Vaporizing the Public Domain," in <emphasis>The Invisible
     1122    Barbecue</emphasis>, forthcoming.</para> </footnote> .  After all, who bought
     1123    the droids their concert tickets? But as the propertarian position
     1124    seeks to embed itself ever more strongly, in a conception of copyright
     1125    liberated from the minor annoyances of limited terms and fair use, at
     1126    the very center of our "cultural software" system, the anarchist
     1127    counter-strike has begun. Worse is yet to befall the droids, as we
     1128    shall see. But first, we must pay our final devoirs to the
     1129    dwarves.</para>
     1130
     1131  </section>
     1132  <section>
     1133    <title>Because It's There: Faraday's Magnet and Human Creativity</title>
     1134
     1135    <para>After all, they deserve an answer. Why do people make free
     1136    software if they don't get to profit? Two answers have usually been
     1137    given. One is half-right and the other is wrong, but both are
     1138    insufficiently simple.</para>
     1139
     1140    <para>The wrong answer is embedded in numerous references to "the
     1141    hacker gift-exchange culture." This use of ethnographic jargon
     1142    wandered into the field some years ago and became rapidly, if
     1143    misleadingly, ubiquitous. It reminds us only that the
     1144    economeretricians have so corrupted our thought processes that any
     1145    form of non-market economic behavior seems equal to every other
     1146    kind. But gift-exchange, like market barter, is a propertarian
     1147    institution. Reciprocity is central to these symbolic enactments of
     1148    mutual dependence, and if either the yams or the fish are
     1149    short-weighted, trouble results. Free software, at the risk of
     1150    repetition, is a commons: no reciprocity ritual is enacted there. A
     1151    few people give away code that others sell, use, change, or borrow
     1152    wholesale to lift out parts for something else. Notwithstanding the
     1153    very large number of people (tens of thousands, at most) who have
     1154    contributed to GNU/Linux, this is orders of magnitude less than the
     1155    number of users who make no contribution whatever <footnote>
     1156    <para>27. A recent industry estimate puts the number of Linux systems
     1157    worldwide at 7.5 million. <emphasis>See</emphasis> Josh McHugh, 1998. <ulink
     1158    url="http://www.forbes.com/forbes/98/0810/6203094s1.htm">"Linux: The
     1159    Making of a Global Hack,"</ulink> <emphasis>Forbes</emphasis> (August 10). Because the
     1160    software is freely obtainable throughout the Net, there is no simple
     1161    way to assess actual usage.</para> </footnote>.</para>
     1162
     1163    <para>A part of the right answer is suggested by the claim that free
     1164    software is made by those who seek reputational compensation for their
     1165    activity. Famous Linux hackers, the theory is, are known all over the
     1166    planet as programming deities. From this they derive either enhanced
     1167    self-esteem or indirect material advancement <footnote> <para>28. Eric
     1168    Raymond is a partisan of the "ego boost" theory, to which he adds
     1169    another faux-ethnographic comparison, of free software composition to
     1170    the Kwakiutl potlatch. <emphasis>See</emphasis> Eric S. Raymond, 1998. <ulink
     1171    url="http://firstmonday.org/issues/issue3_10/raymond/index.html">Homesteading
     1172    the Noosphere.</ulink>.  But the potlatch, certainly a form of status
     1173    competition, is unlike free software for two fundamental reasons: it
     1174    is essentially hierarchical, which free software is not, and, as we
     1175    have known since Thorstein Veblen first called attention to its
     1176    significance, it is a form of conspicuous waste. <emphasis>See</emphasis> Thorstein
     1177    Veblen, 1967. <emphasis>The Theory of the Leisure Class.</emphasis> New York:
     1178    Viking, p. 75. These are precisely the grounds which distinguish the
     1179    anti-hierarchical and utilitiarian free software culture from its
     1180    propertarian counterparts.</para></footnote>.  But the programming
     1181    deities, much as they have contributed to free software, have not done
     1182    the bulk of the work. Reputations, as Linus Torvalds himself has often
     1183    pointed out, are made by willingly acknowledging that it was all done
     1184    by someone else. And, as many observers have noted, the free software
     1185    movement has also produced superlative
     1186    documentation. Documentation-writing is not what hackers do to attain
     1187    cool, and much of the documentation has been written by people who
     1188    didn't write the code. Nor must we limit the indirect material
     1189    advantages of authorship to increases in reputational capital.  Most
     1190    free software authors I know have day jobs in the technology
     1191    industries, and the skills they hone in the more creative work they do
     1192    outside the market no doubt sometimes measurably enhance their value
     1193    within it. And as the free software products gained critical mass and
     1194    became the basis of a whole new set of business models built around
     1195    commercial distribution of that which people can also get for nothing,
     1196    an increasing number of people are specifically employed to write free
     1197    software. But in order to be employable in the field, they must
     1198    already have established themselves there. Plainly, then, this motive
     1199    is present, but it isn't the whole explanation.</para>
     1200
     1201    <para>Indeed, the rest of the answer is just too simple to have
     1202    received its due. The best way to understand is to follow the brief
     1203    and otherwise unsung career of an initially-grudging free software
     1204    author.  Microsoft's Vinod Vallopillil, in the course of writing the
     1205    competitive analysis of Linux that was leaked as the second of the
     1206    famous "Halloween memoranda," bought and installed a Linux system on
     1207    one of his office computers. He had trouble because the (commercial)
     1208    Linux distribution he installed did not contain a daemon to handle the
     1209    DHCP protocol for assignment of dynamic IP addresses. The result was
     1210    important enough for us to risk another prolonged exposure to the
     1211    Microsoft Writing Style:</para>
     1212
     1213    <blockquote><para>A small number of Web sites and FAQs later, I found an FTP
     1214    site with a Linux DHCP client. The DHCP client was developed by an
     1215    engineer employed by Fore Systems (as evidenced by his e-mail address;
     1216    I believe, however, that it was developed in his own free time). A
     1217    second set of documentation/manuals was written for the DHCP client by
     1218    a hacker in <emphasis>Hungary</emphasis> which provided relatively simple
     1219    instructions on how to install/load the client.</para>
     1220
     1221    <para>I downloaded &amp; uncompressed the client and typed two
     1222    simple commands:</para>
     1223
     1224    <para>Make - compiles the client binaries</para>
     1225
     1226    <para>Make Install -installed the binaries as a Linux Daemon</para>
     1227
     1228    <para>Typing "DHCPCD" (for DHCP Client Daemon) on the command
     1229    line triggered the DHCP discovery process and voila, I had IP
     1230    networking running.  </para>
     1231
     1232    <para>Since I had just downloaded the DHCP client code, on an
     1233    impulse I played around a bit. Although the client wasn't as
     1234    extensible as the DHCP client we are shipping in NT5 (for example, it
     1235    won't query for arbitrary options &amp; store results), it was obvious
     1236    how I could write the additional code to implement this functionality.
     1237    The full client consisted of about 2,600 lines of code.</para>
     1238
     1239    <para>One example of esoteric, extended functionality that was
     1240    clearly patched in by a third party was a set of routines to that
     1241    would pad the DHCP request with host-specific strings required by
     1242    Cable Modem / ADSL sites.</para>
     1243
     1244    <para>A few other steps were required to configure the DHCP
     1245    client to auto-start and auto-configure my Ethernet interface on boot
     1246    but these were documented in the client code and in the DHCP
     1247    documentation from the Hungarian developer.</para>
     1248
     1249    <para>I'm a poorly skilled UNIX programmer but it was
     1250    immediately obvious to me how to incrementally extend the DHCP client
     1251    code (the feeling was exhilarating and addictive).</para>
     1252
     1253    <para>Additionally, due directly to GPL + having the full development
     1254    environment in front of me, I was in a position where I could write up
     1255    my changes and e-mail them out within a couple of hours (in contrast
     1256    to how things like this would get done in NT). Engaging in that
     1257    process would have prepared me for a larger, more ambitious Linux
     1258    project in the future <footnote><para>29. Vinod Vallopillil, <ulink
     1259    url="http://www.opensource.org/halloween/halloween2.html">Linux OS
     1260    Competitive Analysis (Halloween II).</ulink> Note Vallopillil's
     1261    surprise that a program written in California had been subsequently
     1262    documented by a programmer in Hungary.</para>
     1263    </footnote>.</para></blockquote>
     1264
     1265    <para>"The feeling was exhilarating and addictive." Stop the presses:
     1266    Microsoft experimentally verifies Moglen's Metaphorical Corollary to
     1267    Faraday's Law. Wrap the Internet around every brain on the planet and
     1268    spin the planet. Software flows in the wires. It's an emergent
     1269    property of human minds to create. "Due directly to the GPL," as
     1270    Vallopillil rightly pointed out, free software made available to him
     1271    an exhilarating increase in his own creativity, of a kind not
     1272    achievable in his day job working for the Greatest Programming Company
     1273    on Earth. If only he had e-mailed that first addictive fix, who knows
     1274    where he'd be now?</para>
     1275
     1276    <para>So, in the end, my dwarvish friends, it's just a human thing.
     1277    Rather like why Figaro sings, why Mozart wrote the music for him to
     1278    sing to, and why we all make up new words: Because we can. Homo
     1279    ludens, meet Homo faber. The social condition of global
     1280    interconnection that we call the Internet makes it possible for all of
     1281    us to be creative in new and previously undreamed-of ways. Unless we
     1282    allow "ownership" to interfere. Repeat after me, ye dwarves and men:
     1283    Resist the resistance!</para>
     1284
     1285  </section>
     1286  <!--<para><img src="anarchism_files/quad.gif"></para><a name="m4"></a>-->
     1287
     1288  <section>
     1289    <title>IV. Their Lordships Die in the Dark?</title>
     1290
     1291    <para>For the IPdroid, fresh off the plane from a week at Bellagio
     1292    paid for by Dreamworks SKG, it's enough to cause indigestion.</para>
     1293
     1294    <para>Unlock the possibilities of human creativity by connecting
     1295    everyone to everyone else? Get the ownership system out of the way so
     1296    that we can all add our voices to the choir, even if that means
     1297    pasting our singing on top of the Mormon Tabernacle and sending the
     1298    output to a friend? No one sitting slack-jawed in front of a televised
     1299    mixture of violence and imminent copulation carefully devised to
     1300    heighten the young male eyeball's interest in a beer commercial? What
     1301    will become of civilization? Or at least of copyright teachers?</para>
     1302
     1303    <para>But perhaps this is premature. I've only been talking about
     1304    software. Real software, the old kind, that runs computers. Not like
     1305    the software that runs DVD players, or the kind made by the Grateful
     1306    Dead. "Oh yes, the Grateful Dead. Something strange about them, wasn't
     1307    there? Didn't prohibit recording at their concerts. Didn't mind if
     1308    their fans rather riled the recording industry. Seem to have done all
     1309    right, though, you gotta admit. Senator Patrick Leahy, isn't he a
     1310    former Deadhead? I wonder if he'll vote to extend corporate authorship
     1311    terms to 125 years, so that Disney doesn't lose The Mouse in 2004. And
     1312    those DVD players - they're computers, aren't they?"</para>
     1313
     1314    <para>In the digital society, it's all connected. We can't depend for
     1315    the long run on distinguishing one bitstream from another in order to
     1316    figure out which rules apply. What happened to software is already
     1317    happening to music. Their recording industry lordships are now
     1318    scrambling wildly to retain control over distribution, as both
     1319    musicians and listeners realize that the middlepeople are no longer
     1320    necessary. The Great Potemkin Village of 1999, the so-called Secure
     1321    Digital Music Initiative, will have collapsed long before the first
     1322    Internet President gets inaugurated, for simple technical reasons as
     1323    obvious to those who know as the ones that dictated the triumph of
     1324    free software <footnote> <para>30. See "They're Playing Our Song: The
     1325    Day the Music Industry Died," in <emphasis>The Invisible Barbecue</emphasis>,
     1326    forthcoming.</para> </footnote> . The anarchist revolution in music is
     1327    different from the one in software <emphasis>tout court</emphasis>, but here too -
     1328    as any teenager with an MP3 collection of self-released music from
     1329    unsigned artists can tell you - theory has been killed off by the
     1330    facts. Whether you are Mick Jagger, or a great national artist from
     1331    the third world looking for a global audience, or a garret-dweller
     1332    reinventing music, the recording industry will soon have nothing to
     1333    offer you that you can't get better for free.  And music doesn't sound
     1334    worse when distributed for free, pay what you want directly to the
     1335    artist, and don't pay anything if you don't want to. Give it to your
     1336    friends; they might like it.</para>
     1337
     1338    <para>
     1339      What happened to music is also happening to news. The wire services,
     1340      as any U.S. law student learns even before taking the near-obligatory
     1341      course in Copyright for Droids, have a protectible property interest
     1342      in their expression of the news, even if not in the facts the news
     1343      reports <footnote><para>31. International News Service v. Associated
     1344      Press, 248 U.S. 215 (1918). With regard to the actual terse, purely
     1345      functional expressions of breaking news actually at stake in the
     1346      jostling among wire services, this was always a distinction only a
     1347      droid could love.</para></footnote>.  So why are they now giving all
     1348      their output away? Because in the world of the Net, most news is
     1349      commodity news. And the original advantage of the news gatherers, that
     1350      they were internally connected in ways others were not when
     1351      communications were expensive, is gone. Now what matters is collecting
     1352      eyeballs to deliver to advertisers. It isn't the wire services that
     1353      have the advantage in covering Kosovo, that's for sure. Much less
     1354      those paragons of "intellectual" property, their television
     1355      lordships. They, with their overpaid pretty people and their massive
     1356      technical infrastructure, are about the only organizations in the
     1357      world that can't afford to be everywhere all the time. And then they
     1358      have to limit themselves to ninety seconds a story, or the eyeball
     1359      hunters will go somewhere else. So who makes better news, the
     1360    propertarians or the anarchists?  We shall soon see.</para>
     1361
     1362    <para>Oscar Wilde says somewhere that the problem with socialism is
     1363    that it takes up too many evenings. The problems with anarchism as a
     1364    social system are also about transaction costs. But the digital
     1365    revolution alters two aspects of political economy that have been
     1366    otherwise invariant throughout human history. All software has zero
     1367    marginal cost in the world of the Net, while the costs of social
     1368    coordination have been so far reduced as to permit the rapid formation
     1369    and dissolution of large-scale and highly diverse social groupings
     1370    entirely without geographic limitation <footnote> <para>32. See "No
     1371    Prodigal Son: The Political Theory of Universal Interconnection," in
     1372    <emphasis>The Invisible Barbecue</emphasis>, forthcoming.</para> </footnote> . Such
     1373    fundamental change in the material circumstances of life necessarily
     1374    produces equally fundamental changes in culture. Think not? Tell it to
     1375    the Iroquois. And of course such profound shifts in culture are
     1376    threats to existing power relations. Think not? Ask the Chinese
     1377    Communist Party.  Or wait 25 years and see if you can find them for
     1378    purposes of making the inquiry.</para>
     1379
     1380    <para>In this context, the obsolescence of the IPdroid is neither
     1381    unforseeable nor tragic. Indeed it may find itself clanking off into
     1382    the desert, still lucidly explaining to an imaginary room the
     1383    profitably complicated rules for a world that no longer exists. But at
     1384    least it will have familiar company, recognizable from all those
     1385    glittering parties in Davos, Hollywood, and Brussels. Our Media Lords
     1386    are now at handigrips with fate, however much they may feel that the
     1387    Force is with them. The rules about bitstreams are now of dubious
     1388    utility for maintaining power by co-opting human creativity. Seen
     1389    clearly in the light of fact, these Emperors have even fewer clothes
     1390    than the models they use to grab our eyeballs. Unless supported by
     1391    user-disabling technology, a culture of pervasive surveillance that
     1392    permits every reader of every "property" to be logged and charged, and
     1393    a smokescreen of droid-breath assuring each and every young person
     1394    that human creativity would vanish without the benevolent aristocracy
     1395    of BillG the Creator, Lord Murdoch of Everywhere, the Spielmeister and
     1396    the Lord High Mouse, their reign is nearly done. But what's at stake
     1397    is the control of the scarcest resource of all: our
     1398    attention. Conscripting that makes all the money in the world in the
     1399    digital economy, and the current lords of the earth will fight for
     1400    it. Leagued against them are only the anarchists: nobodies, hippies,
     1401    hobbyists, lovers, and artists. The resulting unequal contest is the
     1402    great political and legal issue of our time.  Aristocracy looks hard
     1403    to beat, but that's how it looked in 1788 and 1913 too. It is, as Chou
     1404    En-Lai said about the meaning of the French Revolution, too soon to
     1405    tell.</para>
     1406
     1407  </section>
     1408  <section>
     1409    <title>About the Author</title>
     1410
     1411    <para>Eben Moglen is Professor of Law &amp; Legal History, Columbia Law School.
     1412    E-mail: <ulink url="mailto:moglen@columbia.edu">Mail: moglen@columbia.edu</ulink></para>
     1413
     1414    <para>Acknowledgments</para>
     1415
     1416    <para>This paper was prepared for delivery at the Buchmann
     1417    International Conference on Law, Technology and Information, at Tel
     1418    Aviv University, May 1999; my thanks to the organizers for their kind
     1419    invitation. I owe much as always to Pamela Karlan for her insight and
     1420    encouragement. I especially wish to thank the programmers throughout
     1421    the world who made free software possible.</para>
     1422
     1423
     1424    <blockquote>
     1425      <para>
     1426      <ulink url="http://firstmonday.org/issues/issue4_8/index.html"><!--<img src="anarchism_files/contents.gif" alt="Contents" align="bottom" border="0">--></ulink> </para>
     1427      <para>
     1428        <ulink url="http://firstmonday.org/issues/index.html"><!--<img src="anarchism_files/index.gif" alt="Index" border="0">--></ulink>
     1429      </para>
     1430    <para>Copyright <ulink url="http://firstmonday.org/copy.html">©</ulink> 1999, First Monday</para></blockquote>
     1431
     1432
     1433  </section>
    14211434</article>
Note: See TracChangeset for help on using the changeset viewer.