Larry Sanger's Journal: Why Collaborative Free Works Should Be Protected by the Law

The part relevant to the John Seigenthaler case is in Part III, "Objections and Replies." Written spring 2004, originally published June 2005 in Adam D. Moore, ed., Information Ethics: Privacy, Property, and Power, University of Washington Press, pp. 191-206.

The Internet has, famously, made possible a radically new sort of collaboration among software and content developers from around the world. The results of many of these collaborations are distributed free of charge, and under licenses that permit further development and free release. Until only recently, such productions--the Linux operating system, many thousands of free software packages, and free content like the Wikipedia encyclopedia--have been received, for the most part, as idealistic, innocuous efforts to bring free stuff to the world. In coming years, the world will be lucky to escape considerable corporate, political, and legal turmoil over this free stuff. So this paper will defend a modest thesis, on utilitarian grounds: the law should be written in such a way as to make it possible for free, collaborative works to survive, because society will greatly benefit from them.

"Open source software" and "open content" might sound like tech-head buzzwords today, but as free works become increasingly usable, they could pose a threat to the profits of some commercial software developers, as well as publishers. Linux has developed from something that only the geekiest of geeks could love, to an actually usable operating system--usable if only by "power users"--and, with further development, it could become a serious competitor of Windows and Macintosh operating systems. Linux is already running many Internet servers. Moreover, there are, for all platforms, free word processors, spreadsheets, graphics programs, media players, and many more kinds of software. The best of these are usable, but occasionally "clunky," and (opinions vary) usually decidedly inferior to the best of the popular commercial products. But this could change, and there is some reason to think that it will change.

The situation is similar with regard to content and publishing. In under four years, the free encyclopedia Wikipedia (found at Wikipedia.org) developed nearly one million articles in dozens of languages (now over 300,000 in English). It is perhaps surprising that such an uncontrolled collaboration can produce work as useful as it is; but, admittedly, as in the case of free software, this "free content" still lacks the reliability of the commercially-available versions. Yet this could change as well, and probably will: if and when a more reliable review process is added to the current, remarkably robust one, it is hard to see what advantages Encyclopaedia Britannica, Microsoft's Encarta, and other proprietary encyclopedias might have over Wikipedia.

A need and motivation similar to what led to the creation of Wikipedia is bound to lead to many free textbooks. Already, virtually all of the major (public domain) classics are available online, and these free copies are starting to be used in coursework. Many free textbooks are available online, although admittedly few have actually been adopted by instructors. But, again, this could change, and probably will: there are enough scholars and teachers in the world, motivated not by money but simply by a desire to teach, who will probably together create, co-edit, and maintain up-to-date textbooks in most fields. It seems to be merely a matter of time before this happens on a large scale.

In short, whatever people can work on together over the Internet and release under a free license, the results of that collaborative work may very well, in coming years, be developed to the point where it will pose a threat to the profits of entities that develop similar work under the old proprietary model. Publishers and software manufacturers clearly have something to worry about from the "free stuff" movement, I and many others think. This paper cannot take the space to support this claim rigorously; the above will have to suffice to make the claim worth considering, at least.

This paper will be focused on the more philosophical aspects of the situation, and in general, on this question: given that copyright violations, libel, and other abuses of free speech will take place when software and content--work for short-- is developed and released under open source/open content licenses, what relief is reasonable and just? This is actually a broader question than I want to address. For one thing, in many cases, despite the unorthodox licensing, no significantly different circumstances present themselves in the case of free works.

What I want to address is a more obviously difficult, and interesting, sort of case. I mean the sort of case in which free works are developed in a strongly collaborative way--i.e., there is no single copyright holder, the project lacks a central, controlling authority, and there is no "endpoint" at which the project is considered complete. I will call such work shopwork (a neologism I will discuss more carefully below).

The purpose of this paper is to clarify, argue for, and defend the claim that shopwork per se deserves protection by the law, which is to say that the law should see to it that it is possible to pursue shopwork. The claim is similar in some ways to the claim that private property per se deserves protection by the law: the institution of shopwork, like that of private property, deserves to be protected, even while the law provides relief against the inevitable abuses of the institution. So I will first clarify what "shopwork" is, then offer a utilitarian argument for the claim, and finally address some objections to the argument.

I. What is shopwork?

Shopwork, briefly, stands for any strongly collaborative, open source/open content work. The name is a portmanteau constructed from "shared open work"; the name arguably has the advantages of suggesting collaboration in both the original meaning of "shopwork" (which implies something constructed or fixed in a shop, perhaps by several workers together) and, with its parts reversed, "workshop" (which implies participatory learning).

So a shopwork has at least two essential features: (1)it is free, or open source/open content; (2)it is "strongly" collaborative. In this section of the paper, I will elaborate on each point.

Shopwork is free. First, there is much jargon and many buzzwords describing a family of related, overlapping concepts, all having to do with guaranteeing that software and content of various sorts remains "free," in various ways: open source, open content, free software, copyleft, open access, freeware, and Creative Commons, as well as good old public domain. Part of the reason for the proliferation of jargon is that different groups have promoted different licenses and projects using their own specific jargon.

To say that a work is free, in the special sense in which I will use it,[1] is to say that it is released to the world (1) free of charge and (2) under a license that permits others to view, copy, and distribute the work, as well as develop and release their own altered versions of the work--as long as their versions are also released under the same open content license. In the case of software, this requires that programmers be able to view the source code of the program. Not only works but also licenses are said to be free (or open source, etc.); the license that makes the work free in the above-defined sense can be called a "free" license.

There are other permissive licenses that are nonetheless not free licenses. For example, if the license were to require that no one sell derivative versions of the work, then that would not be a free license. Or if the license permitted viewing, copying, and distributing exact or verbatim copies of the work, but not developing new versions thereof, that, again, would not be a free license. Or one might declare the work to be in the public domain, available for any legal use whatsoever; but this would permit others to make proprietary versions of the work, and hence this too is different from free work. There are other variants, of course; the variety of kinds of unusual permissions an author or creator might want to grant others is large.[2] But the sort of license that shopwork uses is a free one. In short, such a license guarantees that a work not only is, but will remain, free.

Shopwork is "strongly" collaborative. Secondly, works can be "collaborative" in various ways. At a minimum, shopwork has multiple authors; if a work has a single creator, it is not, or not yet, shopwork. Moreover, these authors do not contribute bits and pieces which are then stitched together, as with patchwork quilts, anthologies, or many reports written by committee; instead, every aspect of any work contributed is open to revision by the other contributors. These two features, multiple authorship and joint editorship, make shopwork collaborative. A third feature makes shopwork strongly collaborative: it is developed not just by multiple authors, but a constantly changing battery of authors, including even, perhaps, some anonymous authors.

I do not want to claim that the arguments below do not apply to work that is collaborative in any less robust sense than this. But the strongest version of the argument can be developed with respect to work that is collaborative in all three ways listed.

The free encyclopedia Wikipedia provides perhaps the best example of a work that is strongly collaborative in this sense. At present--and this seems unlikely to change--virtually anyone with an Internet connection can visit a Wikipedia article, click an "edit" button, and proceed to make changes to the article. Changes are logged on a "recent changes" page, and often (surely not always) duly checked over by members of the Wikipedia "community." This community--not anyone who has special editorial control--is the ultimate arbiter of what is and is not included in Wikipedia. Hence it is easy to see that Wikipedia has multiple authors who act as each others' editors, and that the battery of author/editors is constantly changing.

Radical collaboration is one of the key values of the open source and free software movements. Open source software, it is widely held, is best developed by a body of programmers who develop and edit each others' work, rather than following edicts issued from on high.[3]

But open source software, and free works generally, need not be created by a collaboration. If a programmer does not succeed in attracting other coders to his project, it will remain uncollaborative. Moreover, there is of course nothing stopping a small group of experts in a field from jointly writing a textbook on a subject and then releasing it under a free license. Such a textbook might feature multiple authorship and joint editorship, but not a constantly changing battery of authors.

This being admitted, any very successful free software project will now probably attract a body of developers. This is in part because programmers are used to working together, creating different modules of one whole software project. The situation is currently different for written (prose) works, with one notable exception in Wikipedia. This has a perhaps obvious explanation. A researcher who releases a text, paper, etc., under an open content or open access license (and there is a growing number who have done so) cannot expect to see it developed further by others unless they are mostly agreement with the first author. Even then, the desire (or professional need) for personal credit suggests the obvious response is another paper--not editing and release of someone else's paper.

The situation could turn out to be different with textbooks. The novel concept and collaborative ethos behind free works developed in the context of software, however, and so it is not surprising that the idea should not yet have caught on among academics. But the future success of a more carefully edited Wikipedia might demonstrate the power of a new, strongly collaborative, model of content development. The next section will suggest further support of this prediction.

II. Why Shopwork Per Se Should Be Protected by the Law

I will now argue that at least some shopworks are, precisely in virtue of their being both free and collaborative, potentially of considerable benefit to society. Hence, it is in society's interest to protect shopwork as a new sort of institution, for, without it, some such benefits could not, or not as easily, be obtained.

First note that, in general, free software and content is useful and becoming more so, as I was concerned to point out in this paper's introductory paragraphs. To be sure, this is a benefit of shopwork, but it is not unique to or due to any special features of shopwork. If one wishes to argue on utilitarian grounds, as I do, that shopwork per se deserves the protection of the law, it will have to be on grounds of some benefit of shopwork per se.

To this end, I want to make two observations, that shopworks are autonomous in a certain sense, and that they are or can be perpetual, from which I will conclude that the projects that develop shopworks are or can be natural institutions.

Shopwork, because it is free, is released to a public that may use and develop it with minimal restrictions. Consequently, any person or group who develops a shopwork knows that other persons or groups are free to take the shopwork and develop it in different, and perhaps better, ways. This, in the open source software community, is known as forking a project. In fact, forking is a relatively rare occurrence (for reasons we need not explore), but it is certainly not unheard-of. Hence we may say that shopworks are, in a certain sense, legally autonomous: versions of them exist, or may (legally) exist, independently of any entities that initially develop them. (That, anyway, is the ideal behind their licenses.) The program I work on now and freely release to the world may look very different in ten years, and this is out of my control; the free license is my renunciation of such control.

It seems likely that some projects, such as Wikipedia if it survives until 2050 (and it is hard to see why it would not), could develop into very many different versions over the years. The local control and histories of some versions might be difficult to trace, obscured by anonymous developers, fading memories, and/or poor record-keeping.[4] Such a project is what we might call untethered, in a robust sense--that is, entirely lacking an entity that bears any clear causal responsibility for it. This is not to deny, however, that as long as a work lives on the Internet, there will always be someone legally responsible (a "designated agent," in the United States under the Digital Millennium Copyright Act) for the website contents.

The other point mentioned above is that, because a shopwork has, potentially at least, a changeable body of collaborators, many shopworks, especially autonomous ones, are perpetual; they have no set endpoint. Unlike most textbooks, for example, the editions do not necessarily end when the author dies. When a constantly-changing collaborative body develops a work, new editions can appear for as long as there are willing personnel. The situation is more similar to that with Encyclopaedia Britannica, a venerable institution, the rights to which have passed from owner to willing owner for many years. But unlike the case of the Britannica, if the particular organization controlling a free work were simply to cease to exist, or to lose interest in developing the project, that would not mean that all project development would also, necessarily, have to end as well. As long as other people can be found to make useful developments to a shopwork, it can live on long past the lifetimes of its founders. This is what I mean when I say that shopworks are perpetual.

Being potentially, at least, autonomous and perpetual, it follows, I think, that successful shopworks are natural institutions, in two senses.

First, we may anticipate that, probably, some of the projects that develop shopworks will become institutions. Perhaps the Free Software Foundation and the Wikimedia Foundation will develop into institutions in this sense.

But, second, one will be able to speak of the results, the shopworks themselves, as institutions as well. A shopwork that develops in enough depth to inspire many forks, or versions, is probably going to stick around for a while; lacking any essential ties to any particular organization, its autonomy ensures that it can last as long as we want it to last, and the fact that it is developed by a strong collaboration ensures that it can outlast any particular person or set of people. So shopworks will probably become institutions in about the same sense that one can speak of law, scientific practice, and marriage as institutions, independently of court systems, research institutions, and particular marriages. It is possible that in one hundred years, books will be written about the rich free operating system tradition, the institution of free encyclopedias, of free, collaborative science textbooks, and perhaps of strongly collaborative research itself.

Let us review the highlights of this section so far: shopworks can and already do have considerable usefulness to society; moreover, both shopworks and their development projects may well become autonomous, perpetual institutions. Now observe that it is one thing to say that a word processing program helps its particular users to write papers, etc. It is another, much more significant thing to say that an institution of free word processing programs can serve a significant, salutary purpose to millions of people around the world over a long span of time. The difference is one of scale. Shopworks, particularly when viewed as institutions, could benefit humanity tremendously.

There is a third sense in which shopwork could constitute an institution--namely, the institution of shopwork itself, one general institution taking all shopwork projects together, in roughly the same way that the general institution of education involves taking, among other things, all individual educational institutions (schools and universities) together, over time. What I want to urge is that society respect and, through its laws, support the existence of the presently-nascent general institution of shopwork.

I will make this thesis a bit more precise later, but more can be said in defense of the thesis as stated in this general form. Being both free and collaborative, shopworks potentially have two significant advantages over proprietary works: low cost and collaboration of unusually large numbers of well-qualified people.

First, perhaps most obviously, shopworks are free of charge. Most people involved in shopwork development are volunteers. This saves a lot of money. For example, it has been estimated that well over one billion dollars in development costs have been saved in the development of Linux and the software bundled with it in Red Hat Linux 7.1; that is, it would have cost a corporation over a billion dollars to develop Red Hat's operating system, user interface, and accompanying software.[5] That cost would have been passed on to users, had the bundle been proprietary. So the users can save that money. Collectively, this will provide many billions of dollars in savings over the long term.

In fairness, it must be admitted that this advantage is still theoretical in large part. As long as, at present, there are difficult-to-quantify costs such as the time spent learning Linux, the inconvenience of using "clunky" amateurish programs, and having little in the way of product support--and other such hidden costs--we can dismiss a simplistic analysis which compares a zero price tag to the price tag of Windows and comparable software. Similarly, few would suggest, at this writing, that the Britannica cannot compete with Wikipedia since Wikipedia is available free of cost: the Britannica obviously has the advantage of reliability that Wikipedia currently lacks. All this, however, with regard to both free software and free content, might well change. It seems to be only a matter of time; remember that free projects have the considerable advantage of perpetuity. Indeed, it could turn out that open source products (and the argument is the same for open content) would end up being on the cutting edge, so that, to keep up, corporations might have to become shopwork developers.[6]

Second, shopworks have been and will continue to be developed by collaboration among unusually large numbers of intelligent, able people. Many programmers and intellectuals have a strong urge to create and to teach, and to a great extent, this desire is independent of a desire for personal financial gain--particularly when such people believe their work will reach many others and not go to profit any one person or corporation in particular. Moreover, very many people enjoy working and learning together, collaboratively. Shopwork combines the latter with the aforementioned desire to create and teach for the general benefit of humanity and is, thus, inherently attractive to many people.

Presently, the firms that are behind the most successful proprietary operating systems, software, textbooks, and encyclopedias have the advantage of having many high-paid employees. What remains to be seen is whether corporations will be able to compete with the depth and breadth of personnel that shopwork projects can front, and not just at any one moment, but institutionally, over time. This is a difficult question. Admittedly, it could turn out that proprietary products will always turn out to be developed by better staff and--hence?--be better products. That suggestion will certainly appeal to some.

But the advantage of proprietary model, in terms of personnel, cannot be taken for granted. Consider, for example, the case of Wikipedia. In its first few years it has become clear that, on many (not all) topics, the ordinary educated public of the Internet can write encyclopedia articles that are equal to or better than those written by experts for proprietary encyclopedias. Already, many of Wikipedia's participants have been very highly credentialed. Over time, as new editorial systems that appeal more to the academic/research turn of mind are tried out, more of the most highly qualified people will be attracted to the project, or to its offshoots or imitators. Just for example, a university might sponsor a vetting process for Wikipedia; some famous scholar might sign on and become a spokesperson for the project; then, suitably "academicized," many more academics might want to get on board. If this happens, it seems unlikely that no traditional proprietary encyclopedias would be able to compete, because--again--shopworks are perpetual and autonomous, and natural institutions. When a high-quality, enormously comprehensive encyclopedia is available free of charge, no business that must pay its employees will be able to compete.

Similar flights of fancy suggest new possibilities for textbooks and--though this raises other issues--the news media, and even research itself. Bear in mind that much research work is intended to be free of the constraints of profit, and a kind of collaborative work (via academic and professional journals) is essentially how the world of research has been organized. So open content research might appeal to academics when they come to understand it and its potential; though not much more collaborative than traditional research (and hence not shopwork), the success of "open access" research publishing indicates the appeal and actual growing popularity of the general idea of free work among researchers.[7]

Now back to the philosophical argument. Suppose the above suggestions are roughly correct. That is, suppose that successful, useful shopworks will very probably develop into institutions. As such, what benefits they offer are magnified. But, moreover, such institutions might well become competitive with proprietary products in terms of both cost and personnel (and hence, quality). Finally, recall that their institutional nature is a function of their being both free and collaborative--which is to say, of their being shopworks. Considering these potential benefits, and that they stem from the very nature of shopwork, we may conclude on utilitarian grounds that society, and particularly the law, should see to it that shopwork is made possible and that shopwork institutions are protected. We may advance a Shopwork Protection Rule:

Laws should be written so as to be consistent with the existence of works that are free as well as strongly collaborative in the above-defined senses.

Given the benefit to society, some may be tempted to conclude a bit more than this: the law should actually support such works, either through funding or other special legislative support. I do not want to urge this stronger conclusion, however. To the contrary, the difficulty is that shopworks are, after all, free. When government gets involved in giving positive support to socially useful projects, whether through funding or through other support, conditions are always imposed; and conditions generally mean restrictions on freedom. Shopwork as an institution is threatened to the extent to which government or any single entity is given authority over it.

The Shopwork Protection Rule has important legal consequences. In particular, the law should treat matters of copyright violation, libel, and other abuses of free speech in the context of shopwork development in a way that is consistent with the autonomous nature of shopworks. That is, whatever remedies are chosen must be consistent with respect for the free, uncontrolled nature of the shopworks. For example, to provide for control over a shopwork, as relief of an abuse, would essentially be to eliminate it as a shopwork. Granted, this might be warranted if, for example, large portions of code were stolen from a proprietary program. This should, however, be the rare exception, not the rule. For similar reasons, the law should jealously enforce contracts and laws that preserve the openness of shopworks. Some might disagree with this, however, which is what I want to address next.

III. Objections and Replies

One might maintain on various grounds that shopwork does not deserve any specific protection, and hence we should not endorse the Shopwork Protection Rule. In the following I will lay out and reply to three objections.

First objection: shopwork is too open to abuse. One might object that, precisely because shopwork is both free and strongly collaborative, it is open to various kinds of abuse; and the only adequate relief is to reduce the degree of freedom or the strength of the collaboration.

Consider an illustration of an abuse. Imagine a wiki--a website in which any visitor can edit any page, like Wikipedia--the contents of which is released under a free license. Suppose the full text of (copyrighted) news articles frequently appears on the wiki, together with many clear examples of libel. The website has repeatedly refused to remove the offending text. Suppose also that the website is for whatever reason popular and influential. The victims of copyright infringement and libel have clear cases against the website and/or its collaborators, and they could have the website shut down.

This fits the definition of 'shopwork' offered above. On first glance, it appears to be precisely the free nature of the website that lends itself to the abuse. But this first impression is mistaken. The fact that the contents of the website can be freely distributed played no role in explaining the copyright infringement or libel. The fact is that neither free works nor collaboration inherently or necessarily invite abuse, any more than an individual's privately owned website does so.

There are, however, related problems associated more closely with strong collaboration and with free works. If it is correct to say that successful shopworks invite large numbers of contributors, the sheer numbers of those involved increases the likelihood that copyright and free speech abuses will occur. But--again--this in itself should not pose a threat to shopwork, any more than it would to any large web content host, as long as there are statutes in place that permit the entity legally responsible for the website (in the U.S., the "designated agent") to avoid a copyright lawsuit by removing copyright violations from the website when given notice to do so.[8] Surely any organization managing a large, successful shopwork will not put its version of the project at risk by tolerating abuses.

What could be a more serious problem with regard to some shopworks, for example Wikipedia, is that wholly anonymous contribution opens the project up to serious liability. A hostile anonymous contributor could threaten the entire project by changing IP numbers and "handles," and deliberately uploading copyrighted and libellous material. But even Wikipedia, a very large and popular website, has been able to manage potential copyright and other free speech problems (so far) that bad-faith anonymous contributors might set up. Generally speaking, the vast majority of the contributors of shopworks are, naturally, people of good will; they far outnumber the bad eggs, to the point that the potentially dangerous contributions of the bad eggs can be prevented and managed without much difficulty.

In this case, the strongest remedy the law might reasonably require is the removal of the option of anonymous contribution. The possibility of anonymous contribution is regarded by some as an essential part of strongly collaborative projects; but it is not, I think, absolutely essential. That is, if a shopwork project were to decide to remove the option of anonymous contribution, it would not be any less of a shopwork. Admittedly, the degree of collaborativeness would be reduced, since those who would refuse to contribute under their own real identities would not participate. Hence, if out of concern over copyright and free speech abuses, the law were to make wholly anonymous contribution impossible, that would not in itself constitute a ban on shopworks.

But what if some small, previously undetected, and particularly damaging bit of libel or copyright violation comes to light, which has been smuggled into some "untethered" shopworks, existing in multiple copies of indeterminate provenance? Surely the law has some reason to take control of the situation, precisely because there is no central authority; there is no other means of relief. Hence we can easily imagine the plaintiff or the government advocating either seizing control of the projects or shutting them all down. And in this case, it is precisely the fact that the offense occurs in the context of a shopwork that is the cause of the problem. The remedy can only be--one might suggest--to undermine the shopwork project in one way or another.

I suspect this is not as serious of a problem as it might sound. Anyone familiar with shopworks knows that unedited copies of shopworks are somewhat like bathroom walls--short of being caught in the act, no one can be held responsible for the mere existence of an obscenity, though the owner of the bathroom might be responsible for painting it over. What, precisely, is the difference between this case and more ordinary cases of copyright and free speech abuse? Only the number of copies in existence, it seems. What difference does it make, as far as the appropriate remedy is concerned, whether a shopwork project cannot trace the origin of a given contribution? In cases of copyright and libel, at the very least, all that the "designated agent" of a website should have to do is remove the offending content. If they are not willing to do so, then perhaps they should be shut down.

Yet the problem might not exist online only, one might argue, and so removing the offending content from websites will not necessarily solve the problem: copies might exist on individual hard drives, CDs, in print, and in other media. The only way to be sure that such abuses are not distributed into posterity, as far as more permanent media is concerned, is not to permit free distribution of collaborative efforts in the first place.

Of course, much libel and copyright violation is even now gathering dust in libraries. The remedies offered for those offenses, against newspapers and publishers for example, did not generally shut down their businesses, much less render impossible the very institution of publishing. Similarly, there will be lawsuits against shopwork projects and those who make use of shopworks (and already have been[9]), and eventually, rich shopwork organizations will very likely be forced to pay large sums of money. Particularly negligent organizations might be forced to shut down. This is as it should be. But it will never be necessary or warranted to shut down the institution of shopwork generally.

Second objection: shopwork is hostile to private enterprise. A second objection can be explained and dealt with more briefly. The institution of shopwork might tend to undermine certain kinds of private enterprise, namely, software, encyclopedia and textbook publishing, and perhaps others; that is part of my own argument, above. But private enterprise is very efficient at and very important for producing goods at low prices. So it should be protected. Therefore--someone might argue--threats to the very existence of parts of it should be prohibited, so shopwork should be prohibited.

This argument, while it might appeal to some corporate executives, will not wash for anyone else, even for defenders of capitalism. If private enterprise is to be protected because it produces cheap goods efficiently, then shopwork should be protected for the same reason. If shopwork does a better job of producing cheap (indeed, free) goods than private enterprise, then so much the worse for private enterprise in the affected sector.

Third objection: there are no shopwork "institutions" yet. An objection aimed specifically at the argument of section II above states that if a given free collaborative work is not useful or does not constitute an institution yet, one cannot argue that it should be protected, at least not on grounds of my argument. After all, a significant aspect of my argument was precisely that the institutional nature of shopwork magnifies the societal benefit of shopwork projects. If my predictions of shopwork institutions do not come to fruition, then, it seems, my argument would turn out not to be nearly as strong as I want it to be. I admit that this is at least possible, though I do not think it is very likely. It is possible that, every five or ten years, software, textbook, reference, etc., and other shopwork projects simply go belly-up and no one is interested in carrying them on. As the best going concern loses its luster for whatever reason, totally independent replacement shopwork projects start up, or, perhaps, no comparable project ever starts up again. In any case, there is no continuous tradition of project development that can be called an institution. In other words, it might turn out that the vaunted hopes of Linux, free software, open content encyclopedias, collaborative textbooks, and the like will have been either a fad of the '90s and '00s, or (puzzlingly) constantly reinventing the wheel, or (most likely) always lagging behind proprietary works.

I should first acknowledge that mine is not the only possible argument for the Shopwork Protection Rule; one could defend the rule on the basis of rights to private property, contract, and free association. I do not wish to elaborate such an argument here; but such an argument would not be open to this particular objection.

Perhaps it is most judicious not to venture a reply to this objection. It is probably most reasonable to want to "wait and see" whether the grand hopes of shopwork and shopwork "institutionalization" ever come to fruition. But if, in the meantime, there are court decisions and legislative efforts that have the general effect of making shopwork impossible (or more difficult), I hope it will not be as a result of corporate lawsuits and lobbying that guard their turf. Then the failure of shopwork to achieve institutionalization would be actually created by law, while the very lawsuits and lobbying in question would themselves point to worries, on the part of corporations or other vested interests, that increasingly powerful shopwork institutions would soon be ruining their operations.

[1] This is similar to (but not exactly the same as) the usage employed by Richard Stallman and the Free Sofware Foundation (see http://www.fsf.org). I use the term merely because it is brief, and neither to endorse Stallman's theory about free software, nor to reject the use of "open source" and "open content"; these latter terms are favored by some of those not fully in agreement with Stallman's theories. Stallman and his followers are careful to point out that there is an important distinction between free in the gratis sense (free of charge) and free in the libre sense (free of most constraints). I think that the argument I am about to make does not work so well if free works are not necessarily understood to be free of charge; hence I include this in the definition. This is not a point I need insist upon strongly, however.

[2] A useful resource to consult in this connection is Creative Commons' page describing its various licenses. See "licenses explained," Creative Commons, retrieved August 3, 2004, from http://creativecommons.org/learn/licenses/

[3] This is one of the main themes of what has become a key manifesto of the open source movement, EricRaymond's essay "The Cathedral and the Bazaar," retrieved August 2, 2004 from Raymond's personal website: http://www.catb.org/~esr/writings/cathedral-bazaar/cathedral-bazaar/

[4] Wikipedia might, perhaps, be a poor example of this, since Wikipedia's "Copyrights" project information page, found at http://en.wikipedia.org/wiki/Wikipedia:Copyrights (retrieved August 1, 2004) presently requires links back to the original article. Hence it is unclear whether a third-generation encyclopedia, i.e., one based on a project that was itself based on Wikipedia, would require links back to Wikipedia; but so it appears. More generally, there are many difficult legal questions that this introductory sort of discussion cannot adequately address.

[5] David A. Wheeler, "More Than a Gigabuck: Estimating GNU/Linux's Size" vers. 1.07, retrieved retrieved August 1, 2004 from Wheeler's personal website: http://www.dwheeler.com/sloc/redhat71-v1/redhat71sloc.html

[6] This point is made by Petr Hrebejk and Tim Boudreau, "The coming 'open monopoly' in software," c|net news.com, October 24, 2001, retrieved August 2, 2004 from http://news.com.com/2010-1071-281588.html?legacy=cnet

[7] See, for example, the Open Access Now website: http://www.biomedcentral.com/openaccess/

[8] In the U.S., this is the function of the Online Copyright Infringement Liability Limitation Act, part of the Digital Millennium Copyright Act (1998).

[9] Most famously, SCO v. IBM. See "SCO v. IBM Linux lawsuit," Wikipedia, retrieved August 1, 2004, from http://en.wikipedia.org/wiki/SCO_v._IBM_Linux_lawsuit