Defining Open Content Licenses
* Article: Open Content, Enclosure and Conversion. Stephen Downes. 2008
URL = http://www.downes.ca/files/FreeLearning.pdf (pp. 122+)
An extented argument for the preservation of the non-commercial clause in open licenses such as the Creative Commons, unless enclosure protection is otherwise included in the sharing agreements, undertaken against the position of David Wiley 
The excerpt below is not a full copy.
"People should not be rewarded for making free things more scarce. Especially when society as a whole benefits so much from their abundance, the creation of artificial scarcities, whether through law, litigation or technology, is a fundamentally destructive and hurtful activity. People should be punished for creating scarcities, not rewarded, as such is the source of illusive, and ill-gotten wealth, founded on misery and deprivation rather than growth and development. The grounds for agreement are thus these: if we could build these into the license, then the non-commercial clause would not be needed."
"One of the challenges is to arrive at a definition of an open license that allows for such agreement.
I say 'definition of an open license' because I now believe that it is possible to arrive at one. But I think it will take a meticulous reworking of a number of the key concepts in order to arrive at this understanding.
One point of agreement I think we can find is in our support for 'free content', in the sense of 'freedom' expressed by Richard Stallman. By 'free as in freedom' he means the following four freedoms (http://www.gnu.org/philosophy/free-sw.html )
- The freedom to run the program, for any purpose (freedom 0).
- The freedom to study how the program works, and adapt it to your needs (freedom 1). Access to the source code is a precondition for this.
- The freedom to redistribute copies so you can help your neighbor (freedom 2).
- The freedom to improve the program, and release your improvements (and modified versions in general) to the public, so that the whole community benefits (freedom 3). Access to the source code is a precondition for this.
Now it is evident that by content we don't mean (exactly) 'software'. So the freedoms are not an exact match, and as Stallman himself says, you can't blindly generalize from software freedoms to other freedoms.
Nonetheless, we can invoke the spirit of the four freedoms to come to some sort of understanding of the content regime we would like to foster: one in which we are able to access (consume?) the content, for any purpose, one in which the source (or encoding) of the content is accessible (ie., non-proprietary), one in which we can share content, and one in which we can modify the content.
Content which respects the spirit of these four freedoms has come to be called open content, and that content intended specifically for educational purposes has come to be called open educational resources. The connection is with the term 'open source', which has come frequently to stand for (what Stallman means by) 'free content'.
No doubt Stallman would prefer that we use the terms 'free content' and 'free educational resources', and there is a good argument for that. But the use of the word 'open' is well-entrenched, and we'll use it here to mean the same thing as 'free', as discussed above.
Point of View
The four freedoms listed above can be interpreted from different perspectives.
This will become very evident in the discussion below. For the sake of the current discussion, I would like to identify two major points of view:
- the content provider - that is, the person who current possesses the content, and would like to use or share it
- the content consumer, that is, the person who does not yet have the content in his or her possession, and who would like to access the content
This distinction is important because there are two ways we can emphasize the impact of the four freedoms on content. The first emphasizes access, that is, that there ought not be any barriers to reading, running, or consuming the content. The second emphasizes use, that is, that there ought not be any limitation on how content is used.
One way of characterizing the point of disagreement is to characterize it as difference in point of view. Specifically, my own view involves an emphasis on access, such that content is not 'free' if there are conditions or constraints that prevent or impair one's ability to read, run or consume content. However, Wiley's view (from my perspective - he is free to characterize this differently) involves an emphasis on use, such that content is not 'free' if there are conditions or constraints that prevent or impair some use of content. Both perspectives live happily together, except for one point of collision: the commercial use of content. Because, on the one hand, the commercial use of content (for example, offering it for sale) can create conditions or constraints that prevent or impair one's ability to read, run or consume that content. And on the other hand, the constraint to non-commercial use of content creates conditions or constraints that prevent or impair some use of content, specifically, commercial use.
That said, each of these perspectives also includes a countervailing perspective. On the one hand, proponents of commercial content may argue that commercialization does not prevent access, because non-commercial sources of content remain available. And proponents of open access argue that the commercialization of content is not actually a 'use' of content, but rather, merely the enclosure of content behind a barrier or wall.
Commercial Use: Agency
Whether or not a conflict in the different perspectives of 'free content' depends critically on what we mean by 'commercial use'. It is evident from the mere existence of the Creative Commons survey that there is considerable uncertainty on the subject.
I commend Creative Commons on its efforts to clarify the issue. The question of what constitutes 'commercial use' is replete with grey areas. For example, 'Is it allowed to show e.g. a CC-licensed photo on a webpage which also includes ads to the side if the image uses the 'non-commercial' clause?" We cannot reply simply on the 'verdict' rendered by Lawrence Lessig in such cases.
We need some clearer understanding.
In order to narrow the range of disagreement, it is useful to divide the set of possible definitions of 'commercial use' into two major categories:
- first, the use of the resource by a 'commercial' organization (this definition parallels the definition of 'educational use' as 'use by an educational organization') such as a commercial publisher, content vendor, or the like, and
- second, the use of the resource in a commercial manner, regardless of the identity of the person or organization using the resource
This distinction is necessary because there is a class of arguments that depend on the characterization of the agency performing the use (such as non-profit or journalist) rather than the use itself. We see this in the Creative Commons guidelines: "Allowable NC users are: (a) an Individual (b) a Non-profit educational institution/library, (c) a Non-profit organization as defined under US or equivalent law... (etc.)."
But it should be clear that we can draw no clear correlation between the nature of the use and the nature of the user. Governments, normally non-commercial users, can engage in commercial activities, such as selling data or charging tolls. And people working for corporations can make non-commercial uses, for example, by merely reading a resource. As in the case of 'educational use', the wording of the Creative Commons guidelines contains a particularly American bias, as though we can identify a 'use' with the type of 'user'. But these guidelines are quite arguably in error. We should not identify 'commercial use' with 'commercial organization' any more than we should identify 'educational use' with 'educational organization'.
The designation of 'commercial' must refer to a type of activity. This is merely a recognition of the success of the commercial sphere, a success so profound that commercial use of resources permeates every sector of society, from individuals running ads on blogs to non-profits selling reports or t-shirts to companies charging subscriptions or running ads.
It is good that Creative Commons is reconsidering its definition. The disastrous interpretation authored by an anonymous contributor, offers no help whatsoever, is not part of the license, and should (and would) carry no legal force, save perhaps in the United States.
Commercial Use: Characteristic Properties
In this this section I would like to argue for the possible contentious position that what we should want to characterize as 'commercial use' is not in fact a use of the resource at all, but rather, as suggested above, a practice of enclosing the resource.
The key to understanding this lies in understanding that the nature of the use is not dependent on the user. A person working for a corporation can engage in quite innocent and non-commercial uses of content; indeed, this is the most common use of content, by corporate and non-corporate users alike.
The question of 'commercial use', indeed, comes up only from the perspective of a content provider, as described above, and never from the perspective of a content consumer. And, indeed, the typing of a use as 'commercial' occurs only in the context of redistribution.
It is this sort of consideration, in my view, that leads David Wiley to say that "the ShareAlike clause is the root of the license compatibility problem - not the NC clause." And I certainly agree to a certain degree - that it is in the (putative) sharing of the content that the impact of the non-commercial clause is felt. I will return to the question of the Share-Alike condition below, and focus on the ides of sharing in particular for the moment.
In particular, I want to be clear, that some types of sharing constitute commercial use, and some types of sharing constitute non-commercial use.
What types or sharing constitute commercial use?
- the charging of a subscription fee to access content
- the charging of a tuition fee to access content
- the placement of advertising on a resource
- the requirement that one become a member of an association - the requirement that the consumer provide information, which will later be sold
- the requirement that a person purchase a device or viewer
This list could probably be extended. But the general gist is clear:
Sharing constitutes 'commercial use' if and only if conditions are placed on access to the resource in such a way that access is possible only if the sharer receives compensation for having shared the resource.
In other words, commercial use isn't actually 'use' of the resource at all, in any straightforward sense, but rather, is the enclosure of that resource, where the purpose of the enclosure is to provide some (financial) benefit to the provider.
From these considerations, the reader should be able to see plainly the basis for my advocacy of the non-commercial clause. From the perspective of the consumer, the placement of conditions on access creates a barrier to access, one that entails that the resource is no longer free.
Now recognize that opponents may say that this is not necessarily the case, that if the resource is licensed under Creative commons and yet used commercially, that there will always be some free (as in the sense above) way to access the resource.
But, there is nothing in Creative Commons that makes this the case - nothing in Creative Commons that would ensure that such a resource can be freely accessed. It could easily be that the only way to access such a resource would be to pay for it, one way or another.
Sharing, Combining and Conversion
Before dealing with the problem of enclosure, I would like to address directly the suggestion that the 'Share Alike' (SA) clause is the source of the difficulties.
To me, it is evident that, no matter what they may say, the authors of the CCLearn Report find the non-commercial (NC) clause, not the SA clause, to be the source of the problem. I realize that, as Wiley says, "the license compatibility section of the report is very clear in stating that the ShareAlike clause is the root of the license compatibility problem - not the NC clause."
However, when we get to what the authors actually recommend, they write: "Therefore, the standard terms of copyright licenses associated with OERs should permit adaptation and translation in ways that allow OERs to be combined, shared, adapted, and recombined without restriction. The license that achieves this purpose most effectively is the Creative Commons Attribution Only (CC BY) license." Why eliminate the NC clause if you don't think that NC is the problem?
In fact, the elimination of SA allows the resource not only to be distributed commercially, it allows attribution to be removed, and indeed, allows it to be distributed under any license whatsoever. It is hard to see how any license that allows such reuse could be construed as 'free' under the definition outlined above.
Let us look, for a moment, at the origin of the SA clause. As Richard Stallman and Eben Moglen write, "The 'share and share alike' or 'copyleft' aspect of the GPL is its most important functional characteristic." The reason for this is that, without the copyleft provision, a software developer could take some GPL code and incorporate it into his own commercial and proprietary product. Under extreme conditions, he could claim ownership over this code, and even litigate in order to prevent its original con-commercial usage.
It is exactly this sort of process SA restricts in the case of Creative Commons licensed content. Here's Wiley again: "By-SA can’t be recombined with By-NC-SA, neither By-SA nor By-NC-SA can be combined with the GFDL, and any public domain or CC By licensed resource remixed with By-SA, By-NC-SA, or GFDL licensed resources is forcibly converted to those terms (as part of the larger remix - the original remains unchanged, of course)."
Now let's understand what this means. It means that, if I have created a resource that I license as Non-Commercial, a person cannot combine my resource with some work of their own to produce a new resource that can be used commercially. The SA clause, in effect, prevents people from subverting the intent of the original NC condition. Remove the SA, and you may as well remove NC, because there are no restrictions on how the combined work can be used at all.
This is depicted as the compatibility problem, or even as CC Infighting. But it is, in fact, nothing of the sort.
The Share-Alike license does nothing to prohibit an NC work from being used in conjunction with a non-NC work. For example, a page of readings provided to a student could link to one of my essays, licensed under CC-NC-By-SA and one of David Wiley's, licensed under CC-By. There is nothing incompatible about the licenses, unless one wants to convert the NC content into commercial content.
In short, SA doesn't prevent you from using NC content, it only prevents you from converting it into non-NC content through some process of combining or merging.
This is (interestingly) exactly how it works in the world of software. A developer cannot merge GPL software into some proprietary application, thus producing a proprietary application. But they can use a proprietary application - Cold Fusion, say, in combination with an open source application (Apache, say) to create a new product or service. So why can't content developers work with commercial and NC content in this way? Why can't they continue to respect the original conditions of the license? There is no good reason - unless they want to enclose the original NC content. Unless they want to convert the NC content into commercial content, and restrict access to it, contravening the author's intentions.
Wiley depicts SA thus: "In this sense of copyleft’s unyielding, unapologetic, impatient, forcible conversion approach to interacting with materials that use a kinder, gentler license, we may appropriately call copyleft clauses the 'Spanish Inquisition of the open education movement.'" This is, frankly, ridiculous.
Share-Alike is not the problem. Share-Alike is what ensures that material created under the NC license remains under the NC license. And it is NC, not Share-Alike, that opponents really wish to suppress. NC, not SA, that they feel is the 'Spanish Inquisition of the open education movement'. Because, you see, they don't want open education at all - but rather, education that was formerly open, but which is now enclosed, which now may be purchased only at a price.
The first part of Wiley's objection to my position concerns the definition of 'non-commercial'. It is my hope that the discussion above has elucidated my meaning to a significant degree. While I agree that there is possibly not a widespread understanding of the term (the survey results may provide some indication) there should be no particular constraint preventing us from agreeing that this is what is meant by 'commercial' use: the enclosure of content behind some commercial barrier.
By identifying 'commercial' use as 'enclosure' I need now to address directly the second part of Wiley's argument: "I have never understood (and I really, deeply, sincerely do want to understand) Stephen’s line of argument describing how OERs licensed with, say, the CC By-SA license (lacking the NC clause), can be cordoned off by for-profit interests." Let me explain. Suppose a person, Fred, creates a resource called 'XYZ'. And suppose Fred posts XYL on his own website and licenses it CC By-SA.
Cordoning can be created very simply. A third party, Omniplex, can copy Fred's resource and place it on their own website. Omniplex then creates the cordoning by requiring that a person purchase a subscription to their website in order to view the resource. For clarity, we'll call this new instance of the resource XYZ-c.
Now the response is very clear. It may be true that XYZ-c has been cordoned by Omniplex. But so long as XYZ is available on Fred's website, anyone can access XYZ. So XYZ isn't really cordoned off, just one instance of it is. And people who access the resource, on seeing the CC By-SA license, could take the cordoned version and place it on any other website. Right?
Quite so. But we must now understand, Omniplex's biggest competition is now Fred, the original source of XYZ in the first place. Anything that can be done to ensure that users access XYZ-c, and not XYZ, will be in Omniplex's corporate interests. And, in fact, XYZ-c has an arsenal of resources at its disposal to ensure that this is the case.
This is just a partial list of the strategies that are employed by agencies in the position of Omniplex:
- a climate of litigation is created such that, only content from trusted corporations, such as Omniplex, can be 'known' to be copied legally, such that users no longer trust that they have the right to use Fred's XYZ, but trust XYZ-c. Or Fred is required to license DRM software in order to 'prove' that the resource is legitimately distributed
- legal and other overheads can force Fred's website off the air - for example, threats of legal action from Omniplex, threats of civil or criminal action on unrelated matters, lawsuits (justified or not) from anti file-sharing agencies, DOS attacks, domain squatting, and more
- SEO manipulation - Omniplex not only employs an aray of spammers to ensure premium placement for XYZ-c, these same tactics are used to push down Fred's Google rank by discrediting him; Omnicorp also has the resources, where Fred does not, to purchase search engine placement
- exclusivity of market - in the same way you'll never find free books in the bookstore, Omniplex would like to set up online markets where only XYZ-c, and never XYZ, is listed
- formal requirement - tuition required to enrol in an accredited course needed for a degree pays, in part, for instances of XYZ-c. Or, students are required to purchase XYZ-c from an 'official' marketplace (so-called to prevent 'piracy')
- proprietary platform - in some environment - Amazon Kindle, say - only XYZ-c is available; XYZ is not available, because the owners of the proprietary platform will not license Fred to place it there
Now what should be emphasized here is that this is only a partial list of tactics that Omniplex could use. What is significant here is that, once it obtains the right to offer XYZ-c, an enclosed version of XYZ, it has considerable incentive to ensure that XYZ is sunk out of view, or preferably, made difficult or impossible to legally obtain. It should be noted that there is nothing in Creative Commons, over and above the NC clause, that prevents this. As soon as content can be legally enclosed, it is as though a veritable horde or commercial providers descends on it, converting it into spam-bait, converting it into Google-ad carriers, converting it into subscription content - and each and every one of them engaged in the common interest of ensuring that the original sinks from sight and disappears.
Again, perspective is useful here. Wiley writes:
"The Open Learning pilot at BYU that will launch in January will be using the CC By-SA license. We own the copyrights on the material we will be sharing, which is why we can set the terms of the license they will be distributed under. There is nothing any corporation, entity, or individual can do to strip us of our rights, to prevent us from distributing our material, or to interfere with our provision of free educational materials to the world in any way. And any corporation, entity, or individual that would ever try to modify and then charge a fee for our materials is required by the SA clause on our materials to simultaneously freely license their derivatives under the same CC By-SA."
Quite so. BYU's rights remain completely untouched by any commercialization of its materials.
But, unless they agree to put it into a proprietary format, or to place it into an exclusive (pay) marketplace, it will not be offered on (say) Kindle or iTunes, etc. (Please note that even if these marketplaces offer free content, were they to gain anything like a strong market share the free content would disappear, just as it does in bookstores).
Unless BYU agrees to license its material to a commercial carrier, and support it with advertisements, then it will be impacted by selective content metering, the unbalanced web distribution what may result after the failure of net neutrality.
Unless BYU licenses DRM technology from ContentGuard, and unless BYU enters into legal agreements with all its authors, it runs the risk of being identified as a distributor of pirated material, or worse, those that reuse BYU's materials run the risk of being accused of piracy, as they will be unable to prove (through some licensing system) that they are using the material legally.
And that's just BYU. Its users are in an even worse position.
It's quite true that BYU currently receives good Google rankings. But this is only because Google is not competing with BYU. Companies that offer products or services that compete with Google are demoted in search rankings. Google's Knol - despite having no initial linkage or credibility - vaulted to the top of search engine rankings, well above the original locations of CC licensed materials.
Users of BYU materials would never find it under such conditions. They could only obtain the Google-supported versions. So, they would have to view the Google ads and subject their browsing to Google analytics, after being required to log in using their Google IDs.
Even worse is the plight of people who are unable to access BYUs materials in their original online format. These are people who do not have internet access, or whose government blocks BYU (and other educational providers, such as Amnesty International or UNICEF).
These people can access BYU content only through commercial cellphone services, or on commercial DVDs or in print. These are expensive - ironically, the poorest people in the world are paying the highest price. And yet, the providers of this content are lobbying the government against internet access because it is a form of unfair competition. Because it threatens their business model. Just the way commercial education and television providers took legal action to prevent the BBC from putting educational materials online. Just as ISPs - who had no intention of providing open access - took legal action against cities who wanted to provide free public wireless internet.
It is, quite frankly, beyond me how anyone can look at all comprehensively at the state of commercialism and the internet and expect any other outcome. Anyone who expects any other outcome is living in fantasyland. There is no domain, where commercial activity was allowed without restraint, where the commercializers did not take over and ultimately wreck the domain.
Grounds For Agreement
I suggested at the top of this article that they may be room for some sort of rapprochement between David Wiley and myself on this issue. Because, after all, we in fact are on the same side, both working toward the development and distribution of free (open) content and educational resources.
Let me begin by being clear about my position on commercialism.
- first, it doesn't bother me at all is a person or an organization makes money by adding value to work that I (or anyone else has created. This is, indeed, the foundation of the productive economy, the idea that, by producing value, a person is rewarded
- but second, it does bother me if a person or organization makes money by subtracting value from work that I (or anyone else) has produced, by limiting access to it, by making it more difficult to obtain, by casting doubt or legal concerns about its use
People should not be rewarded for making free things more scarce. Especially when society as a whole benefits so much from their abundance, the creation of artificial scarcities, whether through law, litigation or technology, is a fundamentally destructive and hurtful activity. People should be punished for creating scarcities, not rewarded, as such is the source of illusive, and ill-gotten wealth, founded on misery and deprivation rather than growth and development.
The grounds for agreement are thus these: if we could build these into the license, then the non-commercial clause would not be needed.
It's important to note that these conditions are already in the GPL. Not only is the principle of copyleft embedded in the license, so also are the mechanisms that make it impossible to prevent a person from using the software. By ensuring that the software user always has access to the source, and that no subsequent development can hide the source, there is no way to create an exclusive commercial marketplace.
The GNU Free Documentation License (GFDL) attempts to recognize this. "If you publish or distribute Opaque copies of the Document numbering more than 100, you must either include a machine-readable Transparent copy along with each Opaque copy, or state in or with each Opaque copy a computer-network location from which the general network-using public has access to download using public-standard network protocols a complete Transparent copy of the Document, free of added material."
In other words, the GFDL attempts to create a condition that would prevent enclosure by at leas some of the mechanisms mentioned above. By requiring an "opaque copy" the GFDL prohibits the wholesale and exclusive copying in a proprietary format. People using "public-standard network protocols" must be able to access the content.
David Wiley's Open Publication License also contains text to address some of the issues raised here. For example, "Mere aggregation of Open Publication works or a portion of an Open Publication work with other works or programs on the same media shall not cause this license to apply to those other works." Mere aggregation does not cause conversion. And additionally, the license requires that, if a modification is made, that the location of the original be identified in the modification.
And the licensing condition he adopts for Flat World Knowledge - the CC By-NC-SA Plus license - also recognizes these concerns. The 'plus' is a commercialization license, with conditions: "it will grant blanket permissions for anyone and everyone to make Commercial Use of FWK-published textbook materials in the context of the FWK Marketplace." Within the FWK Marketplace. Why? Because the biggest threat to FWK is some commercial entity coming along, taking all their content, and creating a competing, closed and completely commercial marketplace.
So I think that Wiley recognizes the grounds I have for concern. Where we have a disagreement is regarding the best way to address them. I have - for lack of anything other than a blunt instrument - opted for Creative Commons Non-Commercial. Wiley has both written his own license and worked with an amended version of Creative Commons."
"We want, in common, to say, I think, that by identifying our content as 'free content', we want:
- to ensure that any person is able to access this content - to read it, or view it, or play it, as the case may be - which means that free content itself must be in some sort of public-standard network protocols
- to have the source of this content - a link back to the original which is always available as an alternative to the shared, or copied version
- to allow people to share this content, to make it available in different formats and in different places, but not to create conditions or an environment where access to the original content is impaired, or access to the extant content is not free, not open
- to allow people to build upon this content, to improve it, and to share these improvements under whatever license they wish, provided that such improvements provide direct, accessible and unimpeded access to the original content
Now these conditions are very general, and are not intended to be the 'letter of the law'. They are expressed as intentions, to be interpreted, not as principles to be defined. And we can, from the discussion above, discern that intent: it is our desire that content not be enclosed, but rather, be set free.
I believe that something like this is the common ground on which David Wiley's and my views of open content rest, and hope that some such statement is the foundation on which we can progress in the future." (http://www.downes.ca/files/FreeLearning.pdf)