English | Français | Português | Español | 中文 | العربية | Русский
WSIS Knowledge Communities
Remember me | Lost password

You are viewing this page as a non-logged user. Login to see the full content and functionality. Do not have an account? Click here to register to become a member of the platform.

Supporting the EFA and MDGs with Open Educational Resources

Share |

Should OER favour commercial use?

Pro

The question of this debate, “Should OER favour commercial use?”, can be answered many ways. Rather than make a moral or ethical argument, I will provide an opening statement which problematizes the mechanism by which commercial uses are disallowed. In other words, I will answer “Yes” on technical grounds. The moral argument can be nuanced and difficult to win. The technical argument is straightforward and unassailable.

I believe, for a variety of reasons, that Creative Commons licenses that include the Non-Commercial clause should not be used. There is, however, one important family of cases where NC-bearing licenses are completely appropriate. In fact, this family of cases is why the NC clause exists. Some history is in order to substantiate this claim.

The first open source-style license for educational materials and other non-software content was the Open Content License, released July 14, 1998. It was a simple adaptation of the GPL that I made while I was a PhD student at BYU. The OCL removed software-specific terminology from the GPL and tried to make the license apply to a broader range of creative works. After a few months, however, I became concerned because adoption of the license wasn’t great. I was particularly disappointed that there wasn’t a single research article, textbook, or other bit of traditional educational material (that I could find) licensed under the OCL.

Early in 1999 I reached out to Eric Raymond, Tim O’Reilly, and others for help thinking through how to solve the adoption problem. I asked Tim specifically, ‘You’re a commercial publisher – what would it take for you to publish a book under an open license?’ Tim responded, ‘After I spend tens of thousands of dollars paying an author, paying for editorial and design, etc., the next publisher down the road can’t be allowed to reprint my book and sell it for $10 less than me. That would put me out of business.’

On June 8, 1999 I released the Open Publication License. The OPL was not a minor modification of the GPL. The OPL was truly innovative in that, in addition to requiring citation of the original author as source, it contained two license options that authors could choose to invoke or not invoke. The first option restricted users’ abilities to creative derivative works. The second option disallowed “any publication of this work or derivative works in whole or in part in standard (paper) book form for commercial purposes.”

Adoption was much better with the new license. O’Reilly published several books using the OPL, including Eric Raymond’s seminal collection of articles, The Cathedral and the Bazaar. RedHat published the documentation for their products under the OPL up until 2010. In 2000 I published my first book using the OPL, free online and pay for print. Both O’Reilly and my publisher used the Non-Commercial option of the OPL on the books they published openly.

Copyright is supposedly a necessary evil – it creates a monopoly on who can make copies (which is bad) in order to provide incentives to authors to produce creative works (which is good). The Non-Commercial clause is a similar evil – it creates a monopoly on who can sell copies (which is bad) in order to provide incentives for individuals and organizations that make a living from their creative works to share them openly (which is good). I recognized early on that without this protection mechanism there would be far fewer open educational resources in the world.

In the opening paragraph of this opening statement I said that there exists a family of cases where NC-bearing licenses are completely appropriate. This is that family of cases – cases where the work to be openly licensed was created by individuals or organizations that make a living from their creative works. Personally, I can’t think of other circumstances where using the NC clause is appropriate. Let me explain further.

The student of open content licensing will recognize the OPL structure (mandatory attribution with options addressing derivative works and commercial use) is the same structure that Creative Commons employed when it launched three years later in 2002. (To be more exact, in version 1.0 of the Creative Commons licenses Attribution was an option, too. It wasn’t until version 2.0 of the CC licenses (May 24, 2004) that attribution became mandatory on every license. So, the OPL structure precedes the current CC structure by about five years.)

When Creative Commons published its first licenses in 2002, I was relieved. I’m not a lawyer, and who knows how the OCL or OPL would have held up in court? Probably not very well. I was happy to post a big message on opencontent.org declaring that everyone should stop using the OCL and OPL and should begin using the CC licenses. After all, they were lawyers, and had taken the OPL structure and created a far better legal instrument.

However, in retrospect Creative Commons made a mistake in generalizing the Non-Commercial clause of a publication license – defined very narrowly in terms of print publication – into a general purpose Non-Commercial clause in a general purpose license. The problem with this generalization is that, once removed from the specific use case for which it was intended, “Non-Commercial” no longer has a meaning.

In 2007 I was amused to find the following contradiction. Creative Commons posted a “Proposed Best Practice Guidelines To Clarify The Meaning Of ‘Noncommercial’ In The Creative Commons Licenses” which included the following logic:

(1) Is the person making use of an NC-licensed work an “allowable NC user under the noncommercial license condition?

Allowable NC users are:

(a) an Individual (b) a Nonprofit educational institution/library, (c) a Nonprofit organization as defined under US or equivalent law [1], (together with (1) and (2) “allowable NC users) (d) A commercial copy shop, ISP, search engine, content aggregator, blog aggregator site or similar service provider who, in the course of providing a service at the direction of the allowable NC user, may exercise a right licensed under the Creative Commons license.

(i) No. License violation – this is not a noncommercial use.

(ii) Yes. Continue to Question B.

This sounded slightly diametrically opposed to MIT OpenCourseWare’s official “Interpretation of ‘Non-commercial’”:

Determination of commercial vs. non-commercial purpose is based on the use, not the user. Materials may be used by individuals, institutions, governments, corporations, or other business whether for-profit or non-profit so long as the use itself is not a commercialization of the materials or a use that is directly intended to generate sales or profit.

When I pointed out this contradiction, guess who blinked? Creative Commons pulled the well-researched draft guidelines off their website. Both unable and unwilling to define the terms of its own license clause, in 2009 Creative Commons undertook a survey of the community to try to understand what other people and institutions thought Non-Commercial meant. The results were predictable. From the Executive Summary:

Perceptions of the many use cases studied suggest that with the exception of uses that earn users money or involve advertising… there is more uncertainty than clarity around whether specific uses of online content are commercial or noncommercial.

In his book the Wealth of Networks, Benkler characterizes the Creative Commons project as follows:

The Creative Commons is an initiative to develop a series of licenses that allow individuals who create information, knowledge, and culture to attach simple licenses that define what others may, or may not, do with their work. The innovation represented by these licenses relative to the background copyright system is that they make it trivial for people to give others permission to use their creations.

As Yochai says, the innovation is supposed to be that the licenses make it trivial for you to define what others may, or may not, do with a work. Of course, the case with the NC clause is that it does not make it trivial to define what others may or may not do. It only makes it trivial to appear that you are doing so. Adam Bosworth once recounted asking Lessig about:

the provision in many Creative Commons licenses that indicates content may not be used for ‘Commercial Use’. I asked, what is Commercial Use? Does reposting to a blog that has ads violate the copyright license? Larry Lessig’s answer was basically, “I don’t know”.

If a lawyer of the caliber of Larry Lessig does not know what NC means, how is your run-of-the-mill user supposed to understand what s/he may or may not do with a NC-licensed work? The only answer is, of course, that they must read the full Terms of Use on the site. The lack of a clear, official definition of NC from Creative Commons transports us backwards in time to a day when site owners were required to write their own terms of use (in this case, their own definition of NC), and when site users were required to read them in detail. In other words, the lack of a clear, official definition of NC destroys the very innovation that Creative Commons licenses are trying to provide.

Should OER favour commercial use? Yes, in part because there is no trivial mechanism for excluding commercial uses.

And what about the family of cases in which using an NC clause is appropriate? Unfortunately, these individuals and organizations have no other choice but the nontrivial one - using the NC clause and authoring their own accompanying definition, which users must read and understand before they can safely exercise any of the rights granted in the license.

(Another interesting implication of the NC definition problem is this: Take the case of two OERs using the BY-NC-SA license. If each author has created a unique definition of Non-Commercial, are the two OERs remix-compatible under the terms of the SA clause in the BY-NC-SA license? I believe the answer is “no,” but leave the proof as an exercise for the reader.)

Comments

Comments Order: Limit: Auto Load:

I think it's unfortuante that more OER proponents don't see commercial entities not as a threat, but instead as a useful vector for transmitting awareness about the materials we're developing.  In the long run, we have as much to gain from the relationship as they.  I elaborate here.

-=Steve=-

Steve Foerster 1102 days ago

In response to Ahrash and Stephen's clarification:

I believe the intention with the formulation, as Stephen has articulated, was to imply preference or inclination toward.

As an open debate about things open -- I have indicated to both debaters that they are free to post comments and join discussions on the floor.

 

Wayne Mackintosh 1102 days ago

Responding to Ahrash:

> Among the many different definitions of "favour" are these two contrasting meanings: "to prefer" versus "to accommodate."

There isn't any widely used sense of 'favour' that means merely 'to accommodate'. The sense of 'favour' used here is one that, as the word suggest, implies a preference or inclination toward.

Stephen Downes 1103 days ago

Responding to Jacky Hood, who says:

> Force is the method favored by those that say education is a 'right'.

> OER should be a commercial endeavor

It would be interesting to see how commercial endeavours functioned without some sort of property rights that protect the ownership of those who have something to sell. If the owner is not able to protect their goods by force, either directly or through a proxy (usually government) then it is not possible to require a commercial transaction; the other person just takes what they need.

It is very misrepresentative to suggest that those people working for rights are engaged in some sort of enterprise requiring force while those engaged in trade and commerce are not. My observation is that the police and security guards are generally protecting those who *have*, and not those who *need*.

Stephen Downes 1103 days ago

I am also interested in getting clarification on the specific meaning of the word "favour" in this debate. Among the many different definitions of "favour" are these two contrasting meanings: "to prefer" versus "to accommodate." These different meanings have very different implications for the purpose of this debate, as the former definition implies that OER are created with an intention and preference for commercialization, whereas the latter implies that OER are created with no prejudices regarding desirable forms of future use, whether commercial or non-commercial.

Can someone clarify which meaning we are debating here?

Ahrash 1103 days ago

There are only four ways in which people can share the services and objects they create:

* by trade (i.e., commerce)
* by charity of the producer
* by third party subsidization (e.g. advertising, grants, donations)
* by force (theft, slavery, confiscation, taxation)

Trade starts as barter but soon a medium of exchange emerges. At that point, one party buys (receives the services or goods) and the other sells (receives the exchange medium). The medium is usually currency but it can be gold, diamonds, cigarettes, candy, carbon credits, frequent flier miles, coupons, or anything that has recognized value and lasts for awhile.

Both parties profit in barter and currency exchanges. They each value what they receive more than what they give up. The person receiving currency takes the greater risk because the value of the currency could go down before it can be used to purchase goods or services.

Even if the marginal cost of the next instance of a good or service is very low: e.g. pharmaceuticals and digital materials, it is the initial creation of the good or service that must be rewarded with other goods/services or a medium of exchange.

Third party models work in some cases but they remove some of the decision-making from the buyer and seller.

Force is the method favored by those that say education is a 'right'. Rights are restraints on others. For example, a person's right to life, liberty, and property means that others must refrain from violence, imprisonment, and theft. To say that someone has a 'right' to goods or services means someone else has an obligation to provide those goods or services without compensation, i.e., the producer (or third party payer) has his/her rights violated.

The best business model for open educational resources is very-low-price, very-high-volume. Digital goods have a global market and so volume can be very high. With very high volume, prices can be quite low, on the order of pennies.

The difficulty is finding a payment method that does not cost more than the item being sold. Here we can turn to the telephone companies. A telephone operating company executive once told me, "We are not a phone company; we are a billing company." The phone companies have reduced the cost of billing to a fraction of a cent. There are many other benefits to delivering OER on phones, including the fact that people expect Internet content to be free but expect to pay for content on phones.

The very best reason that OER should be a commercial endeavor is because both producers and consumers of educational resources deserve respect. Producers should not be seen as people whose creations are not good enough to receive something in return. OER consumers should not be objects of pity and charity.

So what does it mean for the producer to both receive compensation for his/her output and for that output to be 'open'? The seller is allowing non-traditional uses of the output, e.g. making copies, remixing, etc.

In a sense, using OER is like renting a car. Some rental car companies say 'you cannot go out of the country; you cannot drive off-road, etc." An 'open' rental company puts fewer restrictions on the renter.

My husband and I once bought a house with a covenant that said we would never create alcoholic drinks on the premises.

A producer of OER certainly has the right to refuse commercial use of the output. However, the output's spread will be very limited. Commerce and trade cause services and goods to keep moving. Restrictions stop that movement. A house with covenants will be harder to rent out or sell. Instructors and students will reject OER with restrictions (NC, ND, SA) in favor of those with CC BY licenses.

Regards,
Jacky Hood
Director, College Open Textbooks
http://collegeopentextbooks.org

hoodjackylene@fhda.edu 650 949-7091

Jacky Hood 1103 days ago

Great post/argument, David. I would be more bold, however, in making the claim that embracing derivative commercial uses of OER is a _good_ thing in itself, and not simply a position we should take based on the legal complexities of defining/enforcing an NC license.  

 

Ed Bice 1103 days ago

In casting a vote of 'no', I am troubled by your semantics -  the word 'favour' would seem to suggest that the commercialising of open educational resources should be promoted 'over' another model of use.  Is a commercial use model really the 'favourite' model in the realm of OER development?  I would say 'NO'; it goes against the spirit of every open access model for education I've been involved in since the advent of internet and more importantly, that of wiki media!   

Not that I object to the recovery of cost in any model of use  - but that's not necessarily commercial, that's plain common sense. However,  the moment this model adds an insidious 'profit' margin to pay for someone's new toyota landcruiser under the veil of not-for-profit cost 'recovery' then it's not up my street!


Joris Komen 1103 days ago

My first perception is that open knowledge can be used to build a livelihood that is suitable for a civilized society. But fortune hunting using the open knowledge will make it cease to be open.

Anil Prasad 1103 days ago

More ...