Posts Tagged 'open access'

Why Non Commercial Licences are Bad

Brendan Scott, September 2008

In this post I work through one example in an attempt to illustrate why purpose based restrictions in “open” licences are a Bad Thing(tm).  Let me state here that “bad” is meant as “less than optimal”, not bad in any absolute sense.  An otherwise open licence with a non commercial restriction will generally be “good” compared with closed licences for example.

Example – The Cure for Cancer

Let’s say that one night (while doing the washing up) you stumble across the cure for cancer (the Cure).   Being a relatively normal human being, you want to share this with the world to save people.  Let’s say you get a monopoly from the State over the Cure and you now need to think about how to licence it.  If  you are a mercenary type, you would start charging money for the use of the Cure.  All else being equal, the imposition of a licence fee will exclude some part of the community from being able to access the Cure (ie those people who can’t afford the fee).

Let’s assume instead that you chose to licence it at no cost, but on some licence terms.  In this scenario there may still be people who can’t afford a cure (if, for example there are other costs in delivering the Cure to them).   The people who will receive the Cure in this scenario are those who are determined by the licence terms.  If you were to chose licence terms which prohibited commercial use then you would limit the people who get the Cure.   I can’t see any reason why someone would do this, except to be able to charge extra to that group of people who would otherwise gain access through commercial means.* Anyone who can see another a reason please add a comment below!

“NC” does not Promote Philanthropy

You might not be interested so much in licensing for a fee, but might want to encourage other people to provide the Cure to others for free.  Alternatively, you might not want to see third parties profiting from something that you have decided to give away for free.  By adding a non-commercial restriction in these cases you would exclude those people who might provide the Cure to others in the course of commerce.  This would, for example, cover medical practitioners, the vast majority of whom charge for their services.  A non commercial restriction would eliminate from the distribution chain exactly the people who would be most important to delivery of the Cure to the general citizenry.

Any delivery of the Cure will have a cost.  At the very least there is an opportunity cost involved when someone chooses to provide the Cure to others in preference to doing something else.  If a person is prohibited from being compensated for their cost, then their provision of the Cure to others will actually cause them a loss.**  Regardless of whether the cost is large or small, this will restrict the scope of the people to whom the Cure will be provided (some people may shoulder a small cost for a small number of recipients, but as the number of recipients grows even small per unit costs will become unsustainable).  A non-commercial restriction doesn’t encourage others to participate in philanthropy. Rather, it simply limits the number of people who will be willing to distribute the Cure.  By trying to prevent third party intermediaries from benefiting from the Cure, a non-commercial licence throws the baby out with the bathwater.  The only thing that can be said for it is that it is self defeating.

Somewhat counter-intuitively, the way you prevent third parties from gaining an unconscionable profit from your invention is to adopt something along the lines of an open source licence.  An open source licence would permit commercial entities to sell the Cure (and probably make a profit in the short term) but the licence terms preserve a free market for the Cure.  Over time other commercial entities would also begin selling it.  Economic theory says that, in these conditions, the price that they charge will (over time) be the cost to them of selling the Cure.  This is the ideal solution – distributors have the opportunity to make a profit in the short term so they start distributing the Cure.  Over time however, price pressure from other distributors forces their prices down.  In the long run distributors are still able to cover their costs and the number of distributors is maximised.  Indeed, in this case the only time a distributor can charge above their cost is where they are providing some added value (until others start competing on that value add).  In this case the choice of a share-alike style licence will work to preserve the same market dynamics in respect of the incremental improvements.

It is also important to note that those involved in commercial activity are the ones most likely to be able to identify what needs improvement (through feedback from their customers) and to pay for incremental improvements to be made.  By excluding them you not only have a direct impact on distribution and implementation of the Cure, you also destroy a good part of future innovations based on it.

Reasoning Applies to All Purpose Based Restrictions

I have focussed on non-commercial restrictions as they seem to be inordinately popular and also seem to be the most wide reaching in their effect.  However, the same reasoning applies to any purpose based restriction.  The effects of the restriction will be determined by the restriction itself.  For example, a prohibition on “use for celebrating the 4th of July in the year 2075” may have little practical impact on uptake in the short term (but may have some unexpected consequences around July 4, 2075).


I chose the title of the post, not because non-commercial licences are absolutely bad, but rather to draw attention to the fact that licences with purpose based restrictions will generally restrict the scope of the distribution of the content licensed.  I have chosen the example of a cure for cancer, because it is something which ought unequivocally be distributed among as many people as possible.   The two reasons I can see for a purpose based restriction are:

(a) to charge money for the use of the thing; and

(b) because the person has an ethical objection to the purpose (raised in note *)

I do not want to suggest that these reasons are illegitimate.  I suspect, however, that (a) is a pipe dream in the vast majority of cases.  The publishing industry promotes the idea to authors that untold riches await them if they are as covetous as possible of their “intellectual property”.  The reality though is that for every successful creator there are thousands (if not millions) of unsuccessful ones.  The probabilities are very much against any particular person (authors’ associations the world over consistently report below average income for authors).  In coveting their “intellectual property” authors consign themselves to a lifetime of anonymity – one of the worst punishments for a voice which wants to be heard.  Moreover, they miss the opportunity to make the world a better place by giving others access to their creativity.  This ideology which inspires non-commercial restrictions is an ideology in which everyone (except the owners of a distribution channel) loses.


* Actually, there is one other reason, which is that you have some ethical objection to the particular purpose.  In the case of commercial purposes most people would have trouble living in practice by such an objection (eg they would need to live in a gift economy).

** In practice a distributor will produce a number of units of the Cure in advance and must take the risk of not being able to sell them all.  In this case, even limiting them to their cost of sale will cause them a loss – if not all of the inventory is sold.

Houghton and Sheehan on Economic Impact of Open Access

In July 2006 John Houghton and Peter Sheehan published a paper on The Economic Impact of Enhanced Access to Research Findings. Apparently they have built on this work in subsequent papers. The paper explains how the impact of access to information is factored into standard growth models for the economy. The authors posit two parameters (phi – representing the portion of research and development which ends up being useful – and epsilon – representing that knowledge may not be perfectly accessible). They then go on to identify the consequences of a change in these two factors as might occur from a transition to open access. On their analysis the rate of return on research and development increases by a percentage equal to the percentage increase in the efficiency and accessibility of the knowledge (ie in the parameters phi and epsilon). They state:

“… the results… imply that, if a move to open access has a significant beneficial impact on either or both the accessibility or efficiency of R&D, then the benefits of open access will be high also. Assuming, for example, that a move towards open access increased access and efficiency by 5% and that the social rate of return to GERD was 50%, then if there had been open access to all OECD research circa 2003 it would have increased the social returns to R&D by some USD 36 billion. These are recurring gains from the effect on one year’s R&D. Hence, assuming that the change is permanent, they can be converted to growth rate effects.”

The issue the paper does not address however is what the values of these parameters are in practice, and what how a move to open access will impact them. The put forward some hypotheticals in a country by country table.

They Really Mean Open Access+

However, the assumptions they have made is that the knowledge which is accessible may also be used. This is a stronger assumption than is made by some parts of the “open access” movement which impose differing degrees of purpose based and other restrictions – commentary example 1 example 2.

Blog Stats

  • 224,042 hits

OSWALD Newsletter

If you would like to receive OSWALD, a weekly open source news digest please send an email to oswald (with the subject "subscribe") at