- Image via Wikipedia
Richard Stallman and Richard Grant, two people who I wouldn’t ever have expected to group together except based on their first name, have recently published articles that have made me think about what we mean when we talk about “Open†stuff. In many ways this is a return right to the beginning of this blog, which started with a post in which I tried to define my terms as I understood them at the time.
In Stallman’s piece he argues that “open†as in “open source†is misleading because it sounds limiting. It makes it sound as though the only thing that matters is having access to the source code. He dismisses the various careful definitions of open as specialist pleading, definitions that only the few are aware of, and that using them will confuse most others. He is of course right, no matter how carefully we define open it is such a commonly used word and so open to interpretation itself that there will always be ambiguity.
Many efforts have been made in various communities to find new and more precise terms, “gratis†and “libreâ€, “green†vs “goldâ€, but these never stick, largely because the word “open†captures the imagination in a way more precise terms do not, and largely because these terms capture the issues that divide us, rather than those that unite us.
So Stallman has a point but he then goes on to argue that “free†does not suffer from the same issues because it does capture an important aspect of Free Software. I can’t agree here because it seems clear to me we have exactly the same confusions. “Free as in beerâ€, “free as in free speech†capture exactly the same types of confusion, and indeed exactly the same kind of issues as all the various subdefinitions of open. But worse than that it implies these things are in fact free, that they don’t actually cost anything to produce.
In Richard Grant’s post he argues against the idea that the Faculty of 1000, a site that provides expert assessment of researcher papers by a hand picked group of academics, “should be open accessâ€. His argument is largely pragmatic, that running the service costs money. That money needs to be recovered in some way or there would be no service. Now we can argue that there might be more efficient and cheaper ways of providing that service but it is never going to be free. The production of the scholarly literature is likewise never going to be free. Archival, storage, people keeping the system running, just the electricity, these all cost money and that has to come from somewhere.
It may surprise overseas readers but access to many British museums is free to anyone. The British Museum, National Portrait Gallery and others are all free to enter. That they are not “free†in terms of cost is obvious. This access is subsidised by the taxpayer. The original collection of the British Museum was in fact donated to the British people, but in taking that collection on the government was accepting a liability. One that continues to run into millions of pounds a year, just to stop the collection from falling apart, let alone enhancing, displaying it, or researching it.
The decision to make these museums openly accessible is in part ideological, but it can also be framed as a pragmatic decision. Given the enormous monetary investment there is a large value in subsidising free access to maximise the social benefits that universal access can provide. Charging for access would almost certainly increase income, or at least decrease costs, but there would be significant opportunity cost in terms of social return on investment by barring access.
Those of us who argue for Open Access to the scholarly literature or for Open Data, Process, Materials or whatever need to be careful that we don’t pretend this comes free. We also need to educate ourselves more about the costs. Writing costs money, peer review costs money, editing the formats, running the web servers, and providing archival services costs money. And it costs money whether it is done by publishers operating a subscription or  author-pays business models, or by institutional or domain repositories. We can argue for Open Access approaches on economic efficiency grounds, and we can argue for it based on maximizing social return on investment, essentially that for a small additional investment, over and above the very large existing investment in research, significant potential social benefits will arise.
Open Access scholarly literature is free like the British Museum or a national monument like the Lincoln Memorial is free. We should strive to bring costs down as far as we can. We should defend the added value of investing in providing free access to view and use content. But we should never pretend that those costs don’t exist.
This has a bit of homo stramineus flavor to it, Cameron. I can’t disagree with it, but I also want to see examples of people claiming that any of this IS free. This is a brush often used to tar OA, but in my opinion usually falsely.
Well that’s fair enough. So firstly the motivation to write this in the
first place was largely one of _those_ conversations where someone says
“…why can’t we just put it all on the web, that doesn’t cost anything…”
and as the Open Access meme gets around I’m hearing this a lot more as
people discover what the average article processing charge for PLoS/BMC are.
This is also fundamentally where I part paths with Harnard, the impression
he seems to give that we can somehow defer the question of costs and it will
all work out in the end as publishers respond by somehow just reducing them
in response to the success of repositories. I’ve heard a couple of “…well
we can’t just put it in the IR, that doesn’t cost anything…” comments
recently as well.
Finally at some level it’s implicit in Stallman’s piece that somehow open
source (or free) software is free, which I’d disagree with as well. It
represents massive investment, from offices provided at MIT, through
people’s contribution of time to significant commercial investment. The lack
of appreciation for the importance of heavy commercial (or institutional)
investment in most big OS software project irks me a bit.
So I guess this is a question of taking both ends against the middle. We
need both to make sure that people understand the costs in our advocacy work
and counter the kinds of arguments that we get from toll-access publishers
that we’re asking for something for nothing by providing the economic
efficient and business model arguments.
No argument here. :)
As editor of an Open Access journal it was only too painfully obvious how unfree it was. We were lucky in having some resource (somewhere between one and two thirds of a person) to do the real hard editing, but we needed work done on the technology that we never quite managed. I know there are some journals that are done entirely from a back bedroom, but the amount of dedication required is immense.
This is in part what the Blue Ribbon Task Force on Sustainable Digital Preservation and Access was about. No magic bullets of course, but the aim seemed to be to get high community but in. Then you can pull tricks like when NSF threatened to defund ArXiV and the community got up in arms and won enough breathing space for the Cornell solution. That’s now in another chapter as Cornell asks (and apparently gets) donations from libraries as users. Or as another example, when the previous government tried to force museums to charge entry (;-)!
But so often in BRTF we kept coming back to the idea of “government” funding things for the taxpayer community. Always dodgy, the fragility of this idea is being shown by the “policy changes” being introduced by our government now. I would prefer to find a more robust funding model for free than government!
It seems part of the fundamental problem is that we’re trying to fund
something in perpetuity with a single charge business model. An ongoing
subscription model at least has the advantage that it in principle provides
funding into the future. But both that and institutional archiving presume
that the institution (or company) survives.
Maybe we do just need to be realistic about time frames and attempt to set
out costs based on e.g. 80% survival over 50 years. Not a popular view I
imagine with the digital preservation community but does it have any value
as a straw man at least against which to compare other approaches?
The other thing that we desperately need to get a better handle on is a more
flexible model of costs. The RIN study and model is a good start but until
we start really focussing on where costs can be reduced on the production
and distribution side (which really ought be much lower than they used to
be) then we can’t think about shifting those costs over into digital
preservation strategies. We will be doing “more with less” so we need to
figure out where that less is going to come from, or at least whether its
realistic.
Distinguishing the types of free seems useful. Another example to consider is free like the Metropolitan Museum of Art – which is a privately run not for profit organization that receives certain government subsidies but is funded largely by a combination of private gifts and purely optional suggested donations at the door. The national museums and monuments are more largely funded by government grants or direct subsidy.
Nooooooooo! Please, not free as the British Museum for at least three reasons;
1. The British museum is full of pieces stolen of other civilizations and for people of these regions to see the pieces they have to travel to London
2. There is some policy regarding how the museum collection is lended to other places, in order to people who cannot travel to london to see the pieces. And this policy is not public, is not open in any mean you can attach to it.
3. Last, but not least, to enter the UK you must be accepted by immigration staff.
I got your point, but your example is quite bad for people from outside UK.
Manoel
Absolutely agree with what you’re saying here. But also I think you’re
objections also have parallels in the open foo world. So let me take a
devils advocate position without necessarily saying I completely agree with
these positions.
1. The counter argument that is often used is that without that theft and
subsequent preservation most of these objects would have been lost (Not
saying I agree with that, just that its an argument that gets used!).
Equally in the digital world there have commonly been problems where the
only way to effectively preserve something is to “steal” it. Supplementary
data in journals comes to mind where the record of preservation is poor.
Arguably the internet archive is a mass copyright violator. At least online
you’re not taking the object away from it initial place.
2. Actually I suspect in grand British tradition the policy is not even
written down, but that’s another issue. This is not disimilar to the complex
and often misguided use of licences and terms of service online. In many
cases with the scholarly literature it is actually impossible to figure out
what you are allowed to do with something. Policy and legal situation often
don’t match up and technical means (which are often not public) are
sometimes used to enforce policies that are not supported by legal
situation.
3. Yes, although entering the UK is easier than almost any other country
I’ve tried to get into (as a non-UK national). But this is not disimilar to
the online world. You need access to a computer, you need decent access to
the web/internet. As the Budapest declaration says “open access…is
unrestricted…aside from those requirements indivisible from access to the
open internet…” Anyone in Cuba, large parts of Africa, South East Asia, or
in places where the internet is heavily censored such as China or
potentially Australia simply can’t get access no matter what we do. That
access issue is to me a separate one from the one of access if you are on
the internet. It’s a big problem but its not one I’m game to take on at the
moment.
I do take your point, and I originally thought about using the British
Library as my example (although I’m not sure its any better) but that seemed
a bit obvious.