FINO – Free is Not Open

Bill Hooker of Open Reading Frame has yet again and very clearly expounded the difference between Free and Open.
FINO = Free is Not Open
What follows may look like the same old arguments. It isn’t! The difference is that an increasing number of publishers are publicly and clearly promoting the value of Openness and clearly distinguish it from Freedom. THEY are quoting the Budapest and other declarations of Openness. If you don’t understand the importance, please read this carefully.
Bill:

Once more unto the breach, dear friends, once more: the dreaded Free Is Not Open argument rears its ugly head again. I’ve made my position (indeed free != Open, and the distinction matters) clear elsewhere, and was gratified recently to find PMR agreeing; now it seems that the Open Medicine editorial team takes the same position:

The Canadian Medical Association Journal (CMAJ) has just published:

• Stanbrook et al. Congratulations to our colleagues at Open Medicine [commentary]. CMAJ July 3, 2007; 177.

Here is our response:
Although the endorsement by CMAJ‘s editors of open access medical publishing is welcome, we would like to take this opportunity to clarify several points raised in their commentary.1 First, there is an important distinction between open versus free-access publication. Open Medicine has not only adopted the principle of free access, that is, making content fully available online, but endorses the definition of open access publication drafted by the Bethesda Meeting on Open Access Publishing. This definition stipulates that the copyright holder grants to all users a free, irrevocable, worldwide, perpetual right of access to, and a license to copy, use, distribute, transmit and display the work publicly and to make and distribute works derived from the original work, in any digital medium for any responsible purpose, subject to proper attribution of authorship. Given that CMAJ holds copyright and charges reprint and permission fees, it is not in fact an open access journal.

PMR: The last phrase is exactly the point “it is not in fact an open access journal.” The important thing is that this is a publisher making it crystal clear what the distinction is.

In comparison, Open Medicine does not assume the copyright of our authors’ work. We believe that it is only fair and just that authors retain the ownership of their work; as such, Open Medicine does not charge reprint or permission fees, and our work is available for reproduction for educational and teaching purposes without copyright limitations or charges. We use a Creative Commons Copyright License that also ensures derivative works are available through an open access forum. It is through this creative and unlimited use of published material, with due attribution, that we believe scientific discourse can flourish. This truly open access forum also has a contribution to make to a journal’s integrity, independence, and freedom. […]

Chris Surridge of PLoS also agrees, and supplies an excellent analogy:

Free Access to scientific research is great, and all publishers who make their content free to read should be praised for doing so. But this is not Open Access. It is like giving a child a Lego car and telling them that they can look at it, perhaps touch it, but certainly not take it apart and make an aeroplane from it. The full potential of the work cannot be realised.

Where the OM team refer to Bethesda, Chris links to Berlin and goes on to enumerate

…the four unmistakable marks by which you may know, wheresoever you go, the warranted genuine Open Access publication:1. Content is made freely and immediately accessible to all.
This basically means that you can get it on the internet without paying anything in addition to what it costs you to access the internet.
2. Authors retain the rights of attribution.
So the work is the authors [‘ property]. The author doesn’t sign over the copyright to the publisher or anyone else. Rather the author allows the publisher to publish the work under licence. A licence which also ensures that:
3. Content can be distributed and reused without restriction.
So I or anyone else can take Open Access content and use it, in whole or in part, for any purpose including purposes that have not yet been dreamt of as long as I don’t infringe the Authors rights of attribution.
4. Papers are deposited in a public online archive such as PubMed Central.
This ensures, as best as anyone can, that the above three conditions continue to apply to the Open Access content in perpetuity.

It’s been my contention that in the absence of explicit, conspicous and machine-readable Open licensing, condition 3 is violated because in this litigious age, the conscientious and the risk-averse will not download and derive without explicit permission. I got “explicit and conspicious” from Peter Suber:

The newer definitions [of OA] recognize one further element: an explicit and conspicuous label that an open-access work is open access. Readers should be told when a work is free of price and permission barriers. They might be reading a copy forwarded from a friend and not know whether the publisher would like to charge for access. They might want to forward a copy to a friend and not know whether this kind of redistribution is permitted. When an article has no label, then conscientious users will seek permission for any copying that exceeds fair use. But this kind of delay and detour, with non-use as the consequence of a non-answer, are just the kinds of obstacles that open access seeks to eliminate. A good label will save users time and grief, prevent conscientious users from erring on the side of non-use, and eliminate a frustration that might nudge conscientious users into becoming less conscientious.

and “machine-readable” from Peter Murray-Rust:

For me, if my robots cannot read the articles then as a human I have no interest at all in reading the “fulltext”.

Peter MR is not saying that free access for humans is useless, but that to realize the full potential of text- and data-mining, OA materials need to be machine-readable, which includes letting the machines know what they are allowed to have.

I must confess that finding my thoughts echoed by such leading OA proponents makes me feel better about being, on this issue, at odds with Stevan Harnad. I simply cannot agree that Open “comes with” Free, and the distinction bothers me. It should be relatively easy to convert Free to Open — simply add a Creative Commons or similar license — but I think it would be better to do that proactively. If we gloss over the difference between Free and Open at this relatively early stage of OA, we risk creating a (potentially enormous) body of Free text that must be updated to include complete, useful permissions when at last we realize that Free Is Not Open. (The game’s afoot: / Follow your robots, and upon this license / Cry “Free is not Open”!)

PMR: If you have read this far, thanks. Hopefully an increasing number of you won’t have needed to read the detail as you know it by heart. So the formal positions are clear. The problems are with labels and strategy. I hope the following is clear and correct:

• The full texts of BBB logically require Openness – see para 3 above.
• Any publication labelled “Gold” Open Access should be consistent with this. (I am not sure this is true for all Gold publishers/publications).
• Certain uses of the term “Open Access” (many under the term “Green” Open Access) do not explicitly require adherence to para 3 above. And many implementers (publishers) provide licenses under the label “Open Access” which explicitly forbid actions under 3.
• Some proponents of OA (esp. Stevan Harnad) argue that an publication labelled “Open Access” implicitly carries the intention of para 3.
• Many readers are unable to accept this and are deterred from re-using these Open Access publications for fear of breaking laws and license agreements.
• The Greenness, Goldness or Openness of a publisher or publication is often not trivial to determine. It requires careful reading of the license, possibly queries to the publisher (often unanswered) and may even depend on institution-dependent licenses. The “Open Choice” and other options offered by some publishers (pay for Open) do not always offer full OpenBBB – e.g. the publisher still retains copyright. Many “Green” publishers are not proactive in clarifying the situation and – given the shining examples of clarity above – may be accused of at best irresponsible laziness and at worst deliberately obfuscation (usually by inaction).

I hope all of us can agree on this analysis (whether or not we agree on the strategy) – if you don’t, please offer corrections. The questions are what should we do about it and how should we do it.
Stevan’s strategy is that we should devote our energies to achieving 100% Green Open Access and only then turn to achieving full Openness. (Obviously any Gold achievements are a benefit and may, but need not, be increased by Greenness). Stevan is clear and consistent in arguing that this should be the main current focus of the OA movement.
Bill and I and many other practising scientists (e.g. the founders of PLoS) cannot wait for full Greenness to occur. The need for “Open Data” is now rapidly growing can be fought alongside Open Access. It is critical to have clarity in access to and re-use of scientific data, and much of this (although not enough!) is contained in primary scientific publications. This is an area that must be fixed on a shorter time scale than Open Access has taken (that’s not a criticism – it’s a compliment to the staying power of Stevan Harnad and Peter Suber ans many others).
What is clear is that “Open Access” as a term is too vague an instrument for measuring the current position and providing effective instruments for change. By contrast “Open Source” is. We now need the following:

• an effective terminology. The concepts are clear. I like FINO though it’s not quite right as a label for a publisher or publication. “PLoS is a FINO publisher” doesn’t quite make it. Since Openness absolutely implies Free we don’t need both. “FullBBB” is accurate, but is not as catchy. “Molecules is a FBNO” publisher (FreeButNotOpen). It’s pronounceable. I need something where I can look up the status of a work instantly.
• The publications and the supplemental data themselves must explicitly carry the license and or  a statement or Openness.
• We must get an accurate survey of the field. For example the DOAJ does not include re-usability information. For example Molbank/Molecules is listed as Open but carries the phrase: “Copyright Arrangements
  If no alternate arrangement concerning copyright has been made with the Publisher: When you submit a paper and the paper is received, your paper will be asigned an unique manuscript ID and you will be asked to transfer copyright. […] This is a prerequisite for publication of your paper in this journal. The manuscript ID should be provided for copyright transfer form. [PMR: As you have seen from earlier postings here reporting correspondence  with Molecules this is not an oversight but a policy.]

We need to start doing this now. In parallel we have to keep advocating that:

• authors explicitly add creative commons (or science commons) licenses to their manuscripts and supplemental data
• frustrated re-users contact the publisher and ask for clarification about re-use. If this is an oversight and the publisher makes this clear that’s great. If the publisher is deliberately restricting the re-use of scientific data then ask them to make that publicly clear. If they don’t, publicly advertise that the publisher is being uncooperative in advancing science.
• and on the positive side show the community how valuable it is to re-use data. This isn’t easy when you are forbidden to use it. But that is the sort of things we are doing with SPECTRa and CrystalEye – we are creating our own dogfood as well as eating it.

The differences in strategy are not serious problems for the Open Access movement. Any large movement that has taken years to create against ignorance and then resistance backed by money will have them. Open Source has had very strong differences and yet it’s flourished and is a major area for innovation and wealth creation. Open Access is doing the same. And any publisher who wishes to denigrate the Open Access movement by hiring pitbulls immediately advertises their narrowmindness and fear of the future.
When the pitbulls start applying the same tactics to Open Data then we shall know we have succeeded in arriving.