The Panton Principles: A breakthrough on data licensing for public science?

I’ve been invited to a COST meeting in Porto (PT) on Monday. COST is an intergovernmental framework for European cooperation and I’m a member of a working group on automating computational chemistry (D37). But this is different – HLA-NET MC-WG is A European network of the HLA diversity for histocompatibility, clinical transplantation, epidemiology and population genetics. They are interested in how they share data (and materials) through an Open process.I didn’t need to know anything about sharing materials, privacy, etc. [BTW I may use data as a singular noun deliberately].

So last Tuesday we had a visit from Cameron Neylon (who incidentally gave a brilliant talk on Open Science , blogged by Nico Adams) and our group went to the Panton Arms (pub) for lunch. I took the opportunity to get Rufus Pollock of the Open Knowledge Foundation to discuss and clarify what our common stance is on Open Data. Cameron has blogged very comprehensively and usefully on the meeting. Rufus has also summarized views from Open Data Commons.

The critical thing to realise is that Open Scientific Data is not Open Software nor Open Content. It may sound arrogant but it can be difficult for a non-scientist to realise that is is different from maps, from Shakespeare, from photography, from government publications, from cricket scores. Scientists by default collect data, or calculate it, to justify their conclusions to prove they have done the work, to allow others to repeat the work.

It should be free, as in air.

They expect others to use it, without their permission. This could be to provde the original ideas right, or to prove them wrong. It could be to mine the data for ideas the original scientists missed. No scientist likes being proved wrong, or having someone else find ideas that they have missed. But it’s a central part of science. A scientist who says you can’t use my published data has no credibility today.

That’s not to say some scientists don’t try to hold their data back and mine the maximum from it before publishing. But it is becoming increasingly required by funders, by universities (in theses) and by some publishers that the data justifying a publication should be published in some way at the time of article publication. And by default there should be no restrictions on copying, re-use , republishing for whatever purpose and by whomever. I may not like it if my data is used to make weapons, or that a commercial organisation republishes it for money. But that is the implied contract I make by being a scientist. If I don’t like weapons derived from science there are other ways I can make my views known other than by adding restrictions and at times I have.

To summarize. Data itself must be completely free. The question is how to ensure that it is.

The Open Science and Open Knowledge community has been discussing this for about 2 years. We seem to be agreed that legal tools are counterproductive, and that moderation is best applied by the community. This is represented by Community Norms agreed practices that cause severe disapproval and possibly action when broken.

Our current crisis in Britain illustrates this. Huge numbers of Members of Parliament have been fiddling their expenses. They’ve been spending taxpayers’ money on cleaning their castle moats, buying second homes, antique rugs and so on. Huge amounts. This is, apparently, within the parliamentary guide lines.

But it is against the court of public opinion. It violates our Community Norms. The defence that it is within the rules illustrates the futility of the rules.

And it is incredibly difficult to draft good rules. So we’ve decided not to try to use the standard tools of copyright or licences.

For us Data are born Open. The question is how to state that. The simplest way is just to add the OKF’s Open Data button to the data. That’s a statement of intent. It says you can do whatever you like with this data without asking my permission. In many cases I think that is adequate.

However the community has also investigated the legal aspect and to provide a formal means of stating this in legal terms. This isn’t easy but the two approaches Public Domain Dedication and Licence (PDDL) and Creative Commons CC0 are roughly equivalent. I hope it’s useful to say that PPDL comes out of an Open Knowledge philosphy and deals with collections and other non-scientific content, whereas CC0 springs more directly fro
m science. And it IS complex I am meant to be an expert and I still find the details difficult. Here’s the CC0 FAQ:

How is CC0 different from the Public Domain Dedication and License (PDDL) published by Open Data Commons?

The PDDL is intended only for use with databases and the data they contain. CC0 may be used for any type of content protected by copyright, such as journal articles, educational materials, books, music, and art, as well as databases and data. And just like our licenses, CC0 has the added benefit of being expressed in three ways: through a human-readable deed (a plain-language summary of CC0), the legal code, and digital code. The digital code is a machine-readable translation of CC0 that helps search engines and other applications identify CC0 works by their terms of use.

This was the background when we tried to achieve a common view in the Panton. I’ll let Cameron take it from here:

The appropriate way to license published scientific data is an argument that has now been rolling on for some time. Broadly speaking the argument has devolved into two camps. Firstly those who have a belief in the value of share-alike or copyleft provisions of GPL and similar licenses. Many of these people come from an Open Source Software or Open Content background. The primary concern of this group is spreading the message and use of Open Content and to prevent freeloaders from being able to use Open material and not contribute back to the open community. A presumption in this view is that a license is a good, or at least acceptable, way of achieving both these goals. Also included here are those who think that it is important to allow people the freedom to address their concerns through copyleft approaches. I think it is fair to characterize Rufus as falling into this latter group.

On the other side are those, including myself, who are concerned more centrally with enabling re-use and re-purposing of data as far as is possible. Most of us are scientists of one sort or another and not programmers per se. We dont tend to be concerned about freeloading (or in some cases welcome it as effective re-use). Another common characteristic is that we have been prevented from being able to make our own content as free as we would like due to copyleft provisions. I prefer to make all my content CC-BY (or cc0 where possible). I am frequently limited in my ability to do this by the wish to incorporate CC-BY-SA or GFDL material. We are deeply worried by the potential for licensing to make it harder to re-use and re-mix disparate sets of data and content into new digital objects. There is a sense amongst this group that data is different to other types of content, particulary in its diversity of types and re-uses. More generally there is the concern that anything that smells of lawyers, like something called a license, will have scientists running screaming in the opposite direction as they try to avoid any contact with their local administration and legal teams.

PMR: I am completely aligned with Cameron. The added precision of legality is seriously outweighed by its difficulty and downstream problems. Cameron again:

What I think was productive about the discussion on Tuesday is that we focused on what we could agree on with the aim of seeing whether it was possible to find a common position statement on the limited area of best practice for the publication of data that arises from public science. I believe such a statement is important because there is a window of opportunity to influence funder positions. Many funders are adopting data sharing policies but most refer to following best practice and that best practice is thin on the ground in most areas. With funders wielding the ultimate potential stick there is great potential to bootstrap good practice by providing clear guidance and tools to make it easy for researchers to deliver on their obligations. Funders in turn will likely adopt this best practice as policy if it is widely accepted by their research communities.

So we agreed on the following (I think anyone should feel free to correct me of course!):

1. A simple statement is required along the forms of  best practice in data publishing is to apply protocol X. Not a broad selection of licenses with different effects, not a complex statement about what the options are, but best practice is X.

2. The purpose of publishing public scientific data and collections of data, whether in the form of a paper, a patent, data publication, or deposition to a database, is to enable re-use and re-purposing of that data. Non-commercial terms prevent this in an unpredictable and unhelpful way. Share-alike and copyleft provisions have the potential to do the same under some circumstances.

3. The scientific research community is governed by strong community norms, particularly with respect to attribution. If we could successfully expand these to include share-alike approaches as a community expectation that would obviate many concerns that people attempt to address via licensing.

4. Explicit statements of the status of data are required and we need effective technical and legal infrastructure to make this easy for researchers.

So in aggregate I think we agreed a statement similar to the following:

Where a decision has been taken to publish data deriving from public science research, best practice to enable the re-use and re-purposing of that data, is to place it explicitly in the public domain via {one of a small set of protocols e.g. cc0 or PDDL}.

PMR: agreed. The biggest danger is NOT making the assertion that the data is Open. There may be second-order problems from CC0 or PPDL but they are nothing compared to the uncertainty of NOT making this simple assertion. Do not try to be clever and use SA, NC or other
restricted licenses. Simply state the data are Open. Cameron finishes:

The advantage of this statement is that it focuses purely on what should be done once a decision to publish has been made, leaving the issue of what should be published to a separate policy statement. This also sidesteps issues of which data should not be made public. It focuses on data generated by public science, narrowing the field to the space in which there is a moral obligation to make such data available to the public that fund it. By describing this as best practice it also allows deviations that may, for whatever reason, be justified by specific people in specific circumstances. Ultimately the community, referees, and funders will be the judge of those justifications. The BBSRC data sharing policy states for instance:

BBSRC expects research data generated as a result of BBSRC support to be made availableno later than the release through publicationin-line with established best practice  in the field [CN – my emphasis]

The key point for me that came out of the discussion is perhaps that we cant and wont agree on a general solution for data but that we can articulate best practice in specific domains. I think we have agreed that for the specific domain of published data from public science there is a way forward. If this is the case then it is a very useful step forward.

PMR: completely agreed. Now there are some important actions:

  • get funders, universities and well-intentioned publishers to agreed on this approach, with appropriate modifications. It should be sufficient to see the Open Data button to know that the data are free for re-use.

  • Assert that Data covers images and tables and other ways of representing data. It is archaic and bizarre that data presented as an image are copyrightable. We must change this it’s far more important than the second order problems of PPDL/CC0

Cameron has called, this A breakthrough on data licensing for public science. If others agree let’s call it the Panton Principles of Open Data.

This entry was posted in Uncategorized. Bookmark the permalink.

15 Responses to The Panton Principles: A breakthrough on data licensing for public science?

  1. I’ve read Cameron’s blogpost on this topic as well as your own. I am interested to know how successful you have been in convincing your own colleagues at Cambridge University to make their data open? I am specifically interested in things like spectral data? You know of my interest in this area and it would be great to get one of the top universities to initiate release of their spectral data into the Open

    • pm286 says:

      Replies are getting lost – this is a test

    • pm286 says:

      @Chemspiderman Thanks.
      (a) all the code, source, content, specs in our group are Open. We ask that people acknowledge the use of our material; not everyone does
      (b) The JISC-funded CLARION project (http://www.jisc.ac.uk/whatwedo/programmes/inf11/clarion.aspx) is committed to exposing research output data in the Department as Open. Data are initially embargoed but when the researcher flips the publish-swicth then they will be exposed as Open.
      Note however that Open Data does not mean that data are necessarily fit for third-party use. Our CrystalEye data are fit for our purposes but you indicated that you required us to convert them to your format for inclusion in Chemspider, which we did at our own cost. This is not a sustainable model for the Department’s data – third parties have to accept Open data asis.

  2. Peter,
    Great post on an important topic. I’m just going to throw some random thoughts together here.
    1) One of the problems in this area (copyrights, databases, patents) is that the governing legal rules differ from country to country. I applaud the notion of generating a “chemistry community” – maybe even a “science community” – set of best practices. This can, I hope, cross boundaries.
    2) I worry about the need to claim that data is open, as in a sense diminishing what is rightfully there from the start. My understanding is that data cannot be copyrighted here in the US. While the statement “the melting point of benzene is 5 °C” can be copyrighted, the underlying fact is in the public domain and one can freely use that information for any purpose. The same is true about spectra – the IR image of the absorption of benzene is copyrightable (How one displays data can be a creative effort – see below) but the fact that benzene does not absorb at 1900 cm-1 is “data” and can be freely re-used.
    3) As part of “best practices” I would like to see journals insist that all data used in preparing the article be submitted as part f the article. The data should then be made available for re-use with no limitations and inhibitions (I.e. no cost). Currently, that would be as “supporting materials” but I hope that in the future this data might be more intimately connected with the article itself. In an ideal world, data should be packaged in a way to facilitate re-use, but that is something I am willing to let the world grow into slowly.
    4) I think you need to be very careful about the notion to “Assert that Data covers images and tables and other ways of representing data”. Representation of data can very much be a creative process and should be copyrightable. For example, a landscape photographer is simply taking an image of data – the way the countryside appears on a particular date and time. Let that photographer be Ansel Adams, and he obtains a “reprentation of data” that is coloured by the artists perceptions and biases and influences and creates a product that I call art – and that image should be protected. A chemist may do the same thing with an array of data to create a plot that clearly and distinctly manipulates the data to make a point. In such a case, I would protect the image (the plot) with copyright and insist that the underlying data (such as the excel file) be published simultaneously within the supporting materials and that data is open in the public domain. Things get a bit tricky here too – does the photo of a gel run by a biochemists count as data or as a creative image? I think the former, but I’m open to discussion.
    Bottom line is I fully support an effort to create best practices and think tht th “Panton Principles” are a great starting point!

  3. Peter, I’ll respond in greater detail elsewhere but I should correct one major misapprehension:
    > I hope it’s useful to say that PPDL comes out of an Open Knowledge philosphy and deals with collections and other non-scientific content, whereas CC0 springs more directly from science.
    This almost exactly the wrong way round. The Public Domain Dedication and License was specifically designed for use on “data/databases” — scientific or otherwise — whereas, at least originally, CCZero was aimed at traditional cultural material (i.e. the type of stuff other CC licenses apply to).
    I would also take some issue with the CCZero FAQ! The PDDL has a human-readable preamble in addition to its “legal code” and an RDF expression for it is in progress (though how important this type of thing is I’m not sure, after all most other areas get by with much simpler machine readable shorthands for licenses — think of the license drop-down on sourceforge!)
    To reiterate: the Public Domain Dedication and License (PDDL) is entirely suited for application to scientific data, and is perhaps is better suited than CCZero in that it is more focused on the data case.

    • pm286 says:

      @Rufus thanks. I was strongly influenced by the CC0 FAQ and am grateful for these points. I think it’s true to say that PDDL comes from Open Knowledge, right. I thought CC0 was a product of Science Commons, but looks like I am wrong.

  4. The Public Domain Dedication and License is an Open Data Commons license and was primarily drafted by Jordan Hatcher and Professor Charlotte Waelde. The Open Knowledge Foundation today helps run ODC (though ODC’s drafting work is run by the ODC Advisory Council).
    CCZero is not a product of Science Commons but of Creative Commons and was a development of their original public domain dedication tool. The original tool was entirely focused on normal copyrightable cultural material but CCZero was specifically extended to include the data case.

  5. Physchim62 says:

    Can I just reply to Steven Bachrach on some technical points:
    1) Pure data cannot be copyrighted anywhere. This is one of the basic principles of copyright law and dates from the Berne Convention of 1886. Most recently, it is enshrined in the TRIPS agreement (arts. 9 & 10). However, collections of data (such as the Aldrich catalogue) CAN be copyrighted, as there is deemed to be a creative input in deciding which data to include.
    2) In Europe, there is also a separate beast called “database rights”, which doesn’t exist in the U.S. It works a little bit like copyright, but is subject to different laws: it’s also less international in scope. The PDDL contains a specific waiver/license of these database rights, unlike many of the earlier CC licenses.
    3) I think it is very problematic to require “all data used in preparing the article” be published with the article. In some cases, a part of the research might be held back for patent reasons, yet the authors may legitimately publish the rest of the work so long as the part which is held back does not contradict their assertions.

    • pm286 says:

      @Physchim62 thanks – this is very clear.
      (3) “all data used in preparing the article” is best endeavour and there are Community Norms which will moderate this. Patents is one area. Human data is another.

  6. Pingback: s5h.net » Blog Archive » Information Wants to be Free

  7. Pingback: Science in the open » Sci - Bar - Foo etc. Part II - SciFoo - Engaging with the world

  8. Pingback: Panton Principles for Open Data « Be openly accessible or be obscure

  9. Pingback: Panton Principles | Pack 6 – Palo Alto

Leave a Reply

Your email address will not be published. Required fields are marked *