Protection of Freedoms Bill

Lord Sutherland of Houndwood Excerpts

Wednesday 15th February 2012 Lords Chamber

Moved by

[Lord Sutherland of Houndwood]

55A: Clause 102, page 88, line 35, at end insert “which may set conditions on republication or communication of the data to third parties”

[Lord Sutherland of Houndwood]

My Lords, the amendment is in my name as well as that of my noble friend Lady O'Neill of Bengarve. She has already given her apologies to the Front Bench. Having sat through two long days on Report, she finds that her commitments can no longer be put off and are subject to the vagaries of how we put our business together in this House. I am sure that noble Lords would have had a much more precise, analytic and forensic speech from her than from me, but I will try to raise the points that I believe she wished to put and express the concerns that I share with her. Those concerns are shared by the wider university community in the UK. I believe that they are grateful to the Minister for the time that he has taken to meet them and to talk to them about some of these issues.

That being so, there is a shift in atmosphere from the previous amendment because I wish to focus on the implications of the Bill for research and research data sets held significantly by universities and public research institutions.

Research is an international and very competitive business. There is a risk that some of the provisions of the Bill may undermine the competitiveness of much of our excellent research in this country. That is unintentional, but I hope that I can demonstrate that there are some difficulties that need to be resolved.

The specific Amendment 55A is a modest suggestion that any licence for reuse of data sets may have conditions attached to it following the comments of those whose data sets they are. That is a modest way of protecting the interests of our researchers and the research community and, more broadly, UK plc.

The impact of the Bill as it stands will be such as to fall on both individual researchers and on institutions. In the case of individual researchers, it will change behaviour. If you change the rules about how research data are to be treated, you will change the behaviour of researchers. They are pretty clever people, so you need to watch out. Beyond that, there are more serious implications.

It may well be that the provisions of the Bill result in activities that are inimical to and, indeed, unintentionally unjust to individual researchers. The data set, for example, may well have been built up over a long period and involve substantial career commitment by individuals. If you spend your time on large data sets in particular, that is a major—or indeed a lifetime—commitment. The data sets may well have been built up involving the distinctive, significant and, on occasion, unique skills of the individual researcher. I am not quite convinced that the Bill has taken that sufficiently into account. It will certainly be inimical to career development and commitment—and, indeed, in terms of the opportunity costs—in respect of the work of individual researchers. They take time, they do this rather than that, and they use the skills they have, which may well be unique and very distinctive.

In the case of institutions, there clearly may well be problems where a home or sponsoring institution has invested significant resources in the data sets or in building them. As head of more than one such institution, I know that huge resources are invested to build appropriate data sets, in terms of time, space—which is very important and expensive—individual members of staff and money. These are the commitments the institution makes, and there is a risk that they would be set back dramatically despite the effort and the commitment involved. Equally—this point has been made in previous discussion but it is still there—the institutions may well have commercial interests in the research and data sets in question. Beyond that, we fund universities in such a way that the research assessment exercise depends very significantly on the uniqueness and distinctiveness of research. If the data sets that are the foundation for these are too easily available, that sets at nought the efforts of those who worked on them; and makes it easy for those who did not to pillage those data sets. Academics on the whole are nice people, but when it comes to this kind of competitiveness, all rules are set aside.

There is one consequence that I am sure is unintended and which relates to the previous amendment—although not perhaps in a congenial way. I declare, unusually, a future interest, which I may have if this legislation goes through. As I understand it, one of the ways of avoiding the data sets moving out too quickly and in an unregulated, uncontrolled fashion is to have co-ownership with some private sector activity, firm or company. My interest would be that I might set up such a company, the sole purpose of which would be co-ownership of data sets with universities and research interests. I could become very rich—but being the chap I am, I would dedicate all the money to a charity to support research in universities. That is one possible way of beginning to avoid the implications of the legislation as I understand it. This is partly jocular but it is more than that. Ingenious people are out there and will find solutions to retain data sets that they, for good reason, believe are important. This is not miserable secretiveness, this is how research operates. This is how the competition deals with those who are involved in research.

Finally, I believe there is a difficulty in identifying what data sets will fall under this Bill as currently formulated, if it becomes law. What will count? I give the House one example, just as a test. In the description, decoding and understanding of the structure of DNA, Crick and Watson did excellent and magnificent work in Cambridge which was properly recognised with a Nobel Prize. Yet the missing piece in the jigsaw was here in London, with Professor Maurice Wilkins and Dame Rosalind Franklin at King's College London. She is now at last being recognised for her part in this. The data sets of material that she had built up using techniques of electron microscopy were, when they were available to Crick and Watson on a shared basis, what put the final piece of the jigsaw in place. The picture became clear to them and they could move ahead.

I do not think that these data sets would be ruled out under the definitions given in the Bill, because they are simply printings that you could look at. They are not analysed or pre-digested and there is no interpretation given. If they had been requested belligerently by Crick and Watson, they could have saved themselves the price of a rail ticket to London. Your Lordships can see the implications. What counts as a data set, when Rosalind Franklin had created this data set that made all the difference to what has changed the course of life for all of us? I believe there are questions about the definition of a data set because the Bill is really meant to deal with other issues initially but, as it so happens, it is now being applied to research and research data sets in some of the best institutions in this country. I beg to move.

[Baroness Brinton]

My Lords, I shall speak to Amendment 56, but in so doing I start by expressing my support for the amendment tabled by the noble Baroness, Lady O'Neill, and the noble Lord, Lord Sutherland. I will also be incorporating some of the points that the noble Baroness, Lady Warwick, would have made had she been able to be in her place, but for exactly the same reason as the noble Baroness, Lady O'Neill—the unfortunate and substantial delay in getting to this—other commitments have meant that she needs to be elsewhere in the Palace of Westminster.

I also want to thank the Minister for the helpful meetings with him and his officials on the complex subjects of universities, the publication of their research and the implications for the practical working of the Freedom of Information Act. In Committee, I outlined a number of problems that universities face that are not analogous to the use of FOI in non-research areas of higher education institutions, not least because of the size, duration and complexity of many research contracts. Universities are mindful of their duties to respond to FOI requests elsewhere, and in the main they absolutely do. Also, the universities that we have talked to about the problems facing research and FOI are clear that this is not special pleading for the sector as a whole over freedom of information. Nor do they support any institution that does not comply with FOI requests in the mainstream.

The issues here are quite specific. They are about whether the exemptions currently outlined in the Information Commissioner's guidance to the higher education sector can be effectively applied, given the nature of research and whether, in the case of commercial partners, it might give rise to suspicion by those partners that their own confidential data might be seen by others following an FOI request. In Committee, the Minister asked for evidence of where the current exemptions do not apply. Here, from the Information Commissioner's guidance to higher education institutions—which, for brevity, I shall refer to as HEI—are a handful of examples that researchers and their universities have told us really need clarification.

The guidance on Section 22 refers to information intended for future publication. The information is exempt if it is intended for future publication and it is reasonable to withhold the information until that point, subject to the public interest test. While this will certainly apply to research data which an HEI intends to publish, provided that withholding the information is reasonable, it will not apply if there is no intention to publish the results at the time the request is made, which, as the ICO guidance makes clear, is the relevant time for him.

In general, HEIs would expect the data supporting research conclusions to be published, or at least to be available to others, when the conclusions themselves are published. However, in the case of longitudinal studies, the decision to publish may not be made until a late stage in the study, not least because it is not clear what will be reported, or how. Moreover, usually the material is published in the form of a peer-reviewed article, which is often only the tip of a much larger iceberg of data that are not published. I am mindful here of the specific example that my noble friend Lady Sharp gave us at an earlier stage of these proceedings about the very complex data set that she managed for decades, which would certainly fall into that last category.

--- Later in debate ---

[Lord Lucas]

My Lords, before I speak to my amendment in this group, may I first say that I hope my noble friend will treat the points raised by the noble Lord, Lord Sutherland of Houndwood, and my noble friend Lady Brinton with seriousness? It is clear that in an internationally competitive environment it is very important that people have confidence in the proper protection of research databases.

I disagree with both amendments. The amendment moved by the noble Lord, Lord Sutherland, would blow a hole below the waterline of this clause and would certainly destroy all my attempts to get other information out of universities. The amendment spoken to by the noble Baroness, Lady Brinton, misses important points on the other side. It is terribly important that data become available once results have been published. Many of these programmes go on for a long time. Because we intend to use the data in a whole series of publications over the next 20 years, we will never let them go. However, it must be possible for people outside a research group to criticise the results as they are being produced or false conclusions will be dropped into science and never properly got at. To pick one example, given by the noble Lord, Lord Sutherland, the inspiration of Crick and Watson had to be combined with the meticulous work of Franklin. Without that combination and the data being made public, the discovery would not have been made.

[Lord Sutherland of Houndwood]

On a point of order, it was information shared between research colleagues in two different institutions. In an atmosphere that is perhaps not quite as common today, it was willingly shared.

[Lord Lucas]

I understand and I remember from reading biographies that that was the case. None the less, the data were shared. To take an example from my time on the Front Bench as spokesman for agriculture during the problem of BSE, the Ministry of Agriculture, Fisheries and Food had been meticulously researching what was happening with this plague and had years of data. We had good people internally who were researching it. We did not know what was happening and we kept the data to ourselves. Three weeks after we released the data, we were told what was happening, which was transmission by food. That was right. Making data public, beyond a research group, is a very important thing to do at the right time. I should not like to see something in legislation that prevents that and allows people to hog data that should be public so that they can be properly criticised and understood.

My amendment is not on the same subject; it concerns technical bits of drafting in the same clause. I very much welcome the determination to provide greater access to data sets. It is something that I have struggled with, particularly with universities. All the universities that send me data stick copyright notices on them, which I studiously ignore. They have yet to sue me for it, probably because they have better things to do. Alternatively, I proposed as a remedy to one university that, if it insisted on its copyright notice, I would automatically generate an FOI exemption for every one of my users who wanted to access the data. The university thought that a number of 10,000 users a day was getting a bit large.

It is important that we understand that, when information is released under FOI, it can be passed on and made public; and that the generating institution does not retain some sort of control over it merely on a whim. I can understand why that might be the case if the material comprises something done under a publication scheme and is paid for, but otherwise it is very important that the information can be circulated whether in news media or in publications such as mine —the Good Schools Guide—or in many other applications.

I do not see why the proposals in the Bill do not go further and why they are restricted to data sets. It is common for all kinds of information released under FOI to be accompanied by a copyright. However, it is often obvious from the information that it has no conceivable commercial value to the public authority. A requester may have obtained a policy that he or she wants to publish on a website which demonstrates alleged shortcomings in an institution; for example, it may show that a decision has been taken without proper consideration of the consequences. The requester may want to write to Members of this House about the information that has been disclosed. Why should they be prevented from doing so by a copyright notice? It seems to me that the principles we are setting out in this clause should go further.

My second concern is about the definition of “data set”, which I believe is unjustifiably narrow. The Cabinet Office carried out an open data consultation which sets out admirably ambitious objectives for the greater use of data sets. Many collections of data currently gathered by public bodies which may be essential to revealing the inner workings of government do not seem to fall within the legal definition of a data set as set out in Clause 102(2). Any electronic collection of data which is the result of analysis or interpretation cannot be a data set because of new subsection (5)(b)(i) of the definition. The obligation to release it in reusable form will not apply to it, nor will the requirements to release it subject to the minimal restrictions embodied in a specific licence. The Information Commissioner will not be able to require that this collection of data must be published under an authority’s publication scheme. It seems that only raw data untouched by human hand are to be affected by this clause. That may suit people like me who spend their day with programs interpreting data, but most people want access to something which has been prepared for human consumption and has been set out in a way that members of a local authority are intended to understand rather than the geeks in their data department. I do not understand why the Government are seeking to exclude from this clause data which have been made human readable, as it were.

Paragraph (c) of the definition states that a data set remains one only so long as all or most of the information in it,

“remains presented in a way that (except for the purpose of forming part of the collection) has not been organised, adapted or otherwise materially altered since it was obtained”.

I am puzzled by that definition. A publication of data from a database may start off in the form of a spreadsheet consisting of three columns but is reorganised to consist of seven columns. Has enough been done to prevent the data set being published? How is anyone to know that? How is anyone to interpret that? How is any user to know that something is publishable? It seems to me that we are setting ourselves up for endless arguments. I am particularly concerned that authorities may find that, by reorganising data, they are able to conceal it from publication. It does not seem to me that that is the Government’s intention. I very much hope that they will explain to me their understanding of how this subsection may not be used in that way.

--- Later in debate ---

[Lord Sutherland of Houndwood]

My Lords, I thank the Minister for his very careful reply, the care he has taken over the various points made and his willingness to have further discussions. I have no doubt that that offer will be taken up.

However, I have two or three points. First, I have no doubt that the commitment to freedom of information is as strong among those who proposed these amendments as it is with the Government. In fact, one of my prouder moments, many years ago, was when I was given an award by the society that campaigned for what became the Freedom of Information Act because of the way I conducted a particular government inquiry. I am committed to the importance for public services and within the public sector of freedom of information.

Secondly, I will read with care what the Minister said about licences. One question that I will want to have answered—it may well be in Hansard—is about what degree of input a researcher, a research institute or a university will have into the drawing up of the licence that is applied to a particular publication. I am not yet clear whether Rosalind Franklin’s photographs would have fallen under the definition of a data set. If they had fallen under that definition, I do not think the cause of science or human life would have been advanced by having them accessible to whoever wanted to use the Freedom of Information Act to drag them up at that stage.

My third point is on the costs of responding. I chaired a group that advised a research project in Scotland called Generation Scotland. I shall not go into the details, but it was shot through with personal information. I can reassure the Minister that the cost of redacting any of that would be immense. Indeed, such is the sensitivity at a personal level of some of the information that the body responsible for managing it allows no one individual access to the whole data set. It is a double-lock situation, and that is absolutely right. That is not against the public interest; it is in the public interest and that of the 15,000 families in Scotland that have contributed very sensitive information for that research.

I welcome the comments and look forward to more detail on the timing of the commencement of Clause 102. In closing, I simply re-emphasise the point made by my noble friend Lord Oxburgh. We are talking about information for the public. In this case, the public is the entire world. The public is every competitor we have in the research business and in the commercial business that will spin out from it. That is still the real concern.

With the intention of discussing this with my noble friend Lady O’Neill, at this stage, I withdraw the amendment.

Amendment 55A withdrawn.