Wednesday 31 March 2010

Linked Data and Reality

I have a copy of the really interesting book “Data and Reality” by William Kent. It’s interesting at several levels; first published in 1978, this appears to be a “print-on-demand” version of the second edition from 1987. Its imprint page simply says “Copyright © 1998, 2000 by William Kent”.

The book is full of really scary ways in which the ambiguity of language can cause problems for what Kent often calls “data processing systems”. He quotes Metaxides:

“Entities are a state of mind. No two people agree on what the real world view is”
Here’s an example of Kent from the first page:
“Becoming an expert in data structures is… not of much value if the thoughts you want to express are all muddled”
But it soon becomes clear that most of us are all too easily muddled, at least when
“... the thing that makes computers so hard is not their complexity, but their utter simplicity… [possessing] incredibly little ordinary intelligence”
I do commend this book to those (like me) who haven’t had formal training in data structures and modelling.

I was reminded of this book by the very interesting attempt by Brain Kelly to find out whether Linked Data could be used to answer a fairly simple question. His challenge was ‘to make use of the data stored in DBpedia (which is harvested from Wikipedia) to answer the query

“Which town or city in the UK has the highest proportion of students?"
He has written some further posts on the process of answering the query, and attempting to debug the results.

So what was the answer? The query produced the answer Cambridge. That’s a little surprising, but for a while you might convince yourself it’s right; after all, it’s not a large town and it has 2 universities based there. The table of results shows the student population as 38,696, while the population of the town is… hang on… 12? So the percentage of students is 3224%. Yes, something is clearly wrong here, and Brian goes on to investigate a bit more. No clear answer yet, although it begins to look as if the process of going from Wikipedia to DBpedia might be involved. Specifically, Wikipedia gives (gave, it might have changed) “three population counts: the district and city population (122,800), urban population (130,000), and county population (752,900)”. But querying DBpedia gave him “three values for population: 12, 73 and 752,900”.

There is of course something faintly alarming about this. What’s the point of Linked Data if it can so easily produce such stupid results? Or worse, produce seriously wrong but not quite so obviously stupid results? But in the end, I don’t think this is the right reaction. If we care about our queries, we should care about our sources; we should use curated resources that we can trust. Resources from, say… the UK government?

And that’s what Chris Wallace has done. He used pretty reliable data (although the Guardian’s in there somewhere ;-), and built a robust query. He really knows what he’s doing. And the answer is… drum roll… Milton Keynes!

I have to admit I’d been worrying a bit about this outcome. For non-Brits, Milton Keynes is a New Town north west of London with a collection of concrete cows, more roundabouts than anywhere (except possibly Swindon, but that’s another story), and some impeccable transport connections. It’s also home to Britain’s largest University, the Open University. The trouble is, very few of those students live in Milton Keynes, or even come to visit for any length of time (just the odd Summer School), as the OU operates almost entirely by distance learning. So if you read the query as “Which town or city in the UK is home to one or more universities whose registered students divided by the local population gives the largest percentage?”, then it would be fine.

And hang on again. I just made an explicit transition there that has been implicit so far. We’ve been talking about students, and I’ve turned that into university students. We can be pretty sure that’s what Brian meant, but it’s not what he asked. If you start to include primary and secondary school students, I couldn’t guess which town you’d end up with (and it might even be Milton Keynes, with a youngish population).

My sense of Brian’s question is “Which town or city in the UK is home to one or more university campuses whose registered full or part time (non-distance) students divided by the local population gives the largest percentage?”. Or something like that (remember Metaxides, above). Go on, have a go at expressing your own version more precisely!

The point is, these things are hard. Understanding your data structures and their semantics, understanding the actual data and their provenance, understanding your questions, expressing them really clearly: these are hard things. That’s why informatics takes years to learn properly. Why people worry about how the parameters in a VCard should be expressed in RDF. It matters, and you can mess up if you get it wrong.

People sometimes say there’s so much dross and rubbish on the Internet, that searches such as Google provides are no good. But in fact with text, the human reader is mostly extraordinarily good at distinguishing dross from diamonds. A couple of side searches will usually clear up any doubts.

But people don’t do data well. Automated systems do, SPARQL queries do. We ought to remember a lot more from William Kent, about the ambiguities of concepts, but especially that bit about computers possessing incredibly little ordinary intelligence. I’m beginning to worry that Linked Data may be slightly dangerous except for very well-designed systems and very smart people…

Tuesday 9 March 2010

When data shouldn’t be open?

There is a big momentum these days about data being accessible, available, and re-usable. Increasingly people want open data; Science Commons have been recommending using CC0 to make the fully open status of data clear. More recently the Panton Principles start:

“Science is based on building on, reusing and openly criticising the published body of scientific knowledge.

For science to effectively function, and for society to reap the full benefits from scientific endeavours, it is crucial that science data be made open.”

We’ve been big fans of Open Access at the DCC since its early days. We use a Creative Commons licence for our content by default. This blog was one of the earliest to be specific about a Creative Commons licence not only for the core text that we write, but also for the comments that you might add here.

So we strongly support the Open Data approach… where possible. For of course in some areas of science and research, there are data that cannot be open. Usually this is because the data are sensitive. They could be personal data, protected under Data Protection laws. Sensitive personal data (such as medical record data) has extra requirements under those laws. They could be financial microdata, commercially sensitive. Or perhaps data with strong commercial exploitation potential. They could be anthropological data, sensitive through cultural requirements. Research needs to go anywhere, whatever the issues; we can’t be constrained to only research where the data can be open.

So perhaps it’s as simple as that: some science should have open data, and some should have closed data?

Well, maybe not. Because the underlying issue of the Panton Principles must still apply. Research should be verifiable, whether through repeatable experiments or through re-analysable data. Unverifiable research is, well, unreliable- perhaps indistinguishable from fraud. Some access is needed; perhaps we should think of even sensitive data as Less Open Data rather than closed data.

So how do you go about dealing with sensitive data? Keep it secure, transfer securely, provide access under strict licences and controls in dat enclaves, aggregate, de-identify, anonymise, there are plenty of tricks in the book. That’s the topic of the 4th Research Data Management Forum starting tomorrow in Manchester. I’ll hope to have more to write about what we learn later.

A Blue Ribbon for Sustainability?

When we talk about long term digital preservation, about access for the future, about the digital records of science, or of government, or of companies, or the designs of ships or aircraft, the locations of toxic wastes, and so on being accessible for tens or hundreds of years, we are often whistling in the dark to keep the bogeys at bay. These things are all possible, and increasingly we know how to achieve them technically. But much more than non-digital forms, the digital record needs to be continuously sustained, and we just don’t know how to assure that. Providing future access to digital records needs action now and into that future to provide a continuous flow of the necessary will, community participation, energy and (not least) money. Future access requires a sustainable infrastructure. Ensuring sustainability is one of the major unsolved problems in providing future access through digital preservation.

For the past two years I have been lucky enough to be a member of the grandly named Blue Ribbon Task Force on Sustainable Digital Preservation and Access, along with a stellar cast of experts in preservation, in the library and archives worlds, in data, in movies… and in economics. C0-chaired by Fran Berman (previously of SDSC, now of RPI) and Brian Lavoie of OCLC, the Task Force produced an Interim Report (PDF) a year ago, and has just released its Final Report (Sustainable Economics for a Digital Planet: Ensuring Long-Term Access to Digital Information, also PDF). (The Task Force was itself sustained by an equally stellar cast of sponsors, including the US National Science Foundation and the Andrew W. Mellon Foundation, in partnership with the Library of Congress, the UK’s JISC, the Council on Library and Information Resources, and NARA.)

Sustainability is often equated to keeping up the money supply, but we think it’s much more than that. The Task Force specifically looks at economic sustainability; it says early in the Executive Summary that it’s about

… mobilizing resources—human, technical, and financial—across a spectrum of stakeholders diffuse over both space and time.”

If you want a FAQ on funding your project over the long term you won’t find it here. Nor will you find a list of benefactors, or pointers to tax breaks, or arguments for your Provost. Instead you should find a report that helps you think in new ways about sustainability, and apply that new thinking to your particular domain. For one of our major conclusions is that there are no general, across the board answers.

One of the great things about this Task Force was its sweeping ambition. Not just content with bringing together a new economics of sustainable digital preservation, but thinking so broadly. This was never about some few resources, or this Repository or that Archive, it was about the preservation and long term access of major areas of our intellectual life, like scholarly communication, like research data, like commercially owned cultural content (the movie industry is part of this), and the blogosphere and variants (collectively produced web content). Looking at those four areas holistically rather than as fragments forced us to recognise how different they are, and how much those differences affect their sustainability. They aren’t the only areas, and indeed further work on other areas would be valuable, but they were enough to make the Task Force think differently from any activity I have taken part in before.

The report is, to my mind, exceedingly well written, thanks to Abby Smith Rumsey; it far exceeds the many rather muddled conversations we had during our investigations. It has many quotable quotes; among my favourites is

“When making the case for preservation, make the case for use.”

Reading the report is not without its challenges, as you might expect. It has to marry two technical vocabularies and make them understandable to both communities. I’ve been living partly in this world for two years, and still sometimes stumble over it; I remember many times screwing up my forehead, raising my hand and asking “Tell us again, what’s a choice variable?” And the reader will have to think about things like derived demand for depreciable durable assets, nonrival in consumption, temporally dynamic and path-dependent, not to mention the free rider problem. These concepts are there for a reason however; get them straight and you’ll understand the game a lot better.

And there are not surprisingly big underlying US-based assumptions in places, although the two resident Brits (myself and Paul Ayris of UCL) did manage to inject some internationalism. Further work grounded in other jurisdictions would be extremely valuable.

Overall I don’t think this report is too big an ask for anyone anywhere who is serious about understanding the economic sustainability of digital preservation and future access to digital materials. I hope you find the great value that I believe exists here.

Monday 1 March 2010

DCC: A new phase, a new perspective, a new Director

As the DCC begins its third phase today, I am delighted to announce the appointment of our new Director, Kevin Ashley, who will succeed me upon my retirement in April 2010.

Kevin Ashley has been Head of Digital Archives at the University of London Computer Centre (ULCC) since 1997, during which time his multi-disciplinary group has provided services related to the preservation and reusability of digital resources on behalf of other organisations, as well as conducting research, development and training. The group has operated the National Digital Archive of Datasets for The National Archives of the UK for over twelve years, delivering customised digital repository services to a range of organisations. As a member of the JISC's Infrastructure and Resources Committee, the Advisory Council for ERPANET, plus several advisory boards for data and archives projects and services, Kevin has contributed widely to the research information community. As a firm and trusted proponent of the DCC we look forward to his energetic leadership in this new phase of our evolution.

So far so press release. But I'd go further. I can't tell you how pleased I am with this appointment. As some readers will know, I have personally lobbied all and any potential candidates for this post since before I officially announced I was leaving. I understand we had some excellent candidates (I wasn't directly involved), more than one of whom might have made an excellent Director. But I'm particularly pleased at Kevin's appointment for several reasons: he is well engaged in the community including good connections with JISC, our major funder), he's tough enough to keep this tricky collaboration thing going, he has an excellent technical understanding, and he has great experience of actually managing this stuff in all its crusty awfulness. I particularly remember his discussion (on a visit to the Edinburgh Informatics Database Group) about issues like how best to deal with an archived dataset where they came across the characters "five" in a field defined as numeric! You can make it work or make it a record but not both...

So congratulations Kevin, and good luck!