ANALYSIS AND NEWS Is there a library-sized hole in the internet? David Weinberger is senior researcher at Harvard’s Berkman Center for Internet & Society, and has been instrumental in the development of ideas about the impact of the web. Shortly before his recent keynote presentation at OCLC’s EMEA Regional Council Meeting in Florence, he spoke with Sarah Bartlett

You rose to prominence as Could libraries have done about the cultural objects we’ve entrusted an internet thought leader, anything differently? with them. We have thousands of years with pioneering texts such as In the face of considerable challenges, libraries invested in our culture, and we lose it at The Cluetrain Manifesto and have done very good things, but it’s going to our peril. It’s not just about the books, and Everything is Miscellaneous. take more. Libraries are providing open access not even just the librarians; I also mean the What led you into the world of to a closed world, and that is a tremendous communities that libraries serve. Through a libraries? service, but they are severely constrained library graph, libraries could collectively make In Everything is Miscellaneous I explored the by copyright laws that were not designed available every scrap of information they have way the internet is redefining our ideas about for a networked age. They also have limited about those cultural items. how we organise things and ideas, and the budgets. But they were early to digitise their move from physical to digital and networked catalogues in the 1960s, and they were also What do you mean by library resources is a prime example of that. quick to get on the web as portals. And for at ‘library graph’? As a result of Everything is Miscellaneous, I least five years, there has been a blooming of A library graph, which is a concept rather was offered a position as co-director of the library innovation, which is going down the than a physical entity at this point, is a means Harvard Library Innovation Lab. right track. by which the library world can publish its knowledge in an extensible and highly useful Besides format changes, what And what is the right track? way. It draws on linked data, which breaks is the most significant impact of Assuming that content remains locked up, information down to its smallest molecules, the web on libraries today? then I think the right track is to make library forming simple statements that connect two Library knowledge – the content; the information both public and interoperable things by a relationship and point to relevant metadata; what librarians and the community where possible. Libraries can best achieve this web locations. Those statements interconnect know about items held – is being lost to the by sharing their data, making it interoperable within and across disciplines, to form a graph. web. This represents an immense amount of Information previously stored in siloes now culture. The most basic components of the ‘It’s a huge technical spreads right across the internet, and every web are links, but if you want to talk about a time someone adds a new statement, the heap book, what do you link to? There is no clear challenge; many brilliant gets smarter. answer. people are working on it’ Let’s take the book as an example. When They might turn to Wikipedia, but only people talk about Hamlet or Moby Dick, around 70,000 books actually have a page they are generally not talking about specific on Wikipedia, so people rely on commercial so the rest of the world can mash it up with editions, but there are times when editions sites like Amazon. We aren’t even meeting other data. There has been a huge amount of and translations need to be distinguishable. the most basic requirement, linking, much development across the library world in this Representing this complexity in terms of less having a way to refer to the history of the area. It’s a huge technical challenge; many content, structure and ideas is a challenge that work, how it’s affected people and culture. brilliant people are working on it, and I think graphs can meet. With graphs we can traverse Facebook holds huge volumes of there’s real progress. the entire space of ideas and information information about its users and their lives, Most of the data will be metadata; libraries about the objects that libraries hold, but we have no equivalent for what libraries can’t publish the content of a book but they discovering and capturing every conceivable know. That is a huge hole in the internet, can publish information about the book. So type of relationship. and it has at least two negative consequences. that’s bibliographic data and anything else the Firstly, as library information becomes harder library can make available without violating So people will add to the to find, it becomes less relevant. Secondly, copyright or breaching user privacy, including library graph, maybe linking to libraries themselves become marginalised. anonymised and aggregated usage data. They datasets elsewhere? The culture that libraries represent becomes can also encourage their communities to talk Yes. The graph would open up library invisible on the internet, and the perceived more publicly about their interactions with knowledge to every other website, every field, value of libraries diminishes. the library.Libraries know a huge amount every person, in the form of machine-readable

4 Research Information april/may 2015 @researchinfo www.researchinformation.info ??????????????

ANALYSIS AND NEWS

data. The danger is that if we don’t do this, creating recommendation engines that by blogging or reviewing library resources if we only use the web as a portal into the provide just the right degree of difference. online for example, and we can share these library, the culture that libraries represent may community contributions. be overwhelmed by a more accessible and What do you mean by ‘just the If people could use an open e-reader superficially attractive culture. right degree of difference’? that supported the new open annotation I’m concerned about how we get people standard, then their notes might become part Will this machine-readable text to encounter ideas that are just different of a continually-enriched corpus of text. By eventually marginalise the web’s enough from what is familiar to them. It’s a incorporating marginalia in metadata, we human-readable content? very small degree of difference that works. could reconstruct what a professor of history Preserving the human cultural endeavour If ideas are too different – like some deeply found interesting in a set of works. We could is the most important thing. Machines can challenging revision of history, for example mash that up with other historians and help us find content or even summarise it, but – then people will not be persuaded and will thinkers. even that’s not essential. Machine-readable tend to dismiss them. Only with the right metadata helps us find what we value, but amount of difference will people appreciate How do you see the future of ultimately the aim is to get words, pictures something new and perhaps be changed by it. the library graph? and sounds in front of humans. This is a real librarian skill: ‘Here’s something The future is unclear because we don’t know that’s similar to Gone Girl; it’s by this author, what’s going to happen with copyright; How does the library graph help Patricia Highsmith from the 1950s, and you we don’t know which business models will users to access content? might try that.’ With ‘you might try that’ the prevail; I certainly don’t expect every library Firstly, the graph makes it easier for users to librarian is expressing just the right degree of to cut back on the services they provide browse information in interesting ways. Just difference. in order to engage in linked data projects. as Facebook surfaces relationships between Libraries are already trying to do too much people we know, so we might unearth new Will people be able to play an with too little. The movement towards linked connections on the library graph. Secondly, active role in the library graph? data and the library graph is likely to come the graph will make it very easy for other Most of the textual analysis needed to from those libraries that have the resources websites to talk about the works we’ve generate the library graph will be algorithmic, needed, and from funding groups that want to entrusted to libraries. Thirdly, and very but there are other ways of building the library invest.

importantly, users will be able to build their graph. In the digital age, people interact Sarah Bartlett is a freelance writer specialising own services on the library graph, for example with works of culture in a more public way, in technology and the information industry.

The greatest breakthroughs happen when knowledge is shared, giving thinkers and dreamers a clear view of each other’s ideas. When OCLC member libraries share their collective resources, ground-breaking ideas aren’t merely possible—they’re inevitable.

Together we make breakthroughs possible.

Knowledge connects the dots

Because what is known must be shared. Learn more at oclc.org/share

215498_Research-Information-Ad-EMEA.indd 1 10/1/15 9:47 AM