If a library offers anything to 'the contemporary user', it is in the offers it makes to the dependencies or weak points of augmented reality or social networks. To AR, it offers the example (and the practice) of other methods of organising and thinking about relationships between knowledge than spatial ones. And, flipping the coin, to the invisible world of social networks, it offers real space with real resources. That is something libraries have always offered. When I ask, 'have we ever offered anything else?' I ask whether it is not true that the role of libraries has been to facilitate the social forms that have made use of them: societies, universities, communities. What does it mean to service a community in this way? Am I saying libraries should be community centers? Well, in a certain sort of a way, yes, I am. But quite what that means depends entirely on the needs of the communities served. And libraries tend to respond to that quite well. What I see at the moment is a momentum for communities created by their leap forward into the virtual medium: this is quite supportive. But it will still be helpful if these communities can put a foot to the ground from time to time: preferably somewhere where there are facilities to nurture their interests. Perhaps, if it goes well, they will choose to meet there often, to study, to exchange ideas, to gather material for themselves or for their colleagues or families. That still sounds a lot like a library to me. I suppose a metaphor would be to think of it more in terms of a 'drive in cinema'. There isn't so much of a need for the traditional apparatus anymore, but we want those who do choose to use the spatial aspect of the library to be supported by what they can plug in to. This means facilitating the social: libraries as a social keystone, as it were. There is still a potential for the physical aspects of libraries to be important, especially to social networks of all kinds.

As far as the augmentation of reality goes, the techniques of organising, relating and presenting information-artefacts will continue to develop and be useful in the world. Simple overlays into space cannot capture the many facets of reality that exist. There comes a point where visual metaphor becomes saturated, and no amount of advanced user-interface can cope. The parallel is with mathematics, or with the representation of multiple dimensions, where abstraction is far more efficient than visualisation. Yet the fields of user-interface and abstraction collide with the services offered in informational organisation. Can a single search box really cope with the cosmos? It's like pointing a telescope at the heavens: we may see deeply, but it takes another level of abstraction to see widely. Libraries build the systems that incorporate these abstractions. As a matter of fact, we probably ought to see them more like the theoretical abstractions of knowledge they are, and expose them to a bit more Kuhnian or Popperian rigour (dependent of course on the phases of the moon and the acidity of the paper): it is library science, after all, and should shudder through the occasional revolution if it is to adequately describe reality. Are the incremental changes and alternative systems we currently adopt sufficient, or is it fair to point out that we have islands of practice, sometimes speaking a common tongue, but still waiting on a comprehensive renaissance?
In fact, other developments threaten to outstrip this work. The semantic web will allow machines to take a greater role in deciding what we need and don't need to see, and what connections we will probably want to make. (I hope they throw a dash of the irrational, the serendipitous and the downright musty into their calculations). We would rightly welcome these developments as offering a far more systematic and universal set of bridges and tunnels between knowledge than the 'isolated practices' I mentioned above (though it could be worryingly impervious to arguments for change).  As automated as the process could become, I think it is important that people take steps to organise knowledge themselves, to make the decisions that disclose culture. These are sometimes irrational and unexpected. Sometimes they are wrong, sometimes even dangerous. But they are unconstrained by rational processes, in a way that 'automated semantic judgement' can only mimic. (It is worth asking whether such mimicry would be complex enough, juggling multiple cognitive and linguistic quantizations as to seem sufficient.) As Douglas Rushkoff notes we must either 'program or be programmed', and libraries can form a sensible interrupt in that librarians themselves interpret knowledge structures and make decisions about what matters, what is significant and what is not when they construct taxonomical links and other structures. The algorithms that settle page ranking in Google are secret, and though one supposes them to be automatic, they harvest the meaning created in human decisions to mention or not, to link or not, to like or not. There should be similar input in creating the landscape of information reality, too, and libraries offer a tradition of shaping this. Whether the tools libraries  have hitherto used are the right ones to continue with or not, the description of what they do in organising and providing access to knowledge remains important.

Simply teaching people how to engage with knowledge remains important too. The concept of 'information literacy' has been around for a while and now engages strongly with pedagogical ideas about how we learn socially and imbricated in the discourses of the larger-scale social networks we belong to.Libraries can be like a map that takes us above this to see how it works and how best to work with it. I was intrigued and interested to find out about 'Find the Future', a project to immerse 500 youngsters in New York Public Library and have them work together overnight to create a book. It's explicitly and deliberately presented as agame: but it turns research into (back into, I'd say) something playful and social, and it uses the library as a staging ground: it uses the library because it is a: somewhere where ideas can easily meet other ideas, where the structure of knowledge is not immediately determined by physical geography or the dictates of the scopic, and b: somewhere where you can get 500 kids in overnight to hustle some ideas together. In other words, it's just a library doing what libraries do.