The speaker had started the afternoon with a quote from Ian Rogers, ‘Losers wish for scarcity. Winners leverage scale.’ He went on to eloquently, if somewhat bluntly, make the case for exposing data and discuss the importance of making it available in a useable and re-useable form. In particular he discussed the sophisticated re-analysis and mashing that properly exposed data enables while excoriating a number of people in the audience for forcing him to screen scrape data from their sites.

All in all, as you might expect, this was music to my ears. This was the case for open science made clearly and succinctly, and with passion, over the course of several days. The speaker? Mike Ellis from EduServ; I suspect both a person and an organization of which most of the readers of this blog have never heard. Why? Because he comes from a background in museums, the data he wanted was news streams, addresses, and lat long for UK higher education institutions, or library catalogues, not NMR spectra or gene sequences. Yet the case to be made is the same. I wrote last week about the need to make better connections between the open science blogosphere and the wider interested science policy and funding community. But we also need to make more effective connections with those for whom the open data agenda is part of their daily lives.

I spent several enjoyable days last week at the UKOLN Institutional Web Managers’ Workshop in Aberdeen. UKOLN is a UK centre of excellence for web based activities in HE in the UK and IWMW is their annual meeting. It is attended primarily by the people who manage web systems within UK HE including IT services, Web services, and library services, as well as the funders, and support organisations associated with these activities.

There were a number of other talks that would be of interest to this community and many of the presentations are available as video at the conference website. James Curral on Web Archiving, Stephanie Taylor on Institutional Repositories, and David Hyett of the British Antarctic Survey providing the sceptics view of implementing Web2.0 services for communicating with the public. His central point, which was well made, was that there is no point adding a whole bunch of wizzbang features to an institutional website if you haven’t got the fundamentals right: quality content; straightforward navigation; relevance to the user. Where I disagreed with his position was that I felt he extrapolated from the fact that most user generated content is poor to the presumption that ‘user generated content on my site will be poor’. This to me misses the key point: that it is by focussing on community building that you generate high quality content that is of relevance to that community. Nonetheless, his central point, don’t build in features that your users don’t want or need, is well made.

David made the statement ‘90% of blogs are boring’ during his talk. I took some exception to this (I am sure the situation is far, far, worse than that). In a question I made the point that it was generally accepted that Google had made the web useable by making things findable amongst the rubbish but that for social content we needed to adopt a different kind of ‘social search’ strategy with different tools. That with the right strategies and the right tools every person could find their preferred 10% (or 1% or 0.00001%) of the world’s material. That in fact this social search approach led to the formation of new communities and new networks.

After the meeting however it struck me that I had failed to successfully execute my own advice. Mike Ellis blogs a bit, twitters a lot, and is well known within the institutional web management community. He lives not far away from me. He is a passionate advocate of data availability and has the technical smarts to do clever staff with the data that is available. Why hadn’t I already made this connection? If I go around making the case that web based tools will transform our ability to communicate where is the evidence that this happens in practice. Our contention is that online publishing frees up communication and allows the free flow of information and ideas. The sceptics contention is that it just allows us to be happy in our own little echo chamber. Elements of both are true but I think it is fair to say that we are not effectively harnessing the potential of the medium to drive forward our agenda. By broadening the community and linking up with like minded people in museums, institutional web services, archives, and libraries we can undoubtedly do better.

So there are two approaches to solving this problem, the social approach and the technical approach. Both are intertwined but can be separated to a certain extent. The social approach is to link existing communities and allow the interlinks between them to grow. This blog post is one attempt – some of you may go on to look at Mike’s Blog. Another is for people to act as supernodes within the community network. Michael Nielsen’s joining of the (mostly) life science oriented community on FriendFeed and more widely in the blogosphere has connected that community with a theoretical physics community and another ‘Open Science’ community that was largely separate from the existing online community. A small number of connections made a big difference to overall network size. I was very happy to accept the invitation to speak at the IWMW meeting precisely because I hoped to make these kinds of connections. Hopefully a few people from the meeting may read this blog post (if so please do leave a comment – lets build on this!). We make contacts we expand the network – but this relies very heavily on supernodes within the network and their ability to cope with the volume.

So is there a technical solution to the problem? Well in this specific case there is a technical problem to the problem. Mike doesn’t use Friendfeed but is a regular Twitter user. My most likely connection to Mike is Brian Kelly, based at UKOLN, who does have a Friendfeed account but I suspect doesn’t monitor it. The connection fails because the social networks don’t effectively interconnect. It turns out the web management community aren’t convinced by FriendFeed and prefer Twitter. So a technical solution would somehow have to bridge this gap. Right at the moment that bridge is most likely to be a person, not a machine, which leaves us back where we started, and I don’t see that changing anytime soon. The problem is an architectural one, not an application or service one. I can aggregate Twitter, FriendFeed or anything else in one place but unless everyone else does the same thing its not really going to help.

I don’t really have a solution except once again to make the case for the value of those people who build stronger connections between poorly interconnected networks. It is not just that information is valuable, but the timely delivery of that information is valuable. These people add value. What is more, if we are going to fully exploit the potential of the web in the near term, not to mention demonstrate the value of exploiting it to others, we need to value these people and support their activities. How we do that is an open question. It will clearly cost money. The question is where to get it from and how to get it to where it needs to be.