On the Googles, Common Knowledge gets more than 25,000,000 hits. It’s a market research company, a scholarship foundation, a non profit fundraising firm, and in its inverse as Uncommon Knowledge part of a conservative group site, and an interview series at the Hoover Institution.
We can take the Wikipedia entry:
Common knowledge is what “everybody knows”, usually with reference to the community in which the term is used.
or we can take an anti-plagiarism guide to heart:
The two criteria that are most commonly used in deciding whether or not something is common knowledge relate to quantity: the fact can be found in numerous places and ubiquity: it is likely to be known by a lot of people. Ideally both conditions are true. A third criteria that is sometimes used is whether the information can be easily found in a general reference source.
But what’s kind of stunning is how hard it is to pin down what common knowledge really is. There’s 25,000,000 pages to get through to see what everyone thinks it is…which is why we default to using the page that the highest number of other people have linked to.
And this is the easy stuff. Common Knowledge is a human generated phrase. It has web pages that contain it, pages that are in common languages (English, for example, as well as HTML). Those pages are found using common naming systems (domain names and dotted quads) and transferred using common protocols.
Science is harder. Most of the knowledge in science is uncommon. It’s published in papers, not web pages. It’s got citations, not hyperlinks. It is, at least, generated by people, which means we can do some relevance anyway, and create libraries and structures for search. It’s imperfect, and compared to the Web, it’s terrible, but at least it sort of works.
Data in science is worse. Increasingly it is machine-generated, especially in the life sciences. The web we have works for documents, not data. There are no domain name systems for the data concepts of science, no markup languages commonly used. Annotation is time consuming and unrewarded. There is almost no common knowledge there.
So this blog is going to be about that aspect of common knowledge. It’s not just things that we all know. Or things that someone knows somewhere. Or things that someone somewhere has linked to. It’s about getting what we know into a knowledge network – a new addition to Benkler’s layers and Zittrain’s ideas, one that builds in concepts of end to end and generativity for knowledge.
Because we don’t just need to find stuff related to knowledge – we need to be able to compose new knowledge out of old knowledge. We need the cost of model building to be as close to zero as possible, and the ease to be such that pretty much any scientist can build, test, discard, and connect models as easily as writing Facebook applications.
I have some opinions on how to get to that world, most of which come from my experience at Creative Commons, where I run the Science Commons project.
First, we need the things we already know to be legally open and technically available for transformation. We need some new infrastructure. We need tacit knowledge, encoded in tools, to be available on the web for anyone who is qualified, and not just for the elite few.
And we need to start thinking about knowledge not as some abstract mass of links, but as something we need to treat differently. It is made of ideas, which are not pig iron, but something ineffably different.
Ideas want to be connected. When they’re connected, and published openly, and the cost of finding them and re-connecting them is the cost of connecting to the network, then hopefully we’ll have another definition to attach to the Google results.
In the interim, I’m going to blog here. I will be going on about issues like open access, tacit knowledge, biological materials, open source data integration, crowdsourcing science, why web 2.0 and science have an uneasy relationship at best, and anything else that seems of interest. I may also talk about barbecue.
Thanks to the ScienceBlogs folks for letting me join up. It’s good to be here.