We have started a really new exciting venture in making scholarship available to everyone. We're starting from scratch. We're still working out details. And "we" means "you".
About 3 weeks ago things came to a head. Many people are frustrated with the lack of real, 21stCentury, access to scholarship. To the outputs of funded scholarship (somewhere between 300 BILLION USD and 1000 BILLION USD). And to the feeling of exclusion that everyone who isn't a powerful academic feels. The inability to contribute. The feeling that scientific research is a spectator sport for the lucky few who are in rich universities. Some of us swapped emails, initially from a sense of frustration.
But what emerged after about a week was the sense that we had exciting new opportunities to change the world in a bottom-up manner. We are increasingly empowered by the public technology of the web, and we are building on top of it. So we are creating a project, a philosophy, a toolkit, and collections of content. It's happening under the aegis of the Open Knowledge Foundation (OKFN) (mailing lists at: http://lists.okfn.org/mailman/listinfo ) as a new project (open-access ) and overlaps with open-bibliography and open-science.
When we use "open" we are committed to the Open Knowledge Definition: (http://opendefinition.org/)
"A piece of content or data is open if anyone is free to use, reuse, and redistribute it — subject only, at most, to the requirement to attribute and share-alike."
This applied to several aspects of Open Scholarship – Open Access, Open Bibliography (BOAI-compliant), Open Citations and Open Data (OABCD) as a start. If we have OKD-conformant information then, for the first time, we can start to see the power of machines and humans working together. We can use automation without having to seek permission.
We're currently calling it @ccess .
@ccess for all.
Why this strange formation? Because it's simple, memorable, and searchable on the web. It avoids the overloading of "Open". "Open Access" and more generally "Open Foo" is not an effective label for OKD-compliant information. By using a clearly defined string we can label information as OKD-Open, as BOAI-open. (Unfortunately even after 10 years of BOAI there is no simple automatic way of telling that a piece of information is re-usable without asking permission).
So what information can we AUTOMATICALLY label as @ccess – OKD-open? Not very much yet, but we expect that this will grow rapidly. Here's what we can do:
- Anything specifically labelled with CC-BY, CC0, PDDL licences.
- Datasets in CKAN labelled as Open
- Articles from BOAI-compliant publishers (the main ones being BMC, PLoS, EGU, and a few others)
- Data from bioscience databases (e.g. genomes, protein structures) (Bioscientists don't normally use licences but adhere to the Bermuda principles)
Here are some things that, by default, are not AUTOMATICALLY recognisable as OKD-compliant
- Depositions in Institutional repositories. Almost no content is labelled usefully for machines
- Self-archived manuscripts including arXiv
- Bibliographic collections
- Contents of Pubmed (except as above)
- Hybrid publications (95% is NOT OKD-compliant)
So a major problem is that we don't know what is actually OKD-Open and what we can use for modern automated scholarship. @ccess aims to change that.
We're going to build collections of OKD-Open material and label it as such. To show that's it's useful and a new approach to scholarship. Open to everyone, not just academics. Because @ccess is bidirectional – it's about building our principles and community so that we have a say in modern scholarship.
We're using our new bibliographic tools – Bibserver and Bibsoup – as an efficient means of collecting the information and labelling it. We're starting with disease as there are already active communities who want to start using the tools. Our first project is based on MALARIA. This idea been brewing for a year or more – Open Research Reports – but we've had to wait while we developed the technology. We've now got this, and we can collect and, very soon, label and annotate the information.
We are very grateful to Tom Olijhoek, Bart Knols and MalariaWorld for acting as the centre of this. Our first task is to find what Open information there is. This will require considerable human effort. Even Pubmed isn't able to label the documents which are OKD-Open.
We'll be posting more about this – so far we have 70,000 references for Malaria keywords ("Malaria", "Plasmodium", etc.) Mark MacGillvray has ingested them into a Bibsoup and we'll be working on them. The first activity will be to find out how many are OKD-Open. I'm guessing about 3%. That's the sad face of access to scholarly information. That's the amount that we can legitimately text-mine, re-index, use graphics from, etc.
But as people see the value of this they'll want more. And that's an important driver for making more information OKD-Open and labelling it.