Fri 26 Oct 2012
In case you all didn’t know, today is the last day of 6th annual Open Access Week. I’ve been very busy lately with software updates to Open Context, an open access data publishing service for archaeology, so I haven’t had a chance to cover archaeology developments as much as I would like.
However, I recently submitted a paper about open access in archaeology that was accepted to a special issue of World Archaeology. Like most of archaeology’s mainstream, conventional journals, World Archaeology is a closed, toll-access venue. Participating in this kind of publishing is not ideal, since it perpetuates a high cost scholarly communications system that impedes access, opportunities for new research (especially text-mining), and uses public research funding to, in effect, subsidize the creation of private intellectual property. Most people who read blogs like this know the story.
However, I decided to publish there because I thought it important to reach a different audience, one that does not follow blogs or discussions about scholarly communications. Mainstream archaeology needs to participate in arguments about open access, and needs to understand why open access is an important issue. The highly problematic stance of the Archaeological Institute of America serves as a case in point (see Ancient World Online, Doug’s Archaeology, and this letter Jessica Ogden wrote that I co-signed).
My paper introduces some of the basic arguments in favor of open access to a mainstream archaeological audience. None of these arguments are especially new to folks following the issue on the Web, but I think it’s useful to enter into a conversation with other members of our profession less familiar with the topic. Also, the paper introduces ideas about Open Data, a related area of innovation in researcher communications.
One area that I touch on in this paper is an issue of “open architectures.” It’s an emerging area of interest to me, and one where I’m still formulating some thoughts. But I think it’s as important an issue as licensing and access for the future of archaeological communications. It directly touches on the issue of centralization and decentralization in archaeological information systems. Centralization can save money, and has other efficiencies, especially in performance for searches and analysis. However, it can also reduce and constrain freedom and innovation, since implementation choices, technologies, interfaces, and development directions are under control of one group with its own set of agendas. Decentralization, on the other hand, allows wider participation and choice in development strategies. However, decentralization can dilute resources too widely, leading to lots of varied, under-supported, and poorly coordinated implementations. Decentralized systems can also have performance and user experience problems. For instance, a distributed search across lots of different systems involves many trade-offs. It is only as fast as the slowest participant in the distributed networked offering search results.
I wonder about ways we can reconcile the polar opposites of centralized versus decentralized systems. When you think about it, the distinction between centralization and decentralization depends on how narrowly or broadly you see your environment. In archaeology, the big centralized systems are the Archaeology Data Service repository and the tDAR repository. But, in the larger world of scholarly communications and scientific data sharing, these are just two of a wide number of systems serving different constituencies. Which gets me to the point of this post.
Openness and interoperability are vital because even big and centralized systems (within the scope of archaeology) are still small when one considers the bigger picture of the world of research. This is particularly important for archaeology, because archaeology is inherently multidisciplinary. We will always need to link and reference data and other content from other disciplines. Those disciplines will have their own data systems and repositories. So we can’t escape the need to think about building distributed systems.
Can we find ways to have our cake and eat it too, and enjoy benefits of both approaches while mitigating their problems? I think the Pelagios approach may point to a good direction. In Pelagios, several distributed systems offer data according to a simple common standard. The Pelagios team harvested these data and built a centralized index facilitating fast and efficient search and retrieval of resources from these different collections. Pelagios is also interesting because it achieves much with very little effort and cost and its participating collections have such widely varying disciplinary themes and emphases (only some of which were archaeological).
This is an important point. Centralization is indeed useful, but people will need to define the focus of centralization in very different ways, and only sometimes will the need to centralize align with traditional disciplinary boundaries. In a later blog post, I will follow up with more on centralization versus decentralization. But for now, please enjoy a pre-print draft of my paper on open access for World Archaeology.