The Profiler: Open Context's Eric Kansa
The Profiler: Open Context's Eric Kansa
Editor's Note: The Alamedan learned during the reporting of this profile piece that our subject - billed by the White House as an Alameda resident - actually lives in Berkeley, a city we don't typically cover. But we found his work interesting and thought you might too, so we're running our profile on Kansa anyway. Enjoy!
"Alternative academic" Eric Kansa starting working on Open Context more than a decade ago to promote access to archaeological research data that rarely saw the light of day. Last week Kansa, a researcher with the University at California, Berkeley School of Information, was honored for his efforts to make more of that data public. The Alamedan caught up with Kansa just as he was getting on a plane to Italy for a dig.
You've been described as an archaeologist and "computer geek." Tell us a little about yourself.
I grew up in Livermore, Calif. and went to undergrad at (the University at California, San Diego). Product of public schools, then I went to Harvard for grad school and got my PhD in 2001. I started programming as a kid with a Commodore 64, then did a little in college and grad school. While I've always had great fun with computer, I'm definitely not a great programmer. I need to program for publishing archaeological data to keep costs under control, since I need to be able to understand and manage our system (Open Context) and keep our organization very lean.
What is an "alternative academic"?
An "alt-ac" is a term reflecting some of the changes introduced in academic life by digitization. There's lots of need for people who understand information and information technology and have expertise in different disciplines. Lots of people with expertise and advanced degrees in a field in the humanities (art history, literature, history, archaeology, etc.) who work on computing projects are alt-acs.
Alt-acs often have an uncomfortable fit in existing university structures. They are typically not hired as normal faculty. Faculty jobs are still "publish or perish" and still largely oriented toward very narrowly defined kinds of publication of peer-review articles or books. Instead, alt-acs often make data, software, or other kinds of digital media, and that doesn't fit traditional faculty expectations. Nevertheless, there's more and more recognition for the need for computational systems to support research and teaching in the humanities and social sciences. Alt-acs combine their expertise of a discipline and their information technology skills to build these systems. That involves really interesting research problems and thinking, so alt-acs have to be great researchers in their own right, even though they are usually not faculty.
The dark side of the emergence of alt-ac career paths is that these career paths often have very little job security. They are often financed with "soft-money" (temporary grants), and that makes them basically disposable, temporary employees. In that sense, alt-acs have some affinity with adjunct faculty (faculty hired as temps to teach a course). Adjuncts typically face very harsh employment terms (very bad pay, little support, no benefits). But because alt-acs typically have pretty rare combinations of technical skills and non-technical expertise, I suppose they are paid and treated somewhat better than most adjuncts.
What was the inspiration for Open Context? And how did you get it off the ground?
We actually never intended to build our own system. We started our nonprofit (the Alexandria Archive Institute) to help preserve and share archaeological data, because these data are the only way we'll ever understand our history and origins. If these data disappear through neglect, our past is erased. It's that simple. However, we wanted to use someone else's computing infrastructure to keep our costs down. We were originally going to use a system at the University of Chicago, but that system developed in a different direction and couldn't meet our needs. We're still closely collaborating with them, but we needed something that fit better with the World Wide Web. All of this got off the ground thanks to our first grant, from the William and Flora Hewlett Foundation, and we're still so grateful for the opportunity this gave us.
Can you describe what Open Context does, and the project's goals?
We think the Web is awesome for sharing scientific data. If you follow some simple principles of Web architecture, you make it really easy for people to combine data from different places for mashups, new visualization, and new research. That's the main value we provide with Open Context. The data we publish works well with the context to the World Wide Web and can be easily networked with other awesome data published by museums, other scientific domains, etc. It also means anyone can use our data in their own apps (we also require Creative Commons licenses that remove standard copyright restrictions to make it legally possible to reuse the data we publish). Our use of very simple Web architecture approaches also lets us work well with the California Digital Library, the main data archive of the University of California. They're a fantastic group of experts that help preserve the data we publish, which is a key responsibility.
While we're really excited about Web technologies, lots about what we do with Open Context is not really about technology. We're also trying to promote more openness and collaboration in the research community, and also greater positive recognition for researchers who share data back to the larger community. This incentive issue is a big challenge, since many archaeologists hesitate to share their data since they worry they won't get any credit for all their hard work. That's why we're making sharing data a form of peer-review publication. Since it helps to both make higher-quality more usable data and it helps give researchers the right kinds of kudos.
What's up there so far?
We've published all sorts of data from many sites across the world. One of our first publications comes from professor Martha Joukowsky's excavations at Petra, a famous world heritage site in Jordan. It was capital of the Nabateans, a people that got rich controlling trade in spices and other exotica over the Arabian peninsula. We've also published two major excavations in Turkey that span the Neolithic (time of the earliest farming) though the Iron Age. Many of the datasets describe animal bones that were food refuse from ancient settlements, and these help us understand what people ate, but also other aspects of how they managed animals (were they keeping animals for wool, for draught, for milk?). We're currently working on publication projects to help identify ancient trade networks and another project documenting an ancient Etruscan site, Poggio Civitate, who's excavation is directed by professor Anthony Tuck of the (University at Massachusetts,) Amherst.
How receptive have your peers been to the project?
My peers recognize the need to better share and preserve data, that's really clear. They also love the idea of having editorial and peer-review systems to help promote greater quality of data. Some of the ideas we helped to pioneer in the archaeological context, of using creative commons licenses to give legal permission for reuse of information, for example, are now widely used by many projects. We've now got a rapidly growing and dynamic community of researchers active in making open data, and that makes our work so much more interesting! It's a lot less lonely than it was 10 years ago when we started!
Are there other fields where academics are taking a similar approach? And if so, what have the results been?
Data sharing and preservation is a huge issue across the sciences and humanities. Most efforts focus on library like "digital archives" which are absolutely necessary. We think that while archives are needed, they aren't always sufficient, and that's why we think of ourselves as a "publisher" (though free and open access). There are important questions about how we understand and promote quality of information and how we make data easy to understand to enable reuse to be credible. These are hard problems, and there are some other efforts that are also exploring ways to make higher quality, more intelligible data via similar models.
You were honored by the White House for your work last week. Can you tell us know you attracted their attention?
I've been at this for 10 years, and working in archaeology. Archaeology may seem kind of obscure, and it gets less attention than "big science" fields like genetics. But archaeology spans many disciplines, archaeologists work with everyone from anthropologists, to botanists, to zoologists and geologists to architects and chemists. In archaeology, and in archaeological data sharing, you end up interacting and talking with people from many, many fields. That helps get some attention, and I think our experiments with peer-review of data really resonate, since may fields have similar data quality and intelligibility problems.
What's next for you, and for the project?
We'll we're trying to get out of "startup" mode and become much more of a regular part of the expected way researchers communicate. A huge part of this is to find a sustainable way to finance our work in publishing open data. The hard part of data, is that if you start restricting access or create intellectual property barriers by charging for data, you end up breaking and ruining lots of the value of data. Scientific data works best as a "public good," free for anyone to enjoy and reuse. And like most public goods, that requires some form of public financing to create.
But let's face it. The public (taxpayers) fund research, including archaeology. Currently though the vast majority of that research results in private intellectual property, in the form of articles that cost $35 or more to even look at. These articles don't even have the raw data on which conclusions are based! So, if we don't support open data in the science, we're throwing away scarce public money, and worse, public financing of research ends up to ever more limited hands of the big commercial scientific publishers. So, whats next for us is going to be continued advocacy to help make sure the public actually benefits from their support of science!