I just returned from a small Digital Humanities Conference at Rutgers at which everyone agreed that “The Digital Humanities” should not exist. It was decided that next year’s Digital Humanities conference would be on this theme. Kidding. I saw some great things, though, chief among them the work of Dan Cohen and Gregory Crane, two people who have not only thought about the problem of digital humanities a lot but also made something, and in each case something fantastic.
Dan Cohen is the man behind the Center for New Media and History at GWU, and in particular the mega-super-awesome tool Zotero. Cohen gave an overview of Zotero development, and talked in some detail about the advantages of making it open source, not only for all the right reasons, but also because of what it has done for development of the tool. I mean really: Zotero is so clearly head and shoulders above most undertakings of this sort in the humanities, and I think a big part of that is the openness and inclusivity of the project. It also makes it dramatically more flexible than most digital humanities projects, which are mostly modeled on the monograph–doing one thing well and in detail, and along the model of the individual scholarly project.
Zotero, by contrast, is quickly becoming a “platform” technology which anyone interested in “digital” anything can think about extending in ways that Cohen or his close associates may have had nary an inkling of prior to seeing it emerge. Add to that Dan is embroiled in an “interesting” lawsuit with Thomson-Reuters which I do not envy, but am eager to see resolved in favor of open scholarly tools.
Gregory Crane is the man behind the Perseus Project, one of the longest standing digital database projects, and an attempt to intelligently digitize and mine the Classical Greek and Roman corpus. Crane talked a lot about all the different kind of research that have become possible with such a corpus, spanning the gamut from basic computer science questions about intelligent mining algorithms to linguistic “treebanks” that can make texts talk to one another in all kinds of new ways. Crane followed my talk by explaining that he had one published article in an anthropology journal, but that the reaction to it was so savage he never tried again đ
He also asked us, playfully, “who is the most important classicist of the 20th century”? The answer: Ali Hoseyni Khamenei. According to Crane, Khamenei’s regime was an implementation of Plato’s republic. I have no idea whether to believe him, but his point was that Classicists have lots of ways to recover bits and pieces of the classical heritage in greek and latin, but almost none to do so in persian, arabic or urdu. And maybe ten people in the world who can speak the relevant combination of languages. He made a compelling case that we now have the technology to research such a question if only scholars would turn their efforts in that direction.
Crane and others were also well aware of the need to make scholarship open access, and to do so now, but also the need to leverage the existing legitimacy of the credentialling publications so that younger scholars don’t have to worry about getting tenure, or getting respected, if they want to make their sccholarship open access. Crane made a very strong case for the necessity of open access for another reason: the computability of our work. Which is to say, that the research article is no longer the end of the game if we can now start thinking of ways to mind, analyze or process large databases of scholarly literature, for whatever reasons, as objects not only to be read and digested, but used to generate new kinds of knowledge. Crane added that this approach opens up research possibilities far beyond graduate work to create “participatory humanities”— the idea that if you create good platforms you can enable research at all levels, not just the highest. He gave the example of how undergraduates learning greek and latin can make the diagramming of a sentence—a routine part of understanding how grammar works in a language–into a contribution to a treebank, or a contextually sensitive digital version of a text….principles of participation that are much more familiar today in the wake of Web 2.0 and wikipedia.
I’ve been using Zotero for about a year and half now and it is indeed promising. I would caution anyone intrigued by it that 1) while it can be customized, you really need some programming skills to do so & 2) features tend to appear much more slowly than promised by the developersâthere is, for example, a potentially great “hierarchical records” feature that was to appear over a year ago but that seems to have been lost in the pipes.
Zotero is great for citation harvesting, perhaps unequaled. I stopped using it intext citation some months agoâit was slow and buggy at the time and I can’t comment on any possible improvements in recent monthsâwhen I became competent enough with LaTeX to quit word processors (Zotero has very good but not quite perfect BibTeX export). As a broader research tool it doesn’t yet approach the wealth of features available via the CHNM’s older Scribe application such as the ability to build chronologies and various sorts of lists. There has been some mention on its forums of extending Zotero to include those capabilities but I don’t know if there is any movement towards making them a reality.
Agreed… Zotero’s greatness isn’t about user-friendliness. It’s not End Note, and that’s a good thing in my opinion… but you can’t beat the ease with which it adds things to a bibliography and preserves bibdata in things like wordpress or html… it’s a major step towards the semantic webdream, for those who dream of semantics (I dream of a pragmatic web, but no one is dreaming with me).
It does build timelines though… or did Scibe do something more than what the timeline plugin for Zotero does?
Cohen also showed some integration with google earth that was pretty impressive: an archive of documents from 9/11 tagged and geo-located which allowed you to look at interesting maps that cross-referenced the content of texts (“I was praying that day” was the example he used) with location. The ability to do this does indeed require some skills, but I think the harder part is coming up with meaningful research questions that can take advantage of it…
Yes, you’re right, Zotero does integrate with CNMH’s SIMILE application. At this point it is easy (but not very useful, at least to me) to display a timeline of publications/manuscripts by date. You can also hack a little to insert other sorts of data: http://forums.zotero.org/discussion/2523/
I guess I just prefer Scribe’s more traditional interface to SIMILE’s, which I find clever but confusing. I also haven’t figured out how to do a range of dates with Zotero though I only tried so hard. What I really love about Scribe is its capacity for linking pieces of a chronology to a quote or quotes, each of them in turn linked to a record for their source (which, if secondary, can be linked to a record for the primary source from which it draws). You can build biographies in the same way. I imagine all of this is possible via Zotero if there is any will to make it happen.
I think the new or upcoming version of ATLAS.ti is supposed to have a GoogleEarth feature like that you describe. I think it would be useful to help “spatialize” your understanding of whatever you’re working on, though it was probably just thrown in as a sexy gadget.
It did occur to me that I forgot to mention that Zotero has what I see as a major design flaw for Mac users (which is odd given that it apparently has and continues to be developed largely on Macs). The intended workflow presumes the ability to view PDFs in your browser, and this is impossible for Firefox on an Intel Mac. (I think you can do so if you run Firefox under Rosetta with the accompanying loss of performance.)
Dan Cohen was here at IU recently taking the digital humanities crowd for a tour under the hood of Zotero and telling the history of its development. I was very impressed by the overall effort and am starting to put it to good use. The almost-here features that she showed us were very impressive, as was the basic concepts at work in its architecture.
Thompson-Reuters’s suit against George Mason University should be enough to convince Endnote users to abandon ship.