collaboration – THATCamp CHNM 2009

Developing, but not overdeveloping, a collaborative space

Boone Gorges — Fri, 26 Jun 2009 02:18:00 +0000

For the past few months, I’ve been involved in the development of the CUNY Academic Commons, a new project of the City University of New York whose stated mission is to “to support faculty initiatives and build community through the use(s) of technology in teaching and learning”. This is no small goal, given the mammoth size and unruliness of CUNY: 23 institutions comprising some 500,000 students, 6,100 full-time faculty, and countless more staff and part-time faculty. The Commons – built on a collection of open-source tools like WordPress MU, Mediawiki, Buddypress, and bbPress – is designed to give members of this diffuse community a space where they can find like-minded cohorts and collaborate with them on various kinds of projects.

My work as a developer for the Commons pulls me in several directions. Most obviously, I’m getting a crash course in the development frameworks that underlie the tools we’re using. These pieces of software are at varying stages of maturity and have largely been developed independently of each other. Thus, making them fit together to provide a seamless and elegant experience for users is a real challenge. This kind of technical challenge, in turn, leads me to consider critically the way that the site could and should serve the members of the CUNY community. How do you design a space where people with wildly different interests and wildly different ways of working can collaborate in ways that work for them? By making the system open enough to accommodate many ways of working and thinking, do you thereby alienate some of those individuals who need more structure to envision the utility that the site could hold for them? How do the choices you make when developing a tool – decisions about software, about organization, about design – mold or constrain the ways in which the site’s uses will evolve?

In light of these varying challenges, there are a couple different things that I would be interested in talking about at THATcamp. For one, I’d like to get together with people working with and on open-source software to talk nuts and bolts: which software are you using, how are you extending or modifying it to suit your needs, and so on. I’m also very interested in talking about strategies for fostering the kinds of collaboration that the CUNY Academic Commons has as its mission. I’m also anxious to discuss more theoretical questions about the design and development of tools that are meant to serve a diverse group of users. In particular, I’m interested in the interconnections between the designer, the software, and the designer’s knowledge and assumptions about the desires and capacities of the end user.

Visualizing time

Brian Croxall — Wed, 24 Jun 2009 16:17:03 +0000

For the last two years, I have been very interested in visualizing data that emerges within my particular field: literature. This interest emerged as I read Moretti’s Graphs, Maps, Trees at the same time that I was experimenting with using GIS tools like Google Earth as a portion of the analysis in the last chapter of my dissertation. In my last year as a graduate student, a fellowship in the Emory Center for Interactive Teaching gave me additional time to begin experimenting with timelines. Timelines in literary studies were nothing new, but I wondered if it would be possible to have a class collaboratively build one in a manner similar to writing a wiki. The result was–in turn–a collaboration with Jason Jones (@jbj) where I coded a timeline, he designed an assignment, and his students created the data for a timeline of the Victorian Age. I’ve since had the chance to play with the tool in my own classes.

Jason and I both thought that timelines would be a fruitful subject for conversation THATCampers. And as many others have done, I thought I would share my original THATCamp proposal and then propose some ideas about where a discussion might go:

I would like discuss the different web-based tools and software that can be used to produce interactive and collaborative timelines. The presentation would involve demonstrating the different tools, showing the strengths and the weaknesses of each one, and producing a finished product. The tools would range from CHNM’s Timeline Builder to xtimeline and from Bee Docs Timeline 3D to the Timeline and Exhibit widgets that were developed in MIT’s Simile project. Having already spent some time with these tools, I think that the tools from Simile might be the most interesting to THATCamp participants due to their flexibility in representing data in multiple ways, including color coding events, sorting events, and with GIS data, as well as the ability to grab data from sources as diverse as a Google Docs spreadsheet or Twitter. Perhaps the best demonstration of the usefulness of a timeline would be to create–during the session/event–a timeline of THATCamp.

My current thinking:

As I’ve been preparing for THATCamp, I have gone ahead and evaluated as many of the timeline tools as I’ve had time for. I’ll be looking at another one or two tomorrow. I’ve gone ahead and created a spreadsheet listing the abilities of these different tools, along with some evaluation. Admittedly, some of the categories that I was using to evaluate the timelines stem from my deep involvement with the Simile widgets, and so the cases might not stack up as being completely fair to the competition.

Also, wanting to blend together both streams of data visualization that seemed valuable to me, I’ve also expanded on the original timelines that I designed for my courses by adding a Google Maps view this week. You can choose to either look at one view at a time or a dual view.

While a conversation could certainly be held about the different strengths and weaknesses of these different tools, most of the timeline tools that are available are going to be fairly easy for THATCampers to pick up and run with. The most complicated among them is the Simile tool, but I’ve heard there’s a fairly straightforward tutorial on building your own. Instead (or in addition to), I wonder if it could be possible to have a conversation about other possible research and pedagogical uses for timelines than those to which Jason and I have put them to use thus far. One obvious apporach would be to timeline a particular text (say, Slaughterhouse-Five) rather than a contextual time period. But what else could we do with timelines to make them valuable?

Moreover, I wonder if a discussion about visualizing time could be a part of a larger discussion about visualization that seems to be on the minds of other THATCampers (at least per their blog posts) such as Tonya Howe and Amanda Watson. How best can we use such visualizations in our research and/or teaching? At what point are there diminishing returns on such projects? Since these tools are relatively easy to learn (as opposed to programming languages), are they a good gateway tool for “traditional faculty” to begin comfortably integrating new technologies into their research/teaching? And, perhaps most broadly, what is the relationship between digital humanities and visualization

(I should meniton that while Jason and I proposed related ideas to THATCamp, this post is my own. So don’t hold him responsible for my shortcomings in expression.)

Archiving Social Media Conversations of Significant Events

Jeffrey McClurken — Tue, 23 Jun 2009 14:30:42 +0000

I’ve already proposed one session, but recent events in Iran and the various discussions of the role of social media tools in those events prompted this post.

I propose that we have a session where THATCampers discuss the issues related to preserving (and/or analyzing) the blogs, tweets, images, Facebook postings, SMS(?) of the events in Iran with an eye toward a process for how future such events might be archived and analyzed as well. How will future historians/political scientists/geographers/humanists write the history of these events without some kind of system of preservation of these digital materials? What should be kept? How realistic is it to collect and preserve such items from so many different sources? Who should preserve these digital artifacts (Twitter/Google/Flickr/Facebook; LOC; Internet Archive; professional disciplinary organizations like the AHA)?

On the analysis side, how might we depict the events (or at least the social media response to them) through a variety of timelines/charts/graphs/word-clouds/maps? What value might we get from following/charting the spread of particular pieces of information? Of false information? How might we determine reliable/unreliable sources in the massive scope of contributions?

[I know there are many potential issues here, including language differences, privacy of individual communications, protection of individual identities, various technical limitations, and many others.]

Maybe I’m overestimating (or underthinking) here, but I’d hope that a particularly productive session might even come up with the foundations of: a plan, a grant proposal, a set of archival standards, a wish-list of tools, even an appeal to larger companies/organizations/governmental bodies to preserve the materials for this particular set of events and a process for archiving future ones.

What do people think? Is this idea worth a session this weekend?

UPDATE: Ok, if I’d read the most recent THATCamp proposals, I’d have seen that Nicholas already proposed a similar session and I could have just added my comment to his….. So, we have two people interested in the topic. Who else?

Disciplinary Heresies and the Digital Humanities

sterling fluharty — Wed, 03 Jun 2009 09:48:10 +0000

Cross-posted at Clio Machine:

(This post is a continuation of some of the questions I raised in my original THATCamp proposal.)

Are the humanities inherently valuable, both in terms of the skills they impart to students and because the value of humanistic scholarship cannot be validated by external (often quantitative) measures? Or are the humanities experiencing a crisis of funding and enrollments because they have not adequately or persuasively justified their worth? These debates have recently resurfaced in the popular press and in academic arenas. Some commentators would point to the recession as the primary reason for why these questions are being asked. We should also consider the possibility that the mainstreaming of the digital humanities over last couple of years is another (but overlooked) reason for why questions about the value and worth of the traditional humanities are being taken more seriously.

As humanists have pursued academic prestige, they have long resisted the notion that intuition is important in their analysis and interpretation of texts. (Although I think this is more true in history than in literary studies, perhaps because the latter is considered more “feminine” than the former.) Humanists have distanced themselves from the notion that their subjective study is somehow speculative or irrational. They have been much more comfortable describing their work as imaginative and creative. What all of this posturing overlooks is the advances that cognitive scientists have made in explaining intuition over the last few decades. For instance, they have shown that humans are hardwired for instantly recognizing the emotions felt by other people. They have also explained how our minds are programmed to find patterns, even where none may exist. This tension was captured in the title of a recent book by a respected psychologist, Intuition: Its Powers and Perils. From this new perspective, then, intuition is taken for granted or ignored by almost all humanists but it is actually central to much of their work.

This debate over intuition raises important questions for traditional humanists working in the digital era. Would traditional humanists argue that their close reading of texts, which has become the hallmark of humanistic scholarship, is an example of this new concept of intuition at its best, since it is theoretically rigorous and excels at finding new patterns in old texts? Or will traditional humanists increasingly feel that their research methodology is threatened by what Franco Moretti calls “distant reading,” precisely because it risks exposing the limitations or perils of their intuitive models of scholarship? How would traditional humanists react if they knew that various digital humanists have searched Google Books to test the arguments set forth in some monographs and found them lacking when text mining revealed an significant number of counterexamples that were missed or ignored by the authors? These and other examples should get us thinking seriously about the advantages and disadvantages of relying so heavily on anecdotal, case study, and close reading research methods in the humanities.

Data and databases have become the holy grail of the new class of information workers. One recent books applies the term super crunchers to these data analysts. Recent articles in the popular press describe how large data sets allow trained professionals to find new patterns and make predictions in areas such as health care, education, and consumer behavior. In fact, we have probably reached the point this country where it is impossible to change public policy without the use of statistics. Even the American Academy of Arts and Sciences jumped on the statistics bandwagon when it launched its Humanities Indicators Prototype web sitelast year, presumably in plenty of time for congressional budget hearings. The fact that the humanities were the last group of disciplines to compile this kind of data raises some troubling questions about the lack of quantitative perspectives in the traditional humanities.

The humanities and mathematically-driven disciplines operate at almost opposite poles of scholarly inquiry. In the humanities, practitioners privilege crystallized intelligence, which is highly correlated with verbal ability. This has given rise to the idea that a “senior scholar” in the humanities accomplishes his or her most important work in their 50’s or 60’s, after a lifetime of accumulating and analyzing knowledge in their particular specialization. By contrast, the most mathematically-inclined disciplines prize the abstract thinking that characterizes fluid intelligence. This other form of general intelligence peaks in a person’s 20’s and 30’s. As a consequence, the Fields Medal, widely considered the highest award in Mathematics, has never been awarded to a mathematician over the age of 40. So if the digital humanities require young scholars to learn and excel at computational and algorithmic forms of thinking, we should be asking ourselves whether most senior scholars in the humanities will resist this as a perceived threat to their systems of seniority and authority.

Digital humanists have already written and talked quite a bit about how tenure and promotion committees have rejected some digital scholarship for being non-traditional. Further compounding this problem are what appear to be significant cultural differences. Almost all traditional humanists work on their scholarship in isolation; digital humanists collaborate often, sometimes because this is the only way to assemble the requisite technical knowledge. Traditional humanists distinguish their scholarship from that produced in the social sciences, which they often think lowers itself to the level of policy concerns. Digital humanists, by contrast, are almost universally oriented towards serving the needs of the public. And while traditional humanists place a premium on theoretical innovation, digital humanists have so far focused much more on embracing and pioneering new methodologies.

Digital humanists will have to seriously ask themselves whether their embrace of social science methods will be considered heretical by traditional humanists. Online mapping and work with GIS in the digital humanities is clearly borrowing from geography. The National Historic Geographical Information System, which maps aggregate census data from 1790 to 2000, is obviously influenced by demographic and economic analysis. The Voting America web site, overseen by the digital humanist Ed Ayers, builds on decades of studies in political science. Text mining is catching on as digital humanists adapt the methods of computational linguistics and natural language processing. What remains to be seen is whether the digital humanities will take this flirtation to its logical conclusion and follow the example of the computational social sciences.

All of this might sound quite unlikely to some of you. After all, most, if not all, of us have heard the mantra that the digital humanities is a misnomer because in ten to fifteen years all humanists will be using digital methods. But for that to be true, digital humanists will have to fall into the same trap as traditional humanists: believing that others will follow our example because the correctness of our way of doing things seems self-evident. But as we have seen, there may actually be significant differences in the ways that digital humanists and traditional humanists think about and practice their disciplines.

Let me conclude with a few questions that I would love to see discussed, especially as part of a session at THATCamp. Will the methodologies and mindset of the traditional humanities become increasingly anachronistic in today’s data-driven society? Will the digital humanities have to team up with the computational social sciences and create a new discipline, similar to what happened with the emergence of cognitive science as a discipline, if traditional humanists realize that we could radically change their research methods and therefore decide that we are too heretical? What if this departure from the traditional humanities is the only way for digital humanists to become eligible for some share of the 3 percent of the GDP that Obama has committed to scientific research? If digital humanists decide instead to remain loyal to traditional humanists, then what are the chances that young humanists can overthrow the traditions enshrined by senior scholars? Won’t traditional humanists fight attempts to fundamentally change their disciplines and oppose efforts to make them more open, public, collaborative, relevant, and practical?