Author: Marie Lambert 15

After weeks of agonizing over our final project proposal and potential research questions, the proposal itself fell together with unprecedented ease. Once we’d had a taste of thinking practically and narrowing our scope by doing individual project proposals, it seemed a natural step to integrate the elements of our own proposals that we’d found most interesting into the final project. What we have settled upon for a loose overarching question is What did Edward Hitchcock leave behind? Although, yes, it is rather broad, it’s been a useful umbrella under which we can arrange our major themes of time and legacy while still making use of the collection.

For the structure of the project, we ultimately did decide to take most of the individual proposals and connect them where relevant as sub-projects within the overall final exhibit. The idea is that we’ll have a kind of interactive website that lets viewers choose upon entering which of the ~4 sub-projects they’d like to explore first. Ideally, they’d eventually explore all of them, but we’d like to present all of the options on an equal footing, not determined chronologically or by “importance.” We want our audience to be able to engage with any part of the site and come away feeling that they’ve experienced a full narrative, not just part of one.

In terms of sites that would model something like that visually, while I can’t think of any specific examples, I feel that I’ve certainly encountered something similar before. I’m imagining a pretty simple homepage, maybe one that starts off with some basic biographical information about Hitchcock (or features that lead you through that) so there’s some kind of baseline knowledge, and then maybe leads into a page with four boxes: one for each of the mini-projects. You click on a box, and it takes you to that project page. Embarrassingly, the closest visual model I have for this is the block boxes you click on to select answers in Buzzfeed quizzes, but hey, at least those are intuitive.

I do at least have a clearer idea of what I’d like the iteration of my “mini-project” to be. I’ve harped on ad infinitum about how the Cocitation Network project in Signs@40 was in the inspiration for my mini-proposal and all that. But while it is a beautiful piece of data visualization, the kind of data that I’m looking to analyze doesn’t quite fit into the network model the same way. Signs@40 was examining the citations listed in all the articles in their journal over the past 40 years: so while the nodes of the network were from a variety of authors and time periods, they were all neatly catalogued by the single source of the journal itself. Not only does this give them a clear, demarcated set of data to examine, but it all comes back to the same point that they share in common: Signs@40.

With Hitchcock’s citations, things get a bit trickier. Hitchcock was pretty prolific and published a lot of his writings, which ranged in subject from geology and chemistry to the temperance movement. To completely measure the effect of his words, one would have to track whenever ANY of his many publications was cited: a task which not only seems very time-consuming, but kind of boring. So I plan to narrow the search to only tracking when his most important works were cited (exact criteria for that TBD). That helps a little bit with the problem of scope, but still there’s the question of having multiple sources for the citation network(s) it/themselves. Each work of Hitchcock’s would be at the center of its own network, which would necessitate a multitude of graphs, and more time. Unless, of course, we choose to put them all together in a single graph, which then raises the question of topic: should the original documents be all scientific texts, or a mix of scientific and religious? I’m more interested in looking at a medley of topics, but I’m also concerned that having too many different categories that the data points differ on could make it both difficult to model and analyze.

Data visualization of networks seems like a pretty popular field in the DH/tech community. This site has a lot of really well-constructed and easy to read data/network visualizations, so I spent some time here looking for inspiration.

My original idea of what the network would look like was similar to this LinkedIn visualization tool, which allows you to look at an aggregate of your connections on the site and see how they’re all collected to one another. Pretty much a standard network visualization. What I like best about it is the color-coding, which would come in handy if we end up doing a single Hitchcock citation graph of his top publications, as the colors could be used to designate each publication (and also offer a useful comparison between the reach of each one). What I don’t like so much about this graph is that it’s incredibly difficult to read, given all of the nodes that are included. Granted, with our project, it may not be so important to read each individual node, and more so that we at least have a visual sense of the mass of them. So actually, maybe not a problem at all.

Another network visualization I looked at was this one, which charts Google+ “Ripples” that extend as a way of sharing news in social media. The idea is that when something groundbreaking or headline-worthy happens, someone/some site is the first to post it, and then some number of people will see the news and share it over social media, in this case, Google+. Then, some number of people who see that “share” will go on and share it themselves, and the news will extend out like that, reaching an even greater number of people. I found this visualization particularly interesting out of all the network analyses I saw because it seems the closest in goal to what we’re trying to see in the Hitchcock project. In mapping where Hitchcock is cited, we’re trying to get a sense of how far his influence spread, how important and worth sharing his ideas were, both back then and now. Google+ Ripples is doing something similar in that the visual effect isn’t so much focused on who did the sharing but the sharing itself, how the ripples spread out from this original source or news event. It’s all about magnitude and direction of influence, which I think will be key in creating our own visualization of Hitchcock’s influences on the scholarly world.

One potential problem with these models is that most of them are made in Gephi, or some more complicated software. While we had a brief Gephi training, none of us on the team are particularly confident or even conversational in using it, and we eventually decided that in order to not waste time, we would likely use Tableau to create the final visualization. While Tableau, as far as I’ve seen, doesn’t do anything exactly like the Gephi network, it does offer a lot more options for graphs and other visual ways to display data, and is also a lot easier to manipulate.

But before we get to that point, we have to wade through a sea of data collection and processing: all of that and more exciting things to come next week!

After the past couple of weeks spent getting to know the Hitchcock Collection, learning about/experimenting with various digital tools, and generally agonizing about finding a research question or focus from which to begin our project, it was nice this week to go off on our own (mentally, if not physically) a bit and work on the smaller project proposals. This week gave us the chance to produce something that was concrete, something solid that we could point to and move around, instead of getting caught in the mire of repeated concept mapping that had slowed us down so much before when we tried to articulate potential projects.

Although in my last post, I articulated some lack of direction for where to go with my individual project, after reading through some secondary sources, I soon found a direction that related back to one of our first real questions about the collection: was Hitchcock important (in contributions to science, religion, etc.) or just Amherst important?

During my secondary source research, I found conflicting evidence in service of this question. Many of his colleagues and contemporaries touted Hitchcock as a groundbreaking scientist as well as an honorable and modest man—”one of America’s heroes,” J.P. Lesley claims in his biography of Hitchcock for the National Academy of Sciences. Hitchcock was, after all, one of the incorporators of the NAS, which certainly says something about his reputation and renown as a scientist. But was he really a “household term” in the world of geology, as Lesley suggests, ranking him above other internationally known geologists of the era? I did quick Google Ngrams search to see how Hitchcock stacked up to the other names Lesley dropped, and the results were not particularly encouraging on Hitchcock’s part.

But as Google Ngrams is a limited tool for measuring the true importance of a man’s impact on the world, this experiment raised more questions than it answered served as the inspiration for my individual proposal.

My proposal suggested approaching this question using social network analysis, which was one of the digital tools we learned about that fascinated me the most. In terms of data visualization, network analysis software really appeals to me because it allows you to give weight and value to the data you’re presenting, showcasing the dynamic nature of the network and connections involved and not just treating them as if they’re all equal. For someone (aka ME) who is still skeptical about the ideal of treating qualitative information as quantitative data, this tool seems like a way to combat some of my concerns about homogenizing the nuances of humanities research into equally flat little data points. The real inspiration for using this tool came from the project Signs@40, which uses a social network analysis to approach to create a comprehensive network of the sources that their writers have been citing for the past 40 years in their articles.

I see a lot of potential for using this tool to approach Hitchcock and his legacy, and proposed mapping when and where he and his works are cited in geological/scientific writings both from his time and now, and comparing them to see the reach of his ideas. Since the root of this project proposal came from the moderately facetious question we’ve all been asking from the beginning (“why should we care about Hitchcock?”), I feel that this line of inquiry is one that is at least moderately interesting to the rest of the group. Further, it also fits pretty nicely into the one general theme that we managed to come to a consensus on for shaping the project: time. Be it context or legacy or the inevitability of death, the broadness of the theme allows for a good amount of flexibility for individual mini-research questions within the project, and I think my proposed mini-project could provide a necessary perspective on Hitchcock.

Of course, I’m totally biased on the importance and relevance of my particular project, and the more that I think about it, the more that I really, reallllyyyyyy want to pursue it, to the point where I feel like I would be willing to do the whole thing myself, if this ends up being an arm of the larger final project. On a related note, I did feel a little bit of disappointment (not quite the right word? The feeling wasn’t quite that strong) when we’d finished our individual proposals and had to reintegrate into the group. I’m used to doing my research or academic work alone, and now that I’ve gotten so attached to the idea of this project, my first instinct was to begin work on it on my own and to bring it to the group when it’s finished. But I also don’t want to do that, because:

1. It’s probably not logistically possible for me to do alone. My individual social network analysis proposal was broad in terms of the kinds of questions that it could ask and vague on actual methodology for implementation. This was because while I think the social network analysis would be a great tool to use with this hypothetical data (citations), I have no concrete idea how I would go about finding that data. And if I did, there would be a lot of data to process, even if I severely limited the time intervals I drew from. Finally, while we did attend a workshop for how to use a network visualization tool, Gephi, I struggled to understand a lot of the mechanics of the tool, and could definitely use some help working with it.
2. I actually really like working with this group.I find that working with the other interns makes research and planning much more dynamic and exciting than it ever is when I’m on my own. I feel more invested in the project, more confident in its trajectory, and more enthusiastic and encouraged on a daily basis when I work with them. I don’t really want to go off on my own and make something that can just be pasted together with three other individual projects. I want to be involved in all of them, I want to learn as much as possible, and I want to see how the project can still grow and change in ways that I can’t even imagine at this point. And I can’t do that alone.

As much as I’ve hyped up my own proposed project, I was also really interested in everything that everyone else proposed as well. It’s fascinating to see not just where our individual interests gravitate towards, but how we go about asking certain questions and proposing to answer them. I keep thinking back to the learning-style assessments we did last week and seeing how each of us are expressing our individual learning-style preferences in the way that we’ve constructed these proposals. I’m actually really grateful that we’re all pretty different when it comes to that; I can’t imagine myself having come up with some of the things they have so far, and I’m glad to have to opportunity to approach this project from so many perspectives. I know it would be a ton of work, but I would really love to try to integrate as many of the individual proposals as we can into the final project, albeit perhaps adapting them a bit so that they fit together more smoothly. I’m looking forward to the next couple of days of brainstorming and planning, and feel that we’re really close to coming up with a concrete plan here.

Last (last) week, we had the opportunity to take a tour of Edward Hitchcock’s contributions to the Beneski Museum of Natural History, guided by the wonderful and extremely knowledgable Kate Wellspring. I’ve been to the Beneski Museum several times in the past, both for classes and on my own, and I thought I had a pretty good sense of the collection (HA, even writing that statement now, the naivete is painful) and what it contained. Mineral samples, drawings of the Oxbow, and of course the ubiquitous Ichnology Collection of Edward Hitchcock that lines the walls of the ground floor in massive slabs of footprinted stone.

And on those points, I certainly wasn’t disappointed: all of the exhibits and artifacts I’d perused before were still there, but my experience of them was markedly different than it had been in the past. Instead of wandering through and skimming the object labels that held my fleeting interest, I experienced a much more active and energetic interaction with the museum, thanks to Kate’s guidance. Kate not only seems to know just about everything there is to know about Hitchcock, she’s also incredibly enthusiastic about the topic of his life.

As Kate led us around the museum, stopping to give us details about the objects and exhibits that I had seen and passed by times before, I found myself much more engaged and attentive than I had been in the past. Granted, this change in my interest is most primarily explained by the fact that I’ve been living and breathing (nearly) Hitchcock for the past three weeks: in terms of audience, I/we the group was/were the ideal audience for Kate’s tour, given that we are actively and, one could say, urgently interested in learning about Hitchcock.

But outside of current situational elements that make me more engaged with the Hitchcock-related exhibits than I have been in the past, there was something about Kate’s energy and enthusiasm during the tour that made me even more actively interested than I think I would have been on my own. In addition to the basic explanation of each Hitch-related object she showed us, Kate would offer anecdotes from Hitchcock’s life, details on his relationships and personality, and her own commentary on what she thought it all meant. (For an example, when showing us all of the places Hitchcock traveled around New England to do geological surveys, Kate added that he was a very active man, despite his hypochondria, and would walk all over the countryside. “I think he probably needed to get out of the house and away from all those kids,” she quipped, referring to the full house of his and Orra’s children.)

I would say that this tour made Hitchcock feel like a real person to me, but I think I reached that milestone after reading his records of loans to his kids and some of his notes to Orra. Rather, I began to realize that Hitchcock was not just a real person, but a real person who I don’t and can’t really ever know: for all I’ve read about him so far, he can still surprise me; there is still learning to be had here, I just needed to be directed toward it by an authority, i.e. Kate.

I’ve been thinking about our/my experience with Kate at the Beneski Museum as an analogue for how our ideal exhibit would work: in the midst of the ocean of information contained in the Hitchcock Collection, our exhibit (or map, or data analysis, or combination of whatever) should be a guide, providing a unique and interesting narrative that the visitor could not just get from reading the Edward Hitchcock Wikipedia article.

On the other side of things, the exhibit should also leave room for the audience to explore (dare I say browse) and discover some things on their own. I keep having to remind myself that planning this project is not like planning a paper: there should be less of a rigid structure and more than one endpoint for the audience to come to.

Similarly, I continually have to remind myself that the research question that we will finally/soon embark upon doesn’t necessarily have to be a single or fixed question. After a week (this past one) of brainstorming about how on earth we were going to funnel all of our interests into a single question, this week we’re trying a more experimental approach, and allowing ourselves to explore what interests us and seems relevant in the collection and various secondary sources, with the hope that near the end of the week we can come together with more articulable ideas of what is more/less fruitful to research in-depth. I’m looking forward to the less-structured time, although I’m not sure I even know which direction to go off in. I still feel like I don’t know enough about the time and place and context of Hitchcock’s life (esp. in relation to the college, as they seem so inextricably linked in my and I think I can say the group’s mind[s]), and I think it might be most helpful to start there, with secondary sources and see where Hitchcock comes through the most.