Digital History — Spring 2010

February 23, 2010

A question (okay, many) for Professor Walsh…

Filed under: Uncategorized — ths117 @ 3:43 pm

Hello Professor Walsh,

Along with Franco Moretti’s piece in the New Left Review, reading the intro and conclusion to the Fall 2009 issue of Representations stretched some old comparative literature muscles that I haven’t used in quite a while! As a historian, I’m fascinated by the larger social context of these modes of criticism — specifically the symptomatic reading as a result of the 1960s Marxism-inflected imperative towards ideological demystification, etc. (Conclusion 143). I’d be interested to hear what you think about the influence of computers/ the Internet, and free-market/free-information ideologies as perhaps a similarly important influence on “how we read now.” Although the intro piece does touch upon literary critics’ possible use of computers as a means of analysis and taxonomy that gives more attention to rich surface information (17), there’s not much attention given to the impact of the more democratic creation and dissemination of  creative and critical texts that the Internet enables. Could an distinctly expanded (though still exclusive) arena of textual production and engagement have contributed to the decline of the heroic model of criticism? Or has the model of the heroic critic simply moved elsewhere? And do the activist/re-activist hallmarks of Web 2.0 — comment threads, forums, Twitter feeds, wikis — have an impact on this new version of symptomatic reading that envisions itself as reading “beside” the text? (Conclusion 145)

Finally, any thoughts about the impact of the explosion of amateur-created textual media on the work of either historians or literary critics? I concede without debate the fact that much of the Internet’s user-created content is of minimal merit — even as I muse that historians and scholars of cultural studies have done good work with fragments of low culture. Nonetheless, much of what’s produced may rightfully be below the radar of  literary scholars. Still, to appropriate Moretti’s term,  I see the possibility of  a “wave” of text threatening to overwhelm us in the era of digital abundance! To take a different tack though, despite Mayer-Schonberger’s assertions to the contrary, I do worry about the plasticity of digital records — the ability of persons to delete or alter the text and record of reaction in ways unnoticed by the casual observer. Issues of “edition,” authority, and citation become problematic in the digital age.

Apologies for a long set of rambling questions!  I will be attending the entire session on Thursday (barring any unforeseen schedule snafus), and I look forward to your response to any of this that might be relevant to your own work.


February 18, 2010

On the Victorian Women Writers Project, Past and Present

Filed under: Uncategorized — ths117 @ 7:16 pm

Aside from the idea that GIFs are CompuServe’s proprietary image format (who knew?), one of the most interesting aspects of Perry Willett’s article on the VWPP was a chance to see how the commitment to preserving women’s writing shaped the scope of the project. Willett says that texts written by women are among the most vulnerable, both as a result of physical degradation and because the marginality of their authors means that they have less inherent market share and therefore less chance of being reprinted through conventional means. I wonder how ~15 years of change in academia might have affected these conditions — whether the marginality of these texts and their unusual authorship might now be considered a virtue, pushing them to the “head of the line” for new models of preservation and dissemination. The VWWP itself is to some degree a historical artifact then, signifying changes in academic awareness of the importance of history at the margins, and of the potentials of digital history for de/re-centering the canon of important texts.

Of course, the VWPP is more than a historical artifact — it’s also a useful database which is being revamped and expanded to take advantages of the technical advances since 1996. I look forward to hearing about the goals of the project tomorrow! I am particularly curious about how the VWPP 2.0 will factor in some of the concerns about metadata that were raised in the readings for last week. Also, I wonder whether there will be any attempt to provide an interactive element to the new project — whether crowd-sourcing information, forums for discussion of these texts and apposite context, or a “behind-the-scenes” sort of blog for visitors to see how the project continues to evolve.

These are, of course, elements to keep in mind as I consider my own project for this class — a course website for a class on American intellectual history in the nineteenth century. Since I’d like to integrate the website into students’ experience of the class, obviously I need to make sure that there are interactive elements that guide them through the material and allow them to record their reactions, questions, and interpretations.  It would be interesting to find a way to feature user-created content, as well. Do you think a “class blog” would be useful or merely a gimmick? Alternatively (or in addition), it would be great to have them do some reading of the class texts online, using the CommentPress software that Kathleen Fitzpatrick uses for Planned Obsolescence.  This would let me pose reading questions to them “in the margins,” and to have them respond with their own marginalia. I need to think more about the benefits and drawbacks of this technology, but I think it could have distinct uses for helping the undergrads learn how to read a text in the same way that historians do — a major bottleneck identified by the History Learning Project.

February 11, 2010

Comparative archives, digitization, and a project inspired by lunch at Panda Express

Filed under: Uncategorized — ths117 @ 10:00 pm

For this assignment, I wanted to design a project that I knew would show “change over time” in a given period — so, using Kristin Hoganson’s Consumer’s Imperium as a guide, I figured that I would use the Nineteenth-Century United States Newspapers Database to see how the advertising of Chinese restaurants had changed during the time period covered by the newspapers. Ideally, the database would allow me to see how often Chinese restaurants were mentioned in these newspapers; where and when the first mentions occurred; the tone of the advertisements or journalism regarding Chinese restaurants; and any changes in the advertisements over time (size, reputation, etc).

My original idea in using the Nineteenth-Century United States Newspapers Database would be to test how good the database was at registering text in classified ads — material that might be considered “unimportant” by digitizers not thinking historically, or material filled with graphics and font changes that might puzzle OCR scanners, resulting in a high degree of inaccuracy. The results that I got from typing in several search strings (“Chinese restaurants,” “Chinese restaurant,” “Chinese cookery,” etc) showed me that either I was correct in diagnosing classified ads as a weak point of the newspaper database…or that most Chinese restaurants did not take out newspaper advertisements in this period…or that the database’s holdings only represent newspapers from towns where Chinese restaurants didn’t take out newspaper advertisements. To test whether my results were at fault, I checked the New York Times, and rather few hits on Chinese restaurants came up before 1900.

Despite the lack of classified ads, however, I was surprised at the frequency with which Chinese restaurants were mentioned within regular news articles. Although the database turned up no hits before 1850, 310 citations were found between 1850 and 1899. Some of the most fascinating of these pieces were essentially a hybrid between a restaurant review and a feature piece, as in “A Cuttle-Fish Dinner: A Reporter Dines with the Editor of The Chinese American” (Milwaukee Daily Sentinel, 5/20/1883; reprint from New York Morning Journal). Some, like “Row at a Chinese Restaurant” (San Francisco Daily Evening Bulletin, 6/10/1859), simply mentioned events at a Chinese restaurant, while others mentioned Chinese restaurants within the context of the article, like “The Cleanly Heathen” (Milwaukee Daily Sentinel, 7/17/1879):

“NEW YORK, July 16 — The Tribune, describing the visits of the tenement house inspectors to the worst places in baxter and other streets, says, “The Chinese have taken possession of South Mott street. It is  curious fact that while the most debased among the Irish women of the quarter live with brutal Italians, the younger, best looking, and neater of the live from choice with the Chinese on Mott street, who it is said treat them kindly, do most of the housework themselves and make provident masters. Even the basement rooms inhabited by them smell sweeter than the upper stories where the Italians congregate. At a Chinese restaurant across the street, four Chinamen were found at dinner. The neatness of the table and appearance of the viands would have done credit to a second-class American restaurant, at least. The ware was apparently of choice patterns of porcelain and the food quails and rice.”

The great benefit of an archive like this is the searchability of massive amount of text. From the small article above, which easily could have escaped indexing in more conventional periodical sources (and which could have been indexed under “tenement” or “Chinese” or “marriage, relationships, and family” instead of “restaurants,”), the reader learns about middle-class impressions of working-class approaches to race and sexuality;  she sees early signs of Asians being presented as the “model minority”; she sees that Chinese restaurants were presented in some sources as palatable venues; she catches hints of journalists’ tendency to portray Asians as insufficiently masculine; and (after seeing that many of the earlier articles come from the West), she learns that readers in the Midwest were interested in learning about conditions in the slums of New York.

The Nineteenth-Century United States Newspapers Database is fairly easy to navigate, with a user-friendly interface that allows the researcher to search by keyword and to restrict the search by specific date, date range, and to select individual papers or states in which to search. It keeps a list of past searches, which can be recalled at any time during the search session — very useful when tabbing between search strings and following up impulses from one article to the next. Sources can be printed, or a whole set of them can be emailed as a full-text PDF file to the researcher — a very useful tool for someone compiling their own digital archive on a topic. Plaintext citations can also be emailed.

However, there are limitations — one of the big ones being that the articles are images or PDFs, and not designed for interaction with the text. Quotes must then be retyped as opposed to copying and pasting them. Also, as the historian-bloggers from this week’s assignment suggest, there are quality issues with the automatic scanning and text-recognition, as in the article titled “Chinese CooAery” (the top half of the “k” was lighter than the bottom angle, resembling the pointed angle of an A). And as mentioned, while the range of texts offered is amazing for the scholar willing to search through throwaway references to the topic at hand, for the scholar who wishes a level of specificity somewhere between “mentioned in the title” and “mentioned fulltext,” the lack of filter may be frustrating. (Nonetheless, I have to imagine that it beats the heck out of simply paging through old papers hoping to find something!)

While my original classified ad idea is less plausible with this archive, the project it does suggest — tracing the appearance of Chinese restaurants throughout journalism more generally to gain a sense of regional variation in opinion and gradual attempts to “know” the Chinese — would be well supported by Google Books as well. Google Books yielded a selection of periodicals with articles on Chinese restaurants and cooking, many of which would be useful in the same ways that I envision sources from the first archive being useful. There were also a variety of books that includes chapters on the Chinese and their restaurants. However, it’s important to remember that a scholar who had really done his or her work would also combine research on these digital sources with work at a physical library, in order to take advantage of the resources available on microfilm and microfiche, and perhaps smaller ethnic newspapers that haven’t been well-digitized. For example, if copies of Chinese-American newspapers have been preserved, it would be interesting to see how and if Chinese restaurants (or restaurants in general!) advertised there, and how the copy from one imagined audience to the next might have changed.

In summary though, there’s no doubt that digitization has changed how we research. The types and amount of sources we expect to be able to master for a project, the types of sources we imagine (and forget) to consult, and the way that we organize our ideas and research have all changed because of digitization and widespread availability of sources. I feel privileged to have university access to these databases, but I must agree with Dan Cohen that some similar public job needs to be undertaken in order to create similarly rich public archives. (Anti-recession digitization boondoggle jobs, anyone?)

February 5, 2010

All your database are belong to us?

Filed under: Uncategorized — ths117 @ 6:46 am

[explanation of the blogpost title]

I think both of the seminar papers that I’ve done here at IU could probably have benefited from some database style organization. The first one focused on changes between 1890 and 1940 in mainstream impressions of Appalachians and African-Americans as problematic (or “distinctive”) American populations. I approached the issue by tracing changes in the staging, performance and audience reception of a play/musical/movie called In Old Kentucky.

In retrospect, I think that managing my sources for this project via database might have generated some interesting statistical data, or at the very least might have helped me organize my sources better. I spent a LOT of time combing through digitized theater reviews on Proquest and the Nineteenth Century American Newspapers Database. While I made sure on my first pass through the archives to save a wide selection of the most interesting or suggestive reviews (often as links in a draft email), in my writing process I sometimes found that I had failed to flag the important part of the review for later perusal.  Also, sources I had overlooked initially were now important to my argument. Organizing the theater reviews into a searchable or at least tagged database would have allowed me to master the sources more thoroughly. Also, after reading Laurel Thatcher Ulrich’s piece predicated on numerical data in probate wills, I realize that quick access to statistics like the number of reviews of the show in a given year, the types of publications that covered the show, or even an index of the frequency of certain words within the reviews could have led me to some interesting conclusions about the show’s popularity and resonance over time in different communities. As Bill Turkel points out, surveying “statistically improbable phrases” can be quite illuminating, as could statistically probable phrases.

Databases would also have been helpful with my next seminar paper, an examination of major changes in male romantic advice literature of the late 20th century, with particular attention to the politicization of the interactive Internet-based culture of pick-up artistry. In fact, since much of the material that I was dealing with was born digital (blog posts, websites, forum discussions), Zotero seemed to be a natural fit. However, like Kalani, I discovered that Zotero had limited usefulness — or at least,  my understanding of the software’s potential may have been limited. (Still, I agree with Kalani that it needs to enable better notation, especially of non-standard sources!) My problem with Zotero was that I needed a less folder-based and more web-based interface — one that allowed for multiple connections between pieces of evidence. I tried to use tagging to overcome this issue, but found that I needed to refine my own tagging categories as time went on. Individual pieces of evidence often fit into multiple categories — or none at all.

Essentially, when it came time to write the paper, I realized that I wasn’t as organized as I thought I was. While databases are designed to help you organize your sources and make connections between different pieces of evidence, I could also see the possibility of database maintenance and preliminary organizational behaviors becoming an enormous time-sink! (As an analogy, my mom likes to point to my tendency to make lists about what I need to pack as a way to procrastinate about packing.)

Still, I believe that personal databases have a lot of potential — both for projects with clearly delimited boundaries and goals, and for people who realize that they’re onto something interesting, and who just want to have a pool of raw data to begin to think about. I will need to practice my “spidering” and “scraping” skills (the former enabled more easily these days by tabbed browsing!), gain a better sense of what free or low-cost tools are out there to help me refine and organize my findings (Evernote seems particularly cool), and learn to think creatively about how databases can support either a single project or a series of modular works that draw on the same or associated data sets.

Create a free website or blog at