A recently organized group of digital humanities practitioners in the field of Slavic Studies made a big splash at the annual meeting of the Association for Slavic, East European & Eurasian Studies earlier this month. In addition to a half-day THAT Camp the group sponsored a series of themed panels on the regular convention program. I participated in two of those, one focusing on Platforms for Digital Scholarship, and one on the impact of digital humanities work in the classroom.
See below for a reprise of my comments on the first session about the rewards of re-launching and the challenges of maintaining Seventeen Moments in Soviet History.
We also heard from Yelena Kalinsky, Managing Editor for H-Net Reviews who discussed the use of Drupal as a CMS for the H-net Commons; Emil Kerenji, from the U. S. Holocaust Museum, who introduced a richly annotated digital archive of Jewish Perspectives on the Holocaust; Ruth Lorenz, who maintains Луч света, a marvelous site of news-related video clips to teach listening comprehension and social-cultural literacy to Russian language students; Joan Neuberger founder of Not Even Past, the exemplary public history site with affiliated podcasts and blogs she founded at the University of Texas, Austin in 2008; and Svetlana Rukhelman, who spoke about digital archiving projects at Harvard using Omeka.
I am implicated in various digital humanities projects but will talk today about Seventeen Moments in Soviet History. As many of you know, Seventeen Moments is a legacy digital humanities project. It was conceived by James von Geldern and Lewis Siegelbaum in the late nineties (the heyday of Netscape Navigator and KOI8 fonts) – at the same time the Library of Congress launched the bi-lingual multi-media site, Meeting of the Frontiers, and it first went live in 2002. The NEH supported the site’s initial development – sending Lewis and Jim to Russia to acquire materials, especially the film, photographs and audio files from the Russian State Film & Photo Archive at Krasnogorsk and funding the design for the original site. Over the years Lewis’ and Jim’s host institutions, Michigan State University and Macalester College, provided funds for updates. Kristen Edwards collected content from the Hoover Institution and coordinated a major upgrade to the site in 2008. This was the site that was hacked in January 2015, just as a new version of the site was nearing completion. And this is where I got involved. (Why I got involved is a story for another time.)
For the first several months of January work proceeded on two fronts: 1) we tried to finish up and secure a new version of the site that had been paid for but not finished before the hack of the old version; and 2) we made plans to move the site to an open source CMS that could be updated, secured and modified more easily.
To our great regret, No. 1 proved impossible, so by the spring we shifted attention exclusively to No. 2.
The new WordPress installation has been live since August of 2015.
Following our panel organizer’s guidance, my comments are framed around three issues:
- Why we chose WordPress / MATRIX
- How well things working (or not), with advice for someone starting their own project
- How we think about digital humanities resources like 17 Moments and the kinds hybrid / multivalent spaces they shape.
The concerns that informed our design decisions – why we ended up at MATRIX with a WP install.
Obviously we really wanted to save the “new” site that seemed agonizingly close to being ready for prime time. But there were two problems: The first was solvable – de-bugging glitches such as getting the subtitles to display on the audio files. But the second was not. The “new” site used CakePhp (an outdated variant of the scripting language, Php – b. 1995), and was all custom work. Some of the coding was good, but a lot of it was messy – different solutions to the same problem – caveats left in all caps (i.e. DON’T CHANGE ANYTHING BELOW THIS LINE!). And it was vulnerable to the same kind of MySQL injection attacks that had brought down the old site. The vulnerability was compounded by what should have been a wonderful feature – namely a tiered access system, which would allow different levels of users to suggest and submit new content. (So someone with malicious intent could get access to the database by creating an account.)
Four very accomplished computer science students worked on these problems as their senior project, but couldn’t solve the security issue.
Because we were dealing with php and MySQL we leaned towards WordPress and Drupal for the CMS. These are both open source platforms with robust user and development communities. WordPress, which gained popularity mainly as a blogging platform runs more than twenty-five percent of the web as of November, 2016. While Drupal has more bells and whistles, and more flexibility in terms of customization, it is not super user friendly. WordPress, on the other hand is “easy” – especially at the entry level (all of my undergrads use it). It has thousands of (mostly free) plug-ins that make it a decent CMS rather than just blogging software. It also has thousands of templates that provide a terrific assortment of design options. And although it also has security vulnerabilities, the basic script and the plug-ins are constantly updated with an eye toward security.
As part of saving Seventeen Moments and taking it to the next level, we wanted the site to benefit more directly from the attentions and expertise of its huge user / fan base. We want to expand the site — to develop new modules, add new resources – both embedded in the site and as related links next to the various topics. In my early (wildest) dreams, I imagined that (properly supervised) students could help curate these materials once an administrator (Lewis, Jim or I) had approved them. I am a big believer in the empowering potential of digital literacy and proficiency with web authoring tools. So while Drupal would have made sense in terms of its sophistication, WordPress won out as the people’s platform.
And because we had very little in the way of budget, we were grateful that MATRIX, who had hosted the original site, gave us the friends and family discount for hosting the new site and developing it.
So, how well is it working and what advice do we have for those starting out?
If use is a measure of success we are in fantastic shape.
We’re at nearly 500,000 page views for and over 70,000 visitors for 2016 More than half of the traffic comes from repeat visitors. Most sessions involve 4.72 different pages (which does not count additional clicks on videos or images). As of mid November there are more than 9,000 30 day active users – These would be current students, which is pretty cool.
The site draws traffic from all over the world: mainly the US, Great Britain, Australia and Canada, but with healthy representation from Europe, Russia and the former Soviet Union, and just about everywhere except Central and West Africa.
Most people find the site via search engine (Jim was delighted when we regained our #1 Google Ranking), but many come via curated lists (Exeter, Nottingham, BU, Rochester) and Reddit, Twitter, Facebook, wikipedia, etc.)
So, folks definitely missed the site during the nine months it was down. We’ve also received lots of positive feedback about the look and feel of the new interface, which is responsive. Many students do look at stuff on their phones, even though most of the access is still via laptop / desktop).
And the work that remains to be done?
People do miss the music box, and that brings me to some of the challenges we’d like to overcome in the near to medium future.
We could not find an off the shelf plug-in that could replicate the music box, and will have to wait until we find the support to have one made for us. It shouldn’t be expensive or hard, but given the size of the site and how many plug-ins are currently on it (we know it loads slowly sometimes), this isn’t as straightforward as you might think.
We hope to have more success with regular updates and improvements going forward. Ease of updating (and the benefits of that for security) were key factors in choosing an open source, easy to use, CMS in the first place.
So you have a great project you want to develop…. (advice to people starting out):
1) try to inform yourself and your team about what you don’t know or can’t have and the implications of that – because what you don’t know can hurt your site, and you can’t assume that the folks who are trying to fix things are that much more expert than you are.
2) Cherish your blessings as a late adopter.
3) Think about your project as having an indefinite digital future – it isn’t just about building / launching, but about maintaining, updating, and growing.
Finally: How do we think about digital humanities resources like 17 Moments and the kinds hybrid / multivalent spaces they shape?
I’ve been struck by the accidental and dialogic evolution (over fifteen odd years) of what the site is and what it does. For many, Seventeen Moments began as and still is a valued pedagogical resource – an Open Educational Resource before its time. But it is also much more, and here I think the balance between the site’s accessibility and its high value content have been key.
Having it on the open web does make the site vulnerable at some level, but it also makes engaging global audiences possible. During March of 2014 for example, Lewis Siegelbaum’s subject essay and the related module, “Gift of Crimea”, had more than 50,000 visitors. And now, when just about anything is at your fingertips, I take some comfort in knowing that lots of fingertips find Seventeen Moments instead of whatever fake news might be filtered forward surface instead.
Going forward, I hope we can strengthen the site’s presence by expanding it (along the lines outlined above), getting the Creative Commons license attached to all of the relevant items, and doing more with social media. Our Facebook page and Twitter are linked, but have been neglected because the admins are busy with their day jobs. We have commissioned some more modules, especially for the postwar period and would like to set up a “sandbox” site where people could submit new material for potential inclusion in the existing modules or ones that might be developed down the road.
It also seems that how we identify sites like seventeen moments and what we expect of them is starting to shift.
First there’s the relationship between a site like ours, that was developed as an “online primary source repository” in English, primarily for students (or their profs.) and more scholarly sources like Vadim Staklo’s Russian Perspectives on Islam, which uses OMEKA (also free and open source, but better geared toward library, museum and archival collections).
But both of these sites, and many other digital humanities resources, are also public history venues – sites where history is made, presented and consumed. To audiences outside the traditional classroom, the distinctions between an OMEKA site, with the rigor of Dublin Core metadata and something like Seventeen Moments are not that significant, and both have the potential to engage more users in more complex ways than their pre-digital predecessors.
Second, are the historical underpinnings of the site: Seventeen Moments is rooted in the Digital Humanities as well as Digital History. Digital Humanities started with text encoding projects in the 90s (Women Writer’s Project), and became widespread as a term when the NEH launched the Digital Humanities Initiative in 2006 – several years after Seventeen Moments first went live. Digital History of course emerged from the quantitative approaches of the 60s and 70s, but was (like everything else) transformed by the multi-media, interactive potential of online digital environments. Ed Ayers conceived the Valley of the Shadow project in the early nineties; Roy Rosenzwieg founded the Center for History and New Media at GMU in 1994. (Joan Neuberger launched Not Even Past in 2010.)
And these histories point to one of the big challenges for developing best practices with digital humanities resources going forward, namely, how do we secure their future?
Like campus construction or the nation’s highways, finding funding for new projects and buildings is much easier than securing the resources to maintain and renovate them. At a time when print collections are shrinking, and more and more library resources vanish behind paywalls in the cloud, securing and maintaining spaces for open access digital humanities resources seems especially imperative and challenging. Seventeen Moments proves that “if you build it, they will come,” and we know that they will also stay, and even return after a major disaster. My hope is that we start designing and building our digital humanities resources with their future in mind, so they don’t just get stuck on a server somewhere waiting for the end to come.