Hiking with Howard Rheingold
Back home again, I’m still processing all the ideas and people we met with in Chicago and California. Our last visit in California was to Howard Rheingold, who took us walking in the mountains by his home outside of San Francisco. Rheingold has a beautiful enclosed garden smelling of roses, with an office for his writing and a studio down the back for his painting. A work in progress in his studio is a pataphysical slot machine he’s building with some friends – all wood and paint now, but to have three dimensional dioramas and circuit boards as well when it’s finished. I should have taken a photo – but I did find a video of it on YouTube (of course). It’s painted blue now, but is still a ways from being completed.
When Rheingold visited Bergen in 2004 for our Digital og sosial conference I remember he found himself a rock on Fløyen to bring home to his garden. I was quite impressed that eight years he knew exactly where it was, neatly arranged with other rocks and souvenirs of his travels. Rheingold lives in the hills near Muir Woods, home of the redwoods, and he took us walking up the steep hills at an impressive speed – we Bergeners are used to hills but were still huffing and puffing behind him.
Howard Rheingold has been one of those people who sees what’s going to happen next since the early days of social technologies. His 1993 book The Virtual Community (now available online) was seminal, as was Smart Mobs ten years later. I’m currently reading his latest book, Net Smart, which outlines five literacies needed to manage technology usefully – when to allow yourself to be distracted while surfing for example, and when (and how) to maintain a focus. One of his major current interests is mind amplifiers – that is, the ways in which technology ranging from writing to blogging to more can augment our thinking.
Unfortunately the videos from the Digital og sosial conference in Bergen in 2004 are lost in cyberspace (we’re looking for them and may yet get them up again) but Jon did find and upload this video of Howard Rheingold answering questions after his talk at the conference. I’ll let you know if we find the video of the talk itself…
Visting Stanford Literary Lab
Franco Moretti‘s work on “distant reading” (or macroanalysis) has been very inspiring as we have explored what kinds of research we can do using the ELMCIP Electronic Literature Knowledge Base, especially his book Graphs, Maps, Trees: Abstract Models for a Literary History. Moretti generously supported a grant application for developing visualizations of the data in the Knowledge Base last year (we didn’t get the grant, but are still working on it) and we were eager to visit the Stanford Literary Lab, which is led by Moretti and Matthew Jockers, to learn more about their work and hear some of their thoughts about our Knowledge Base.
After a beautiful drive up the Californian coastline from Santa Barbara we arrived at Stanford on Thursday afternoon and easily found the Literary Lab.
Matthew Jockers’ work is primarily on text mining of literary texts, and he immediately starting thinking about how and whether one could do text mining of electronic literature. This is something Scott and Jeremy Douglass also discussed earlier this week in San Diego. A problem is that works of electronic literature are very heterogenous in comparison to most traditional literary genres (Jockers noted that this is the case with modernism, too, which is why there is very little digital humanities work in the field, other than on individual authors) making it difficult to draw patterns out of large quantities of texts. On the other hand, as Scott and Jeremy Douglass had discussed, and Scott and Andrew Salway have discussed in Bergen, it would be possible to automatically extract and compare particular code structures, functions, use of visuals and so on. There are also genres of electronic literature that are (or have been) relatively stable, such as Flash poetry, interactive fiction or hypertext narratives.
Members of the Stanford Literary Lab listening to our presentation of the Knowledge Base. Matthew Jockers is on the left, Franco Moretti on the right, and grad student Ben Allen is next to Moretti.
But the ELMCIP Knowledge Base as it stands now is primarily focused on metadata rather than the texts themselves. We show the connections between creative works and events they were presented at, critical writing they are discussed by, organizations with which their authors are affiliated, journals in which they were published and so on, and being able to visualize all this in a social network graph would be immensely useful. And although the works themselves might be characterized as having “extreme morphological variation”, as Jockers or Moretti said, the network of connections is homogenous. For instance, as Moretti put it, all the authors in our database are humans, most are alive, and many live in Europe or the US.
Jockers suggested that simply showing a graph view where different categories of node were displayed in different colours (blue for creative works, green for critical works and so forth) would be useful. We’d also like to be able to see how nodes would cluster. How important are geographical locations in the community, for example?
Another way of viewing the data is in a timeline. When he saw our rather simple graph of when creative works in the Knowledge Base were published, Moretti immediately jumped up to take a closer look. 
“There’s not really a long enough time span to identify trends,” he said, “but would you say that there was a steep increase and then a stabilization?” The problem with our data, of course, is that we don’t really know how representative it is, other than that we most certainly do not have an entry for each actual work of electronic literature in the world. The data is skewed by the interests of contributors and by our knowledge. For instance, we know there is a great deal of Brazilian electronic literature but we have almost none of it registered in the Knowledge Base. Moretti pointed out that with enough data over time, you can link literary trends to events like wars or social upheaval, but you can’t really do that with a fifteen year time span.
Moretti did suggest comparing this graph to the publication of critical works, which we’ll certainly try next. Graphs like these are easy to generate by dumping a spreadsheet from the Knowledge Base and playing with it in Google Fusion or even Excel.
But if trends are difficult to track with such a brief timeframe, Moretti suggested that typologies would be easier to spot. While trends link literary history to social events outside of the texts themselves, typologies are internal to the data, and allow us to identify genres and themes. You need to work differently with the data with typologies than with trends, being much more precise.
Stanford Literary Lab mostly researches 19th century literature, because that is what they have good corpuses for. Until recently, we haven’t had access to corpuses or metadata for electronic literature, because there have been no systems for documenting these works. Libraries haven’t known what to do with it, apart from the few works that are published on disk or CD-ROM with an ISBN number. Hopefully the existence of data in the ELMCIP Knowledge Base will allow future scholars to engage more with electronic literature as well as the classics.
Relationships between nodes in the RoSE project and the ELMCIP Knowledge Base
Today we met with Alan Liu, Rita Raley, Dana Solomon and Lindsay Thomas of the RoSE project at UCSB. RoSE stands for “research-oriented social environment” and, according to the project description, allows “tracking and integrating relations between authors and documents in a combined “social-document graph.” So authors and documents are equal nodes in a system, and there are different kinds of relationships between them. It’s “Facebook for the dead”, as Alan quipped, although the nodes in the system range from the long-dead (Aristotle) to the very much alive and kicking (William Gibson or Alan Liu himself). RoSE uses network graphs both as a visualization of the connections between nodes and simply as a means of navigating the database.
Unlike Facebook, the “edges” or connections between nodes are more nuanced than just “friends”. For instance, someone could be described as a “scholar of” Aristotle, or, more complicatedly, as a “lover of” or a “cousin of”.
The list of possible relationships is long. Scrolling down beyond the menu you see on the screen are further options, such as “enemy of” and many more. Alan and Rita noted that perhaps it was actually too simplistic to view a node as a single entity. Relationships evolve over time – even after the death of the author.
(Btw, that’s Dana Solomon lurking beside the projection in that photo – he’s just started working on a PhD on data visualization in the digital humanities, and I noticed lots of interesting links in his Twitter feed.)
One of the challenges of the project is the time involved in manually entering relationships. The database is seeded by metadata harvested from Project Gutenberg, YAGO, and the SNAC project, but this only provides a very thin framework because not a lot of relationships can be automatically read out of library metadata other than ”X is the author of Y”. This provides a large number of nodes, but relationships between them are thin. Many nodes are singletons or just one relationship.
On top of that, RoSE allows groups to add a thick description on top of that, because people can manually add relationships. So there’s a top-down controlled vocabulary as well as a bottom up crowdsourced vocabulary.
There are a few examples of projects visualizing relationships between humanities data, such as KNALIJ (which needs to be pronounced “knowledge” in an American accent) which visualizes relationships between scientific medical publications. This works because the data is well-behaved and in a standardized format, unlike, say, electronic literature, or people. Projects like RoSE and the ELMCIP Knowledge Base have had to manually enter all the relationships that KNALIJ is able to automatically extract. It’s hard to see how such relationships could be automatically harvested, beyond the very thin framework that projects like SNAC can harvest.
RoSE is built on Ruby on Rails with Flash for the visualizations, and was coded in house by a collaboration between the English department and Media Arts and Technology at UCSB. The ELMCIP Knowledge Base is built in Drupal using modules that support extensive node-referencing.
A clear connection between both the RoSE project and the ELMCIP Knowledge base is our shared deep interest in how creative communities form, take shape and evolve, and how the connections and relationships between people, texts and activities are crucial for documenting and understanding this.
Google form fiction (at least, one assumes it is fiction)
A few days ago, I received an odd email, apparently a followup to a first mail that I don’t think I ever received. I almost deleted it instantly as spam, but after a second glance I decided to try that URL. It is written by a man who is seeking his replacement. What he does, exactly, is however not made clear. Look, here’s the email:
The website is not a lot more elucidating, but the Latin title of the website, Moderari et tueri omnia (regulate and protect all) and the introductory text, it appears that the job is quite important:
While I cannot tell you the specifics of this arrangement, I can say my daily tasks are vital to everything. And by everything I mean exactly what the word implies. Without the simple and reoccurring tasks I complete, what you know now will no longer be.
The majority of the website is devoted to a lengthy questionnaire made in Google Forms. There are many questions. I believe many of those who might begin to answer would be put off by the complicated responses required:
A grocery store near your house is having a sale on breakfast cereals. The hobby store next door has run out of specialty dice, but has an overstock of scale replica airplanes. And while the parking lots trees are mature it appears the birds are avoiding them. Please explain how all these events are related.
When I shared the link on Facebook a friend pointed out I should beware lest answering the questions committed me to pushing a button every two hours like on Lost. I suspect something more closely resembling superheroes might be behind this though:
Do you have full freedom of movement or have access to devices enabling basic tasks such as pressing and pulling?
Note: you will be asked to fight in a strength enhancing suit. So answer accordingly.
However there are a few references to buttons and levers as well.
If anyone has the patience or creativity to actually answer all the questions and submit the form, I would love to hear whether you receive any response.
Turn your workouts into zombie adventures
I haven’t tried this yet, because I, uh, left my running shoes in Norway, but Zombies, Run is certainly an amazing concept. It’s a running app for your phone that sets you in a story: 100 metres into your run you hear ZOMBIES CHASING YOU! You have to escape them, and pick up medicines and other supplies before returning home where you can log in and use your collected supplies. Apparently there’s a story being narrated as well, and the more missions (or runs/workouts) you complete the more the story develops.
I suppose Endomondo and Runkeeper slightly gamify your workouts, but this is taking it to a whole new level!
A tour of the ELMCIP Electronic Literature Knowledge Base
We’re going to be demoing the ELMCIP Electronic Literature Knowledge Base on our tour of the US, and I’ve mapping some paths through it that demonstrate some of my favourite aspects of the Knowledge Base. The Knowledge Base is a relationship database built in Drupal, with about 4000 nodes now, after a year and a half of development. There are entries for creative works of electronic literature and for critical works in the field, and these are connected both to each other (“This creative work is referenced by these critical works”) and to entries for authors, events the works may have been presented at, syllabi for courses they were taught in and more. So there’s bibliometric information, but also the potential for extracting social networks and more.
Here’s the page showing the most recently updated creative works. Right now there are 1175 creative works registered, and we add new works daily. Here I especially like being able to search by year published and by language.
Here’s a graph showing how the creative works are distributed by date of publication. This was made in Google Fusion, we don’t yet have a plugin to do this.
Here’s the page for a single creative work: Stephanie Strickland’s slippingglimpse:
The left column shows all the bibliographic information and links to the different versions of the work that are available online. There’s a brief description of the work, and a list of links to critical writing in the Knowledge Base that references the work.
Further down on the entry page you can also see which course syllabi registered in the Knowledge Base where this work is taught:
Here’s a screenshot of the Knowledge Base entry for the middle course listed there, Lori Emerson‘s graduate course on Digital Poetry and the Limits of Interpretation.
If you look at the entry itself you’ll see the many creative and critical works Lori uses in the course, and follow through to all the links. This is a hugely useful tool for developing new courses on electronic literature, because you can so easily see how other people teaching a particular work have used complementary works and essays, and when you add a work to your syllabus, you can very easily see which critical writing references it. Teachers, authors, critics and students can all request accounts and edit their own entries, much like in the Wikipedia. (You do have to request an account and show some basic interest in electronic literature, simply to avoid spam.)
Another useful way of using the syllabus function is to let your students build the syllabus in the Knowledge Base. Teachers can be given special features where they can see their students’s work, and students can add links to works already in the Knowledge Base and create new entries for works not already there.
Another way to explore the Knowledge Base is by starting with an author. Chris Funkhouser’s entry lists two creative works he has made, a string of critical writing, shows a journal he edits, and links to the entry for a syllabus of a course he has taught. We can click through to the entry for his new book, New Directions in Digital Poetry, and see links to all the creative works and critical writing substantially dealt with in the book.
There are more – just click the screenshot to see the full entry. Actually, the ELMCIP team at UiB had a “Friday Funkfest” where they (unfortunately, I couldn’t join them that day) spent several hours jamming together to get entries in place for all the works referenced in the book.
Today we’re off to the School of the Art Intitute in Chicago, where we’re meeting Judd Morrissey in Art and Technology and doing our first presentation. I’d better go and get ready!
US tour for the ELMCIP Electronic Literature Knowledge Base
Tomorrow Scott and I start our seven-university tour with the ELMCIP Electronic Literature Knowledge Base. We’re visiting a string of excellent digital humanities, digital culture and visualization labs in Chicago and California, and are hoping to learn a lot about what they’re doing and get lots of good ideas for how to further develop the Knowledge Base.
One of the things we’re most interested in is how we can present the increasing amount of data in the Knowledge Base in even more useful ways. The KB includes bibliographic information about creative works and critical writing but also relational links to entries about events at which works are presented, organizations and artist’s collectives that authors have been affiliated with, and syllabi that teach works of electronic literature. We’re interested in developing visualization modules, and in finding ways in which researchers can query the database. Right now we can do simple things by uploading a data dump to Google Fusion or Manyeyes, like this graph showing the frequency of works tagged with “hypertext” in the KB according to their year of publication:
For instance, it would be great to be able to extract answers to questions like these:
- What are the community structures in the field? Do actors form close-knit clusters or are groupings more random and transient? Are these structures stable over time?
- Is there a connection between productivity and particular types of position in the social network of the field (e.g. being part of a closely knit group)? What characterises the community participation of actors whose works are highly referenced?
- What common characteristics do actors who frequently interact and thus belong to a group share? E.g. nationality, residence, age, language, gender, genre they work in?
- Has the speed with which ideas and theoretical paradigms are developed and disseminated internationally increased with the adoption of network technologies? Can literary genres in new media be understood as “memes” which circulate and are developed virally on a transcultural basis?
- Can necessarily reductionist, quantitative, semantically structured approaches to mapping literary and cultural practices enable richer, more expansive analyses of individual artifacts represented within an unfolding historical context?
Here are some of the places and people we’re visiting:
- Art and Technology at the School of the Art Institute in Chicago
- Joe Tabbi, English Dept at University of Illinois at Chicago and director of the Electronic Literature Directory
- Electronic Visualization Lab, University of Illinois at Chicago (Daria Tsoupikova visited us in Bergen last year as a visiting Fulbright Scholar)
- Quinn Dombrowski at the University of Chicago
- Mark Marino, UCLA (who spent a month teaching our students as Fulbright Scholar last spring) (only Scott – I couldn’t bear to leave the kids for more than four nights…)
- Lev Manovich and the Cultural Analytics people at UCSD (Scott only)
- The RoSE project at UCSB, with Alan Liu and Rita Raley (Rita also spent a month with us last spring – we’ve had some great Fulbright visitors!) (I’ll be there for this one!)
- The Stanford Literary Lab (Franco Moretti’s Graphs, Maps, Trees: Abstract Models for a Literary History was very inspirational to us)
Definitely a busy trip, but it should be very inspirational. It’s also our first try at leaving the little ones (aged 2 and 4) for several days with Grandma and Grandpa in Chicago – hopefully this will turn out to be fun for all concerned!
Fitbit as an automated diary
I’ve been using a Fitbit step counter since New Year and have been enjoying the various visualizations it gives me of my days. It provides an interesting form of automated diary of my days.
Here, for instance, is last Monday, Easter Monday, which was a holiday here in Norway. The y-axis shows calories burnt each five minutes, and the x-axis shows the time of day. That pink spike at 7:30 am is me walking over to my mother’s to have breakfast with my sister and her family before they began their eight hour drive home after Easter in Bergen. After that you can see I barely sat down until our two and nearly-four-year-old were in bed at 7 pm, although we didn’t do anything particularly strenuous. Then a nice relaxing evening with Scott, with a bit of pottering around the house and some getting ready for bed in the evening. I logged a total of 11417 steps.
On Tuesday I was back to work, and as you can see, we got off to a late start – I was in bed until nearly 8. The yellow lines are walking the kids’ to preschool, then I took the train to work, so there’s only a narrow pink spark where I walked up the hill to the university from town. I was rather sedentary at work, although I did use my standing desk a little in the afternoon. The second pink line is me running to the train, then I picked up the kids, we got dinner going and so on. Once they were in bed at 7, I crashed, then did some housework (the blue lines after 8 pm) and then sat down to do some more work before turning in. A sadly sedentary day, at only 8698 steps.
On Wednesday I was determined to be a bit more pro-active, and after an efficient morning and preschool-dropoff I walked to work, which takes about 50 minutes and is a pretty good workout, I’ve realized. It costs a lot less time than going to the gym and I’m more likely to actually do it, especially now I’ve got the Fitbit cheering me on. I moved around a bit more at work and was more active throughout the afternoon and at home in the evening too, ending up at 15505 steps.
Fitbit lets you download (some of) your data, and I’d love to see other examples of how you could visualize it – and of course there are other views provided by Fitbit in addition to this. There are other fitness trackers too, such as Nike’s Fuelband, which is a bracelet that actually vibrates to alert you that you’ve been too sedentary.
They’re not primarily sold as diaries, though, and although you can add journal notes for each day, they’re not displayed next to the visualizations so that it’s not too easy to annotate days, either, without taking them out of the Fitbit website as I’ve done here.
Connecting all your social media to make a scrapbook
After my last post about visualizations of personal data, Guttorm Hveem suggested trying Memolane, a service that gathers all your social media feeds into a scrapbook that you can either keep private or make public. So of course I signed up and connected up a dozen of the services I use (a few weren’t there, but many were) and here you go: my last few days in social media:
As you can see, I’ve not really been very active online over the last few days – I’ve been enjoying my Easter break mostly offline. And the social media I’ve used, like Endomondo and Path, aren’t among the ones Memolane can hook up to. Neither does it show phone calls and text messages, as Evertale does, though future Facebook events I’ve said I’ll attend are up there.
Memolane found all my old Tweets, and that in itself is wonderful – in some ways, services like this let us reclaim the scattered info we spread on all these (commercial, out-of-our-control) sites. Ironically, my old blog posts don’t show up, because the RSS feed only goes back so far.
Of course, Memolane itself is, I assume, commercial and out-of-our-control. And as always, I’m a little leary about connecting all these services, although Memolane does let me mark individual feeds as private, public or friends-only. And yet I sign up anyway because I love seeing how all these things work. Here’s a prettier screenshot, from a week where I was using Foursquare more and taking more photos.
Are there other scrapbooks that merge social media feeds like this?
Reading old blog posts: cameras on phones will be huge in the future (2003)
Yesterday, Instagram was bought by Facebook for a billion dollars. Me, I just reread a post by Justin Hall from 2003 where he argued that cameras on mobile phones would be huge in the future. We don’t call Facebooking and Instagramming “mo-blogging” any more (the fact that it’s mobile isn’t even really the point any more) but they are certainly huge.