I’m inspired by Mark Sample‘s playful remixing of ‘Hacking the Academy‘ along procedural lines, ‘Hacking the Accident‘:

Known as N+7, this mode of procedural writing was popularized by the Oulipo group, an attempt to divest creative expression of two hobgoblins that haunt the modern age: the myth of the muse-touched creative genius, and the equally debilitating reductivism of the Freudian unconscious. N+7 circumvents these fictions, these stories, these—what “stories” under N+7 become—straightjackets. N+7 celebrates the random, valorizing the lack of what artists, writers, and intellectuals often seek in their work: control. It is a yielding, a submission—not to the muses, not to genius, not to dreams and desires, but to the world beyond us. It is pure engagement with the outside world.

The results of N+7 would seem absolutely nonsensical, if not for the disruptive juxtapositions, startling evocations, and unexpected revelations that ruthless application of the algorithm draws out from the original work. Consider the opening substitution of Hacking the Academy, sustained throughout the entire book: every instance of academy is literally an accident.

I’m no literary scholar. But what I am interested in are the ways deeper patterns can sometimes be seen through machine lenses (and in particular, topic modeling), and I think that the Oulipo group might’ve been on to something (just ahead of their time). So, I scraped every paragraph from the last 10 posts of Play the Past, via its RSS feed (which took me back to November 29 2011). I searched for 15 topics, and then mapped the interconnections between paragraphs by topics:

Oh what a tangled web... topics connected to paragraphs in Play the Past - topic 12 seems to be central... (generated using http://gephi.org)

So: it’s time to choose your own adventure. If you click on http://j.mp/AADmUS you can follow the tangled web of ideas and text. What happens when you playfully read Play the Past?

List of Topics, generated using Mallet

To that end, here’s something one can do with Omeka, to push items from its database into the Wikitude augmented reality platform.

1. In Omeka, have the Geolocation plugin installed and working.
2. Navigate to http://[your omeka site.com]/geolocation/map.kml
3. You should see the xml structure of your geolocated items.
4. In a new tab, go to wikitude.me, and sign up for a developer account (it’s free).
5. Click ‘add new world’.
6. Click ‘upload KML file’.
7. Fill in all required fields (you’ll have to create a 32 by 32 pixel icon to serve as a dot-on-the-map, and upload that too).
8. Under ‘KML/KMZ’ file, click on ‘Enter KML URL’. This will give you a box into which you may paste the URL from #2.
9. Hit save.

If you’re successful, the next screen will tell you how many points have been uploaded. If, at some later point you’ve added many more items to Omeka, you’ll have to go back to your World in Wikitude and hit save again, to upload the most recent stuff.

Now, with Wikitude on your phone, you might not be able to find your world right away. There’s a solution. If you log back into the Wikitude developer zone, and click on the world you just created, you’ll find a string of letters under ‘developer key’. On your Iphone, go to ‘settings’ , select ‘Wikitude’. Under ‘Developer Settings’, there’s a box for the developer key. Enter that developer key there. Start Wikitude up, refresh the display, and your items from Omeka will be under ‘Around Me’.

…And there you have it. Right now, this just does the basic text descriptions, and the location. By fiddling with the Geolocation plugin code, one might be able to add the other information that Wikitude can display, like images, video, audio, etc.

For a similar approach, but directly from Google Maps, see this video by drmonkeyjcg:

Today though, I’m trying a different approach. I want students to understand that the choice of tool, and the way data gets represented or manipulated in a computer, are not inconsequential choices. To that end, I’m working with Prezi. Its metaphor – zooming – couldn’t be further from Powerpoint – recreating the 35 mm slide.  It’s a simple example, but I think it should make the point elegantly. Now, as to ‘getting started with digital tools’, I could’ve just sent the students to the DiRT wiki (which I am indeed doing), leaving it at that, but again, I wanted to make my larger point have more resonance. In which case, I’m going to take them through my digital workflow as I use topic modeling to try to understand deeper structures in the corpus of ancient writers. I’m not pretending to comprehensiveness this morning. Rather, I am using my own research (and how I came to this research) as a trajectory for launching students into their own research. Below is my prezi; please feel free to use, adapt, alter accordingly to your own needs.

While I have nothing near so polished available, the GUI for Mallet used with Gephi can do nearly the same thing. My body of data comes from Writing History in the Digital Age. An earlier experiment with the same data is recounted here. I re-ran the data using the GUI approach, and have to say, this is a much easier and accessible approach. Run the program; select the folder with your txt documents in it; select the target number of topics; select the appropriate language stopwords list if necessary; hit ‘train topics’. What is very neat about this program is how it presents its output in both html and csv.

So in the spirit of crowdsourcing, I’ve put the output files online, and haven’t tried to decide yet what the topics might mean. Instead, why don’t you view the files for yourself, and let’s identify the topics using the comments of this post?

I then took the CSV files, and got them ready for import into Gephi. Decide which two columns you’d like to represent as being connected, and prune away the extraneous data. I took the ‘topicsindocs.csv’ file, and pruned it so that each paragraph of each author is paired with its major topic. I stripped away the info about the paragraph itself, so that the resulting visualization is just authors to the topics they write about. In the screenshot below, you can see the open gephi file with my own ‘Wikiblitz’ article highlighted, and its connections.

What’s also interesting is when I ran the ‘modularity’ routine – identifying communities based on patterns of self-similarity of ties – only four communities emerged (albeit with a very low modularity measurement, 0.235, which suggests that these communities are all that strong). A natural grouping of the papers, perhaps? (by the way, here’s the pdf/svg file).

This is the second archaeological project I’ve seen this year bypassing the traditional means for funding research (the first being the Maenander Project by Colleen Morgan, via Kickstarter; Colleen’s fundraising was successful, btw). It only makes sense. Archaeology is exciting. It’s paid for via the public, usually through the proxy of government fiat or funding agency. Why not take your research directly to the people who ultimately pay for it? Each dollar raised is a direct vote of confidence in your work.

Would my PhD research have been able to raise funds this way? Perhaps. It’s long-tail economics…

Topics by authors in Writing History in the Digital Age

I and my students have made some contributions to ‘Writing History in the Digital Age‘, the born-digital volume edited by  Jack Dougherty and Kristen Nawrotzki. Rather than reflect on the writing process, I thought I’d topic model the volume to see what patterns emerged in the contributions.

I use Mallet to do this. I’ve posted earlier about how to get Mallet running. I used Outwit Hub to scrape each individual paragraph from each paper (> 700 paragraphs) into a CSV file (I did not scrape block quotes, so my paragraph numbers are slightly out of sync with those used on the Writing History website). I used the Textme excel macro (google it; it lives in multiple versions and requires a bit of modification to work exactly the way you want it to) to save each paragraph into its own unique text file, which I then load into Mallet.

Phew. Now, the tricky part with Mallet is deciding how many topics you want it to look for. Finding the *right* number of topics requires a bit of iteration – start with say 10. Look at the resulting composition of files to topics. If an inordinate number of files all fall into one topic, you don’t have enough granularity yet.

As an initial read, I went with 15 topics. One topic – which I’ll label ‘working with data’ – had quite a large number of files (composition document) (remember, the individual paragraphs from the papers). Ideally, I would re-run the analysis with a greater number of topics, so that the ‘working with data’ topic would get broken up.

I also graphed the results, so that each author is linked to the topics which compose his or her paper; the thickness of the line indicates multiple paragraphs with that topic. I have also graphed topics by individual paragraphs, but the granularity isn’t ideal making the resulting visual not all that useful. The colours correspond with the ‘modularity’ of the graph, that is, communities of similar patterns of connections. The size of the node represents ‘betweeness’ on all paths between every pair of nodes.

So what does it all mean? At the level of paragraph-by-topic, if we had the correct level of granularity, one might be able to read the entire volume by treating the graph as a guide to hyperlinking from paragraph to paragraph, perhaps – a machine generated map/index of the internal structure of ideas. At the level of individual authors, it perhaps suggests papers to read together and the organizing themes of the volume.

This is of course a quick and dirty visualization and analysis, and my initial impressions. More time and consideration, greater granularity, is to be desired.

Enter PhotoFly, from Autodesk. PhotoFly is currently free, and it works magic. It transforms your computer and digital camera into a 3d scanner. You take a series of overlapping photographs, upload them to the program, and the program sends them to Autodesk for processing. The system works out from the photographs three dimensional points, and uses these to stitch your images into a wire mesh model (with your photos as the overlay). The results are impressive – once you figure out the trick of taking sufficiently redundant photographs to provide the necessary information.

I used a Kodak EasyShare camera, Z612, and tried four different scenes before I finally started to get the hang of it. My first was a ceramic coffee mug with a glossy white finish. The shininess of the mug confused the processor. I then tried a Campbell’s Soup Can (my nod to Mr. Warhol). I put it on a lazy susan, set my camera up, and rotated the lazy susan through 5 degree increments, thinking I would get good coverage. This did not work (which I would’ve known had I watched the tutorial videos, but really, who has time for that?). The reason it did not work is that the algorithms that stitch everything together count on differences in perspective, focal depth, and so on to work out the relative placement of the camera for each shot, and hence the distance between the focus point and overlapping points that can be identified in the shots. At least, I believe that’s the reason. I tried again, moving around the soup can, but this time I didn’t get enough overlap.

My next attempt was to do my office (imagine, scanning an interior space!). I had more success this time, but again my overlap and the sheer clutter in here defeated me. Finally, I put a small toy car (about 5 inches long by 2 inches wide) on a chair, and proceeded to take about 20 photographs of it from every angle, varying the depth and distance. By this point, I was starting to get the hang of it, and it uploaded quite well. The ‘draft’ model came back more or less complete, and I saved it and sent it to youtube:

In the ‘draft’ mode, you can select triangles to clip, provide real-world coordinates and measurements (useful for creating scans of buildings and interior spaces especially) and basically do enough pre-processing that the model looks just about complete. One then changes to either ‘mobile’, ‘standard’, or ‘high’ quality and the model is sent back to Autodesk for more processing. At this point, the model can be exported in a variety of formats, especially CAD formats like DWG. This is where I get very excited. Sketchup Pro for instance can import DWG. And Sketchup can be used to create AR. I could imagine scanning an interior space, sending the resultant model to Sketchup and then into AR, tying the model to a QR code. Since my model can be sized 1:1, it should be possible for instance to scan say a cave-shelter, and then step into it somewhere else (the middle of a playing field, for example). I’m very excited about the possibilities; more on these as I explore.

My final model of the toy car is below. All in all, it took approximately 30 minutes to get to this stage. Obviously, my model has some flaws in it, but for 30 minutes work… not bad.

Problems with the Software

From time to time, the software just mysteriously died during the upload process. Save early, and save often. Also, exporting to youtube was often fraught. One enters the mysterious world of picking the right quality and codec to make it work. Once I selected ‘mobile’ for everything, things seemed to work better.

9/10 stars from the electric archaeologist. This is the most exciting piece of software I’ve played with in ages.

mea culpa: there should be a citation of the Laura Secord plaque to Rob MacDougall and his historical ARG, Lies Here, recounted on Play the Past.

H20 was a miniseries starring Paul Gross. It was filmed in Ottawa in the winter of 2004. I thought it would be fun to see how tv movies get made, so I auditioned and got a few days of extra work. Most of the time, it was as a face in the crowd. But once – once! my big break! look ma! – I got to be in a scene with one of the principles (I saw Paul Gross once in the hallway of the Chateau Laurier, where we were filming, but that doesn’t seem enough for my purposes today). The principle was Yves Jacques, who was playing the premier of Quebec. If you watch the film, you can see me nod and answer the phone. 2 seconds of cinematic glory. But sufficient to generate a Bacon number (from the Oracle of Bacon, more or less, since us extras aren’t in the database):

Shawn Graham was in H20 (2004) with Yves Jacques who was in Michael Kael contre la World News Company (1998) with Elliott Gould who was in The Big Picture (1989) with Kevin Bacon

Perhaps we need an archaeological equivalent. What’s your Wheeler Number? How many steps to the icon?

Michael Fulford was one of my supervisors; Fulford worked for Barry Cunliffe, who studied with Glyn Daniel ; Daniel was a host of the game show Animal, Vegetable, Mineral? along with Mortimer Wheeler (in 1971, Cunliffe also hosted)…. making my Wheeler Number 4. (or 3, depending on whether or not Cunliffe co-hosted with Wheeler in 1971).

Of course, it might be more …academic… to look at citations and co-authorships, but somehow it’s more satisfying to know that a game show figures in this linkage.