In Part One I wrote about Polar Rose, a Firefox toolbar that performs facial recognition on photos loaded in your browser.  Polar Rose programmatically creates associations between individual’s faces that you could have easily overlooked, and opens new possibilities for how we organize and search digital photographs.  Equally as important, Polar Rose is free, built partially on open source software, and uses crowd sourcing to seed their database of recognized faces (read Part One for an overview of other emerging tools).  What’s significant to digital humanities is that it breaks from the text-based search paradigm that dictates our research process by using visual patterns to query images, rather than search text querying metatext.

When first approaching the tool, my instinct was that Polar Rose - or an abstraction of it, could be useful to digital archivists.  In the age of Google Books and with increasing pressure from patrons to see digital copies of archival content, archival objects that have already been organized and described in a finding aid at the collection-level are thrown into an item-level digital archive on the web.  How do I search that content?  How are those objects described for me, as the researcher, to discover them?  When finding aid keywords become item-level tags, how accurate are my searches going to be about the specific image that I’m viewing?  That can vary.  Polar Rose and its ability to recognize individuals in photos begins to describe the “ofness” of a photo, in contrast to larger categorical key words that describe the “aboutness” of a photo and may be in a finding aid.  Simply put, it provides a different form of description — one that is unique at the item-level that may be useful to researchers.  The fact that Polar Rose performs actions automatically could begin to free up time for archivists and augment the information used to search those digital objects.

Polar Rose isn’t only useful to digital archivists though.  It’s easy to see its practical application in small collections of photos — even on a consumer-level.  Local historians would likely have luck with this type of tool as well, if there were enough reoccurring faces for Polar Rose to recognize.  Family photos?  I’d love to have software than can automatically keep track of the faces of my extended family.  Best of all, Polar Rose is built on top of some open source tools, although it’s unclear how much of it could be useful for creating a more generalized application.

The integration of visual search into archives may still be a ways off, and I’m not suggesting that Polar Rose is by any means perfect right now, but I’d be interested in what my colleagues think about visual search and its potential.  Similar to how text-mining can help us get “inside” written works, visual search may be able to do the same for images, while allowing us to search based upon images themselves rather than text.

Over the weekend I gave a brief technical presentation at National History Day about developing History websites to a group of high schoolers and their teachers. From now on, whenever I give a talk I’ll try to post my slides to slideshare so others can view and download it (after being inspired by Karin).  Here’s my first stab at it:

| View | Upload your own

I covered the basics of using a WYSIWYG editor and templates to structure your site, and then explored a few ways to embed content into your pages. While preparing the slides, I stumbled upon a great website called CommunityWalk that offers a simple interface for creating embed-able Google Maps. It appends small advertisements to your markers but it’s still worth checking out.

The photograph above is marked up by Polar Rose, a Firefox toolbar that does facial recognition on photos loaded in your browser. Polar Rose automatically detected three faces in the photograph based upon patterns and compared the images to a database of faces stored on their servers. Had someone already tagged a particular face or person, Polar Rose would have told me who appeared in the photo. Although their database is currently seeded with mostly celebrities and politicians, there’s something going on with Polar Rose that historians and digital archivists should take notice of.

At the heart of my excitement is the possibility of making photograph-to-photograph connections that would otherwise have remained undiscovered. Thinking particularly about collections of family photos, where some have persons labeled and others do not, the ability to automate the identification of individuals in photos goes far in understanding the relationship between these digital objects. The concept of tagging the faces of people in photos, rather than writing a short description accompanying a photo, has become popular in Facebook. There, an individual’s profile aggregates all photos they’ve been tagged in, and presents that corpus for public browsing.. Facebook only expands upon the uses of that information in a limited way, by allowing you to see photos including yourself and another specific person.

Polar Rose is not the only company investing and innovating in visual search. Facial recognition and visual search are slowly being integrated into pre-existing services. MyHeritage’s celebrity face recognition allows you to upload a photo of yourself and see what celebrities you resemble. Like.com offers the ability to see items that are visually similar to the ones you’re shopping for online. And the big gorilla of them all, Google, has quietly implemented facial recognition into their image search. Try this: append &imgtype=face to the url of a image search query, you’ll see with those key words only those images that Google identifies as facelike. The first search I tried was for “digital history“, and I wasn’t impressed. However, when adding &imgtype=face to the query, I happily saw the faces of Roy, Dan, and Bill. Here’s without appending the text, and after. A world of a difference.

I’m breaking this post into two parts. Part 2 will be posted soon and will discuss how facial recognition could improve metadata for digital images and help researchers wade through the seas of digitized images.

[original photo of Northern Pacific Railroad workers from TrainNet]

I’m back from an exciting weekend at THATCamp, working through with a wide range of digital humanists. Among us there was shared excitement and pleasure to engage in the type of informal dialogue that a BarCamp-style unconference like THATCamp affords. And given the interest in it, we’ll be doing THATCamp again next year so pay attention here and on the THATCamp blog for further information later in the year.

Jeremy and I will be recording a new episode of THATPodcast that looks at how we organized THATCamp, and discuss how others could use a similar approach for an untraditional conference. Until then, I’d encourage you to browse the website for a full list of campers, the schedule that we made on the day of the conference, and the community blog that campers posted their sessions on before Day 1. Here are some photos from THATCamp below the fold, among others on Flickr.

Building THAT Schedule

THATCamp Badges

Text Mining Session

Getting Started ‘Building Stuff’

Crowdsourcing Session

Attention cultural historians, museum curators, and robot enthusiasts. On Saturday, the San Jose Museum of Art opened an incredible exhibition on robots that continues through October 19th. This compliments some previous observations I’ve expressed about the robot icon in American culture. Here’s the blurb from their website:

Robots: Evolution of a Cultural Icon examines the development of robot iconography in fine art over the past 50 years. In 1920, the term robot was coined from a Czech word robota, which means tedious labor. Since then, the image and the idea of a robot have evolved remarkably from an awkward, mechanical creature to a sophisticated android with artificial intelligence and the potential for human-like consciousness. As robotic technology catches up with the wild imagination of science fiction novels, movies, and animation, dreams and fears anticipated in these stories may also become reality. Artists included in the exhibition have responded to the technological innovation with optimism, pessimism, and humor, presenting work that ultimately explores our ambivalent attitudes towards robots.

And an introduction from JoAnne Northrup, Senior Curator:

What do the game Battleship, carnival clowns, patriotic country music, and gender in the Oregon Trail computer game have in common? They all were represented at the session I spoke at in Rochester, NY for the MAASA/GLASA conference on play at Strong Museum. This eclectic group of presenters approached play in many different ways something I’ve come to expect from American Studies. While there I got the scoop on a new quarterly publication from Strong whose first issue comes out this summer:

The American Journal of Play will feature articles on such disciplines as child development, education, psychology, anthropology, history, communications, and museology and is aimed at a general audience of educators, psychologists, play therapists, sociologists, anthropologists, historians, museum professionals, and others interested in children and the importance of play.

Last Fall I teamed up with Student Pugwash USA to create a Facebook survey evaluating the positions of 18-25 year olds on science policy and politics, and the results are in! The full report is available for download and while its content falls outside the scope of this blog, I think it’s another great example in addition to my Oregon Trail Survey of how we can target niche groups and wider demographics on the social web. The survey was designed to inform the creation of a 2008 Science and Technology Policy Guide about the candidates.

Despite my optimism of the survey, its self-selecting nature makes it in no way a representative sample of all youth 18-25. One of the most striking numbers was that 96% of respondents said they were planning on voting in the 2008 election. I would consider most of the respondents hardcore political junkies or environmentalists, which is still OK. Overall I’d suggest this approach for collecting oral histories, or other information from the most active members of an online community.

In October 2006, I conducted an online survey about the Oregon Trail computer game by using the beta Facebook API to authenticate users and target OT enthusiast groups. Tonight I’ve published raw results. In only 8 days the survey had 500+ participants representing 44 states and 4 countries. While my original intent was to evaluate my methodology and eventually reconduct the survey, I realized that a) others can use this data, so I should share it b) it’d be beneficial for me to revisit this in anticipation of an upcoming presentation c) I’ve had some terrific blog dialog recently and I’d like to see that continue.

One thing to keep in mind while viewing the results is that the survey was entirely an experiment; I wasn’t entirely convinced that outside of my group of friends I’d attract much attention. I was pleasantly wrong. Most of the questions required short-answer responses, and the lack of quantitative data collected makes the sheer number of responses unmanageable in many ways. Along the lines of oral histories, many of the responses shed light into how students played the game in the classroom, and how they remember parts of the game. I’ve had luck navigating the responses by using keywords, although without being able to bundle similar responses it’s difficult measure the number of occurrences. Still, patterns in the results can be discerned. Here are the questions I asked:

Year of Birth:
Gender:
State:
Game Version:
Where did you play the game?
How often did you play the game?
Was the game taught with a lesson on the trail, or did you just play the game?
What was your favorite part of the game?
What was your least favorite part of the game?
How were Indian characters portrayed in the game?
How were female characters portrayed in the game?
How were male characters portrayed in the game?
What did the West look like in the game? Do you think this was accurate?
Do you think the game accurately portrayed the Oregon Trail? Why/Why Not?
Any additional comments or memories of the game:

The object of the survey was to have respondents recount representation within the game, judge whether or not they think this memory of the game is accurate, as well as give insight into how the game was used within an educational setting. Understanding historical representation in the classroom, and measuring the effectiveness of the game to present that knowledge was of key interest.

Beyond the educational goals of the game, many of the survey responses discussed elements of the game that I glossed over in my research, but resonated with students. One of these was the ability to name characters in the game. Here are some interesting quotes related to that:

I liked the excitement of it. It felt like an adventure, plus you could always name your crush as your husband in your party!

I would use my friends names as members of my wagon train

Being able to name people… And then letting the ones I didn’t like die, or rub salt in their wounds, etc. Morbid, I know, but I think most kids will agree that was the best part of the game.

the first time I played, the people I played with named all our characters after the X-men because they were popular at the time, because of the cartoon.

Also fun to name your characters stupid things. My personal favorites were names like “Nobody” or “Somebody” or “Everybody,” so that the game would be like “Somebody has cholera” and you’d be like ‘but who?’ and “Nobody has died” and you’re like ‘Whew, that’s a relief.’

The last response was the best IMHO.

Those who have read my blog for a while know about my ongoing interest in robots, particularly the development of the robot archetype in popular American Culture in the early twentieth century. By and large I’ve sought to understand these depictions of robots within the cultures that produced them. Whether that was the image of the “savage robot” explored in an earlier post about a 1937 Lil’ Abner comic strip, or an earlier attempt to create a brief pre-Asimov robot timeline, the relationship between humans and machine is fascinating. In several weeks I’ll have a call for participants in my newest exploration of the robot archetype, but more on that later.

Elektro was a robot presented by Westinghouse at the 1939 World’s Fair that helped bring the robot into public consciousness, as well as present an idea of what the future of robotics would be. Seven feet tall, this steel humanoid robot walked by voice command, talked, moved its head/arms and even smoked cigarettes. Elektro’s voice used a 78-rpm record player, saying lines including “If you use me well, I can be your slave” — similar to the Lil’ Abner comic. Not only was Elektro made in the image of man, his ability to smoke distinguishes his activities closer to humans than a machine. This fascinating diagram shows the moving components of Elektro, including his bellows.

The submission deadline for The Humanities and Technology (bar)Camp is Midnight, Saturday the 15th; you have less than 4 days to submit your proposal! If you’re free the weekend after Memorial Day, send us an application to join a top-notch crew of digital humanists give a range of presentations “from full-blown papers (not many of those, we hope) to software demos to training sessions to debates to discussions of research findings to half-baked rants.” More information is on THATCamp’s website and on twitter.