Originally titled, “The Irretrievable Archive: Issues of Accessibility and Security,” I now think my project should be called “The Panoptical Archive,” or something of the sort. The phrase “too much information” seems appropriate. The embedded information that sits within a picture, let’s call it, the picture’s digital DNA or metadata, can be used by interested parties for whatever purpose they deem fit. The date and time that a picture was taken, is recorded. The camera that was used to take this picture is recorded, as is the focal distance, aperture, the make and model of the computer on which the picture was retrieved and/or recalled and other information. Just open the picture in your image viewer, go to “Tools” and click “Get Info” – it’s all there.
This metadata can also be completely useless and irrelevant – existing in a netherworld of 0s and 1s, in digital limbo. That is, until someone wants or recalls that information. As in a traditional archive in the form of library or museum, the in-house records can at once corroborate, indemnify, emancipate or hold someone accountable to a time, place, or event. No one’s looking…until they’re looking.
Eric Ketelaar reflects in his paper, “The Panoptical Archive,” that the panoptical archive, “has more than one face, like the surveillance society of which the archive is both a tool and a reflection.” He references David Lyon by saying, “society may be viewed either from the perspective of social control or from that of social participation.”
Two recent examples that illustrate the duality of an archive’s social implications can be seen firstly in Errol Morris’ documentary “Standard Operating Procedure.” The criminal investigator on the case of the Abu Ghraib scandals was able to pinpoint certain individual’s participation and whereabouts (social control) by collecting and analyzing the metadata within each photo taken by each officer (social participation, although the consequences of said participation was probably not anticipated). In this sad, conflicting, critical and embarrassing series of events, the combined metadata was able to provide certain “empirical” evidence. However, the acts, events, or people caught in those pictures didn’t necessarily paint a story of the entire truth of what happened in those jails. A photograph, once thought of as evidence or a trace of documentary “truth”, can’t tell the entire story. A picture only captures what lies within the frame. Inquiring minds ask, “What is beyond the frame?” If that question is not asked, or is not deemed worthy enough of a question to ask, the implications of photographic “evidence” may have varied and drastic consequences.
In another example, we can see how citizens offer up their photos and it’s inherent embedded data to group initiatives like Flickr as an example of positive and voluntary social participation. Researchers at Carnegie Mellon University have created software that helps identify where in the world a photo was taken.
This software matches a given photo against six million geo-tagged photos available on Flickr. By finding similarly composed shots on the peer-to-peer photo site – such as those containing narrow streets or tall buildings – the software can figure out where an image was likely to have been taken (besides the GPS capabilities of cameras nowadays).
In experiments, the software geo-located 16% of test images to within 200 kilometers of where each photo was snapped. Although these are somewhat low results right now, in another few years the technology and capability of the software will improve. “The world is pretty self-similar,” said James H. Hays, a Ph.D. student in Carnegie Mellon’s computer science department who co-created the algorithm. He goes on to explain, “We’re trying to help a computer understand an image as well as a human can.”
The impact of this technology can have as many artistic and social benefits as it can forensic abilities (for better or for worse). In either case, as users and generators of images (and the image’s inherent digital DNA), it is advisable for all of us to remember that a picture is worth a thousand words but it’s metadata can trap you or free you in those very same words
My process of assembling the archive
At first, I was inspired by the idea of a museum, such as the Textile Museum of Canada, cataloging all of its material digitally and for online use. I wondered what would happen if for some unforeseen reason, the original material was lost and what remained was simply the digital reproduction or evidence. The digital representation only acts as a trace of what once was. Is this enough for an archive to continue, to function, to be relevant if the original version is wiped out? I’m not quite sure but I think it has a lot to do with the intended use, function and intended audience of the original piece as opposed to, or in relation to its digital replication and it’s intended use, function and audience.
With the 100 pictures, I wondered how I could digitally replicate the picture instead of just capturing a thumbnail image of it. I was also struck at the process of generating the information contained within a photo – data that would make the picture relevant in hardcopy.
I chose to produce a histogram of each image as I felt a need for some sort of visual. I also thought it somewhat ironic; to provide a representation of the picture’s exposure without actually being able to see the picture. And for the interested or knowledgeable, to possibly identify a photo this way, if one remembered the originals. As well, I decided to catalog all available metadata (the more the better), including the original date/time of the picture and if that was not available (as I later found out and suspected due to a scanning process or frame grab) then to collect the retrieval data. I also wanted to have keyword descriptors of the photo as a hint of what the image contained but also as a teaser or a memory jog (to those who were familiar with the photos and might feel saddened at the loss of the originals). The ensuing result – a binder filled with pages of words, numbers and histograms was so unfulfilling, almost cold. In this respect, the information’s data cannot be considered a viable replacement for the actual picture. The only justification for the collection of this data would be the ability to input this same data in a software program that could restore the picture, essentially bring it back to life. The problem – no such software program exists right now. You can only restore photos from your camera’s memory card. Somehow (and I shouldn’t really be surprised) this card retains traces of your original photo.
Interesting patterns that I hadn’t initially considered emerged: Similar dates and times of captured pictures, similar camera make/models, similar retrieval dates of images I felt must have been grabbed online or scanned, and similar computer make/models. I was able to draw certain conclusions about the migration patterns of the individual who was capturing these images. Halfway through, my original image information capture program conked out and I had to move to another metadata collection template (you’ll notice the difference in the pages within the binder). But I was still able to collect, what I felt, was the most important – the time and date of the photo taken and retrieved. This is when the forensic implications hit me. I was creating a panoptical archive. For good or for bad – I could make certain determinations due to this “evidence” embedded in the photos.
I now had created a database of all 100 pictures. I organized this material according to picture name in alphabetical order. When I set about creating my archive, I felt it necessary to categorize the pictures chronologically so that I could see the progress of time/space creation (whether in original date/time of photo taken or in original date/time of photo retrieved). Archiving the photos this way would be quite interesting if they were visually displayed and in their original format (original prints that were scanned/captured versus original photos that were taken on the day). But once again, true to my original intention of an irretrievable archive, I was left only with the data; a result most hollow and unsatisfying in visual pleasure, but quite satisfying in organizational impulse.
What I wish now, upon reflection, is that I organized my database in order of the archive (chronologically and not alphabetically). That way the viewer/guest/purveyor could discover the patterns much more easily without needless flipping back and forth to reference the archive and then find the applicable page in the database. I should have done an actual walkthrough of how one might experience the archive in relation to the database, as one cannot exist without the other, in this particular piece. Although I believe most archives are not as transparent with their databases and/or organizational strategies.
Overall, what began as an experiment of cataloguing possibly ephemeral material ended up laden with information of which the forensic implications could not be ignored.