For my Investigate brief I decided to interrogate the area that compounds my doorway.
I started of by gathering as much information as I could and then trying to understand it by question it. ie. I saw door, wall and wood and visually tested how would it be if it was just door, just wood or just wall in terms of texture. And then I tried what would happened if these surfaces where swapped around.
My second attempt of understanding my doorway consisted on mapping the space held within this area. I researched how to map space and ended up finding answers on how to make a map of star systems. The answer had to do more with the routes the stars take since the hyperspace connections might not have anything to do with the regular space distances. This taught me that they key thing to track was the movement within the space in order to be able to map the space it takes.
I plotted 15 sheets of A4 paper across 5 columns and 3 rows covering the entrance of my doorway. Labelled them with letters A-E on each column and 1-3 for each row. After a week, I scanned each individual sheet and regrouped them as tiles, bringing back the original layout. I then inverted this image into negative to highlight the dust, stains and scratches. This allowed me to see any patters in the movement my doorway of a week.
Having explored the visual and spacial aspects of the area I was researching, I moved onto the usage of this space and how we interact with it.
I set up an analogue camcorder to record how we behaved within this space. The tape was only 20mins long, which meant I had to rewind it every time and check wether it had trapped any action. After hours of overwriting footage, eventually it captured one of my housemates crossing the doorway.
During our first tutorial we shared the different methods of investigation we had explored so far. I realised the aspect of my project that weren’t working were the risk of finding information I might already know and the fact that I was paying attention at things that were too obvious.
We discussed our progress in pair. My classmate Daisy made a very interesting observation regards the 3rd route of my investigation. According to her, instead of seeing the doorway in my recording, she was seeing me looking through my camera at my doorway – meaning the study had become the subject of my study instead of the original area. This could be because of the tools I had used. The tape was recorded on an old analogue camera, which was then digitalised with my phone, leaving the information that is only seen while recoding, not supposed to be left for the viewing (ie. remaining camera’s battery or minutes of tape left).
Standing too close of an image can distort the perception. The longer we spend without stepping back, the easier it becomes to obsess about individual areas, loosing visual perspective. Many exquisite small areas within a painting for examples do not mean that it will hold up from a distance. Conversely, a few flawed areas that are part of a strong composition do not matter much.
A way of looking at the bigger picture was to export all the frames of the video and view them altogether as a contact sheet.
This didn’t give me any extra information that I didn’t have already. However it did remind me of the techniques we use for painting when we step back from the easel, which are usually followed by squinting your eyes in order to visually simplify the values in the scene, to eliminate the perception of reflected light, and to better see the whole.
There is a digital form of squinting, which is to downsize the resolution of an image. I compressed as much as I could each individual frame of the video. Enough to be able to still understand the images. Then put them back together and exported it as a video.
Through the above process I managed to have an overview of my recordings. Now I wanted to get as close as I possibly could and explore the details.
Some people might think that the smallest element that forms a digital image is a pixel, but that’s not totally true. Those pixels are formed by code. Therefore I opened the code of the video and tried to make sense of it.
At a glance, nothing stood out in a way that I could understand. Maybe I wasn’t able to read the code but would be able to gather something from the sounds of the script. I then entered a fragment of the code into an online text reader and listened to the audio. I felt like I had hit a dead end of my investigation without really finding any new information.
After revisiting the process I had followed, I found (gleaned) a couple of things that did seem like they shouldn’t go to waste.
- Looking at all the different parts to understand the whole
- Seeking languages of information that I could process
Adding up those methods meant opening the code of all the different images that formed the video and mashing them up into one single file.
The outcome of this was a glitched image made out of different fragments of the video. Even though it came across quite abstract, it was a capture of everything that happened in the footage, giving me a new visual understanding of the information I was gathering.
There was more information left in the code I could harvest. I recognised many words, and was still eager to understand how the code formed the image. A simple approach to see how the code works, what to see what could break it. I tried deleting every single word I recognised to see if there was a forbidden one that the images couldn’t work without.
I thought I had found it when the file crashed. To double check the outcome was correct, I tested deleting each individual word at a time. It turned out that what corrupted the file wasn’t deleting an specific word but reducing the amount of characters of the entire code.
I still wanted to see if the code could generate an image without any words. The question was how to remove words without loosing characters.
The way around this was to replace all those words with one that didn’t mean anything and wouldn’t affect the code. I turned out that the average amount of characters per word in English is 5.1 and the most common letters used are E,A,R,I,O. Therefore the safe word had to be Eario.
I replaced every single word I could understand in the code with Eario and effectively the file didn’t crash, just distorted more the image but not drastically.
At this stage I knew that altering the number of characters would break the file, but I didn’t know yet what information is needed to generate an image. There were no words left, but there were numbers that looked like dates. I replaced all of them by 0123. Barely did any damage to the file.
Now the question was if is it the digits that the image can’t work without or is it the letters. I tried both routes, by replacing numbers with letters and vice versa, both files crashed. Finally had a clear discovery: code can generate images without words but still needs both letters and numbers.
I brought back this knowledge to the initial brief by opening the very first picture I took of my doorway and testing what would happen if I remove from the code all the information I don’t need. The only relevant information of my doorway its probably its name, it’s identity, aka my address. 32 Burgoyne Road SW9 9QJ
I replaced every single character that is not in my address by duplicates of the ones that are. My surprise is that this barely affected the image, which raised the question that wrapped up my interrogation: why the need to include so much irrelevant information? It’s MY doorway. I took a picture of it with MY phone, which I then viewed on MY computer. If the only information needed to generate this image are some characters, break lines and spaces, why do they flood the file with other information? Why do they flood my doorway with spam through my letterbox?
I do value the access to information we currently have. It’s by far the biggest advance we’ve made over the last decades. I do appreciate the positive impact that this has had in many aspects of our lives. But why the excess? If I’m taking a private image, of a private space to view it with a private device, what information needs to be there other than the one that generates the image?