I am taking some inspiration for this project from one of Lisa Baldini’s (my collaborator for Contemplative Computing) early experiments in the MAIM minor project module. In a nutshell she was attempting to use the growth of bacteria as a variable factor to modulate the intentionality of the human voice.
This idea really resonated with me and I want to develop it in my own way with regards to the bacteria and consciousness conundrum which I rambled about here. I hope that the experiments I do can help plug into Lisa’s work and move her project forward as well as letting me scratch this itch that’s been growing ever since I read about bacteria data storage.
You can read the ins and outs of Lisa’s project at her degree development blog. Lisa focused on a visual gauge of bacteria because that was part of her conceptual interrogation. Alas this proved to be difficult. Changing the measurement of bacteria growth is going to be one of the changes I will make to Lisa’s project, the other major change being that I want to plug the bacteria into Twitter. Twitter is often referred to as a hive mind and I often find it described in casual discourse as a visible stream of consciousness. I don’t want to get mired in debating whether or not Twitter is in fact the early stages of De Teilhard’s noosphere (for the record I don’t think it in anyway is) but the fact that the comparison is easily arrived at is the main aspect I wish to exploit. The aim is to explore the agency of self & consciousness in relation to our embodied co existence with bacteria
Here’s how I envision the project developing, in roughly three phases, each developing in complexity. Keeping in line with what I’ve learned under Harwood I expect each stage to be contingent upon the next.
First Stage: Hands On Experiments
Where I aim to depart from Lisa’s methods is to measure the bacteria via the gas they emit. This will be costly but I am hoping for a more real time measurement of growth via this. I suspected using a microcontroller to measure gases would be a common enough employment of the systems and so it has proven (a full list of relevant links will follow as my research unfolds)
I haven’t decided on which sensor to buy yet, though I’m leaning towards CO2 sensors and working with anaerobic bacteria as the test subject.
Basically I will culture the bacteria in as airtight an environment as possible. The eventual aim, once proof of concept is established, will be to fix the sensors into a resin cap to place over the agar guaranteeing as little gas contamination as possible.
Such precise metrics are not my immediate concern. I have a few ideas of how to measure the Co2 from the bacteria (as crudely sketched below) but as I know too well these methods will have to adapt on the fly.
I’m going to feed the CO2 values into an arduino and use these values as a means of determining the rate of increase/growth in the bacteria culture. I do not know if the amount of CO2 produced will prove to be an accurate gauge of their growth but I suppose I will find out via the process.
Second Stage: Software To Do Things With The Data
If I manage to get some level of usable data from the above experiment I will do the following with it.
A program will be programmed to scan twitter on the basis of a hashtag (for example #bacteriamind). It will then take any message with that hashtag, apply an encryption or word jumbling algorithm to it and then repost the message to the user who posted the original tweet. The rate of encryption (i.e. how scrambled or secure it is) will be determined by the percentage of bacteria growth
This method is again a proof of concept and I’m leaning towards using Enigma encryption as an attempt to make some conceptual link back to Turing. If it proves too difficult to code then I’ll find some easier way to scramble the message.
This second stage will be proof of concept for the software filter that the bacteria data will plug into, and if I have time (which is admittedly unlikely) I may do it concurrently with the bacteria experiments.
Third Stage: Realisation
This is where I will aim to make the conceptual bridge between what inspired me to undertake this project and the steps taken to get there.
As mentioned the program will search for a hashtag and scramble it. The eventual realisation for this project will be that rather than wholesale scrambling it the software filter will change the content and context (i.e. intention) of the message. What I am attempting here will be very crude semantic AI so plenty of pitfalls undoubtedly await.
What I hope the project will do is that say I post
the program will pick out “loved” and “see” (I imagine the latter category will be quite difficult to accomplish, but possible via a parser that searches for any word before pronouns) and change the words, pick from a database of antonyms for those words and then repost the message to the user and any twitter recipient in the message (it will also remove the original hashtag in order to prevent endless recursion). From the example above we then might get:
The severity of the semantic swaps would be determined by the amount of bacteria (gauged by the percentage of gas emitted)
The last place I hope to take the project would be to make it into a participative game, although this last element is immensely contingent upon all the above parts moving in coordinated synchrony. For instance if the same message is posted by ten people, e.g. “what hath god wrought #bacteriamind” then rather than trigger a semantic scrambling of the message a toxic substance will be released into the agar, killing off some of the bacteria, or at least lessening their number. This is to finally link back to the idea of bacteria outnumbering us ten to one, but if our sentiment can be collectively articulated on Twitter maybe this extended consciousness can overcome our bacteria scrambled embodied intent?