Proposed Activities
 
Ideas for exploring Computer Science at a fete.
 

The proposed activities may or may not take place on the final stall. This is a space where I am working out possible ways to explore ideas directly connected to Computer Science as fete-style tabletop activities. I would normally do this privately in a notebook, but as documentation is such a key part of this project, I am opening that up here too.

What I am looking at in creating any new work is to define the context, then to deliver what seems to be the most right thing for that context. Sometimes those definitions and responses are very difficult to put directly into words, but in this project that bit is actually very straightforward. I have a finite and pre-defined amount of money and time. I know when, where and roughly for how long the activity will go on. I know that the work is a table-top stall as part of Home Live Art's Science alternative Fete. I have previous experience in similar contexts so a fair estimation of audience behaviours. I know that the work is in an education context, as part of Royal Holloway's Science Festival, and that the theme is "Storytime Science - exploring science found in children's stories". I know that the work and process needs to be documented. I know I am paired with the Computer Science department. That is the general context, which gets refined and more specific as the project continues. I will have a site visit to explore where the fete will be held, I will learn more about what Computer Science means to this particular department (it is a huge field) and what individual scientists I engage with are working on.

Proposed activities on this page will shift and develop throughout the process, and I will keep a record of that here.

 

Chatbot
I could write a simple chatbot as a useful agent for documentation - I could have conversations with it throughout the project. It would also help me to find out more about people's expectations of computers and computing. Learning to program and write a chatbot will be useful to get me into the discipline of programming, which I haven't done for a very long time. The chatbot could also talk live on the stall.

> This was proposed before contact with the department, and I began programming it right away, to use in documentation. This became Pinnochion#1.
> I wrote Pinnochion#1 in AIML as an open-source, popular AI language, which has been brilliant for understanding from the ground up. However, the limit of it responding to once sentence at a time made it not as useful as a documentation tool. I have begun building Pinnochion#2 on The Personality Forge, which will make a less controllable bot, but one that I could have longer conversations with. It's very time consuming though, and may or may not be the best use of limited time.
> There may also be an AIML interpreter that could solve this one sentence at a time problem in the same way that PF does, but responding to the last sentence only?
 

The Pinnochion
The audience can converse with an artificial intelligence with interchangeable "heads", each looking very different. The AI inside would be identical, but I would be curious to find out whether people would have very different conversations depending what the thing looked like. It could even be wired up to allow them to talk with everyday objects or a wider variety of percieved "personalities". I am interested in exploring perceptions of what a computer is capable of, as well as individual visual prejudices affecting conversations had with an identical "mind" each time. Pinnochion is a reference to "Pinnochio", the raspberry Pi and the cornish dialect word for apron, which is "pinny", as this is an apron of expectation hiding an identical core. The 'on' is to make it sound a bit sci-fi, and a bit ancient like the necronomicon, and also is a reference to binary.

>Proposed before contact with the department. Possibly too much focus on the cmputer as a machine. But there's something in it regarding expectation that could be useful.
"The Pinnochion" became the name of the project as a whole, as it contains many of the things I want to explore, and also anchors the whole project clearly to a childrens story.

 

The Struggle
Paper version of an AI chatbot. The spectadors experience what an ai program is actually doing from the bot's perspective. I want to open up the processes that happen within a program in a social, non-electronic way. This is an activity for pairs (or more, they can clump up). They sit with a screen separating them. There is a hole in the screen. Behind the screen, the person acting as the "AI" has an alphabeitcal card index loaded with things the "human" might say. So if the human says "hello", the "AI" has to search for the "hello" card and either read the response or hold the card up to the "screen". (Im not sure if this is more interesting written or spoken yet.) If there isnt a card for what the person says, the "AI" can write a new one, so the shared knowledge on that side of the screen grows throughout the day. After, I can program these categories into real chatbot that people can look up if they like in future. This is a nice kind of active documentation.

> Proposed before contact with the department. But as a way of opening up what a program is doing, I think it is a go-er. Likewise the difference between cleverness and learning.

 

The Copy Box
There is a model on the table, and some clay. People can make a copy of the model, take the previous one and leave their own. This leads to a chain of evolving copies throughout the day. The chain and their relationship to each-other only exist in the documentation. If the computer science department does 3d computer modelling or has a 3d printer this can be compared. Likewise if they make mathematical or theoretical models it can be used to illustrate the difference between kinds of model. Also used to illustrate that computing is often about repetition. Computers can copy and repeat pretty much indefinately.

> Proposed before contact with the department. I really like this activity, but Im not sure if it is connected enough to the Computer Science carried out in the department.

 

Memory Beads
The audience store something they want to remember in a string of black and white beads using ASCII. This is to get a real sense of how information is broken down, and how even something as simple as a letter has to be broken down into binary. It could give an idea of the sheer scale of memory involved in modern computing, and the incredible effort and persistence of the early pioneers. Also link to prayer beads and early computational methods using knotted string. Representation and storage of information. And the spectadors get something precious to take home with them.

> Proposed before contact with the department. Still very useful, if a little basic. Needs tweaking so as not to be too much about the computer as a machine.
 

Delete
Counterbalance to Memory. Spectadors can put things they want to delete through a shredder. Reference to things being deleted not actually being gone, just not accessible any more. The shreddings are cast/used as paper mache or made into paper pulp and flattened into paper to be written on again.

> Proposed before contact with the department. Possibly too focused on a function of the computer as a machine rather than computer science.