A splitchef trial- Final Project Test Run (my perspective)


Disclaimer: Do look at all our OSS for the roles and perspectives  of all the different players and moderators involved!


For our final project, Dion, Joel, Bryan and I wanted to utilise an idea of a 3rd space cooking project to fulfil all elements we have learnt from Experimental Interaction. We call it Split Chef.

Essentially, this project entails two group of people: the moderators and the players. The players will further split into two sections: one being the cooker who looks at the drawings while the other one is the drawer. The moderators will be the one ensuring the process is running smoothly and ensuring the project is not compromised in any way.

The project starts with the drawer drawing out the ingredients needed and showing it to the cooker through Insta-live. The cooker will have to identify and guess, afterwhich proceed to buy all of the ingredients. This applies for the later section where the drawer will draw the recipe and the cooker looking at it to guess how to execute the cooking process.

DIWO applies where we use many people’s action to snowball into a final product. Glitch applies when we see an ingredient out of its conventional usage or capability, since the cooker would not know what the ingredients are for. 3rd space is used through the completely “handicapped” Instalive drawing communication without linguistic cues. We also hope to use Instalive’s comment section as a crowdsource opinion/helpline.


If you haven’t seen the video on any of Dion, Joel and Bryan’s OSS, here are the videos of the test run.

As a moderator, we wanted to ensure the sanctity of the project is run through, which was to create a dish through non-linguistic communication via a 3rd space, and without prior notice of what the dish. It was tough ensuring that there were not any loop holes that would tarnish this purpose.


The set-up was difficult as we had to think of a recipe that would involve several elements, and we couldn’t ask Bryan or Joel for opinions since they were the players for the test run.

The recipe we are going to use have to avoid all of these potential problems:

Number one, the dish couldn’t be too easy to replicate or have too little fundamental steps of cooking. Pasta would be too easy since its boiling noodles and adding a heated sauce. There weren’t dexterous steps that can be complicated with non-linguistic cues like drawing.

Number two, we have to select players according to how we want our final project to be. Essentially there are players who have stronger cooking instincts as cooking is an process drawn on experience, and that experience can change how the players approach this project. For example, Joel was experienced enough to understand that some ingredients selected was theoretically “correct”. He picked out lime after seeing the previous ingredient of tobasco sauce.

Number three, we have to ensure a strict budget in case the players assume that there is an ingredient needed that wasn’t on the intended recipe. For example, Joel bought a bag of dumplings thinking that it was the key ingredient, where it is in fact, shell pasta.

Shell Pasta- Dumpling

Number four, adding on tho the shell pasta. It is not actually a key ingredient, but in fact it is an ingredient used to throw the players off a linear thought. This is done so because we did not want the players to assume that they are cooking a specific dish because they are seeing some key ingredients, such a as the tortilla wrap they had bought prior to this.

Tortilla Wrap

Number five, if you have seen Joel’s OSS, you would know that a small chicken was bought in the process of grocery shopping. As a moderator, this snowballed a chain of wasted time since the chicken took a longer time than expected to defrost and this would be a problem on the actual shoot day since we would be asking people out of this team to perform the SplitChef routine. We would have to dissect this problem of time consuming glitches without interrupting the sanctity of this project.

A last minor problem, the lady at the supermarket wasn’t very happy that Joel and I were filming inside the supermarket. But at that point of time, we were mostly finished, so oh well.


The cooking process itself was involved with a lot of trouble shooting.

Number one, the pantry we borrowed was ideal as it was near the supermarket. We had followed Lei’s advice of choosing a location near the market in case the players lose their mojo of cooking since we do not want the excitement of the project to lose its momentum. However, we did not know how thick Hall 2’s walls are as the walls completely blocked out our phone’s data and wifi. This created a heap of mess as our 3rd space communication was impeded, and that was a problem since this communication is what builds our project. We had to borrow wifi from a student living beside the pantry, but that wasn’t ideal also since the wifi itself was 1 bar and barely sustaining.

glitched out so bad

Number two, as moderators, we did not prep the area well as we lacked in a lot of equipments. We did not have enough oil or even a  chopping board, which was quite a comical sight.

chicken on the pantry
panning the pans

Number three, we had to familiarise the environment. Even though we brought pans (courtesy of Joel’s house), the induction cooker there was crazy and didn’t work well with his pan. This resulted in only one working induction and this might be a problem on the actual day.

Number four, as moderators we do not know how extensive we want to be with our rules and regulations. In this project, we function on the idea of non-linguistic cues as highlighted before. However, when we see the process of the cooker and drawer “communicating”, we can more or less tell if they are heading for the right direction. In cases where we can tell that certain decisions wouldn’t end well, we do not know if we can allow the cooker to proceed out of the drawer’s direction. For example, our friend Zhen qi who helped us draw, drew a “20s”beside a chicken. This created an assumption on Joel’s part to cook the chicken for 20 seconds, which was ludicrous. He asked if he can continue cooking beyond the 20seconds and we didn’t know how to proceed from there.

This snowballed into a lot of problem as  we realised that we had to be strict with our rules in case the project falls out of its initial intentions.

Lastly, since I was on site with Joel, the cooker, we experienced a lot of problems communicating with the drawers, Bryan and Zhenqi, as well as the other moderator, Dion. We realised that there were a lot of set up issues in accordance to the camera angle and time spent waiting. Breaking this into two points: Firstly, it is a process thats bent heavily towards a strictly timed process, we tend to forget that and waste a lot of time idling or waiting for things to happen. Secondly, this time wasted maybe unprofessional on the day itself when we invite people to play Split Chef with us.


Overall, it wasn’t that bad of a test run as it allowed us to see the minor defects in the process. There were small hiccups here and there but overall the ideation was clear and smooth. Looking forward to the actual shoot.

Leave a Reply