Art projects

Creator: Eric Kerr

A collection of some of my photography projects.

Creators: Olivier Perriquet and Eric Kerr (2018)

Haw Par Villa (HPV) is a project based on Haw Par Villa, Singapore, that will take the form of a virtual reality experience. The project is based on the architecture of the Tiger Balm magnates the Aw brothers, especially Haw Par Villa in Singapore, originally built in the 1930s. The park is a syncretic patchwork of mythologies, religious instruction, and morality tales embodied by sculptures, dioramas, and gardens. We want to explore how this process unfolds through new technologies of 3D reconstruction and how this might cause us to reflect on our own memories both collective and individual, the suspension of time, and visionary environments.

Demo 1, Demo 2, Demo 3, Demo 4

Creators: Sarah-Tabea Sammel and Eric Kerr © (2017)

SOMETHING BEAUTIFUL is a short film exploring Wittgenstein’s thought experiment, "The Beetle in a Box". Wittgenstein supposed that our minds (by analogy, boxes) carry the concept or notion of a beetle. However, no one can see into someone else’s box. There might be an identical, a different, or no beetle at all in the other's box. The beetle as such becomes constructed and constituted via the language game; comprehension is achieved through communication and exchange that allows for the beetle to become consensus. SOMETHING BEAUTIFUL asks how a word - "beautiful" - is used, floats in-between minds, and gets attached in our society and offers the explanation that beauty is between the eyes of the beholders, onlookers, spectators. What do we agree upon? What is "beauty" put upon? Would something beautiful exist without humans creating and nurturing the term?

Premiere: F3 (Fågelbo Film Festival), Fort Lauderdale, Florida. November 2018.

Creator: Eric Kerr (2022)

The Geometric Electric Guitar is a six-string left-handed electric guitar with a Floyd Rose-style tremolo bridge. Here are some photos of the process:

Creator: Eric Kerr (2026)

Mouldy is a slime-mould variant of John Conway’s Game of Life. Conway’s version is a grid of cells that are either alive or dead. Each generation, a cell lives, dies or is born depending only on how many of its eight neighbours are alive. My goal was to make a player a part of the world. Every move changes the world. Even lingering, and "doing nothing" shapes the world around you.

Mouldy's rules are supplemental to Conway's. In Mouldy, mould grows on a substrate threaded with a slowly wandering meshwork of richer ground. On bare substrate the cells follow Conway’s rules. On the meshwork they follow a different rule set. The meshwork itself thickens, thins, drifts and changes course, so the conditions of life are always being remade. The mould's behaviour emerges out of the meeting between spore, neighbour, and ground: intra-action. Entities don’t exist first and then interact. They take shape through their relations. I use Tim Ingold’s word, meshwork, for lines of growth tangled together rather than points joined in a network.

Slime moulds are a fitting model for a game of life. Physarum polycephalum has no brain, yet it finds the shortest path through a maze. In one experiment, researchers from Hokkaido University created a precise map of the Greater Tokyo Area on a wet surface with oat flakes (which Physarum polycephalum loves) at the locations of major cities and stations. Within about a day, the mould had grown a network resembling the city’s rail system. In slime moulds, what looks like intelligence is a property of many bodies spreading through an environment: agency without a centre.

Player actions are also never perfect. You can’t place a spore exactly or draw a structure that stays put. The cursor changes what the mould can do, and the mould, by where it grows, changes what the cursor is for. The player is one actor among others although a colony might die without any player movement. Donna Haraway puts it most simply: “We become-with each other or not at all.”

One feeling you might get if you play around with the simulation is that of acting through a system you neither fully design nor fully control.

• Bennett, J. (2010). Vibrant Matter: A Political Ecology of Things. Duke University Press.
• Deleuze, G. and Guattari, F. (1987) [1980]. A Thousand Plateaus: Capitalism and Schizophrenia. Trans. B. Massumi. University of Minnesota Press.
• Gardner, M. (1970). "Mathematical Games: The fantastic combinations of John Conway's new solitaire game 'life'." Scientific American, 223(4).
• Ingold, T. (2007). Lines: A Brief History. Routledge.
• LifeWiki. "List of Life-like rules." https://conwaylife.com/wiki/List_of_Life-like_rules
• Morton, T. (2013). Hyperobjects: Philosophy and Ecology after the End of the World. University of Minnesota Press.
• Tsing, A.L. (2015). The Mushroom at the End of the World. Princeton University Press.
• Virilio, P. (1999). Politics of the Very Worst. Trans. M. Cavaliere. Semiotext(e).
• Wikipedia. "Life-like cellular automaton." https://en.wikipedia.org/wiki/Life-like_cellular_automaton
• Woodard, B. (2012). Slime Dynamics. Zer0 Books.
• Yong, E. (2010). Slime Mould Attacks Simulates Tokyo Rail Network. National Geographic.

Built in Processing and p5.js.

Creator: Eric Kerr (2018-)

Write about what you know, they say. And so: Sunday League Striker. You play as a striker trying to get the ball into the net past an increasingly difficult opponent. The game is built on the bones of a classic dungeon crawler. Instead of a dungeon, however, the player adventures through boggy pitches, corrupt referees, and hungover defenders. It grew out of the libtcod tutorial by João F. Henriques (via the Python tcod tutorial).

Sunday League is pub teams, council pitches, mismatched kits, players with day jobs, and a level of commitment that is, I argue, admirable.

Roguelikes are a genre of role-playing game descended from the 1980 video game Rogue. People like to argue about what counts as a roguelike (relegating allegedly inauthentic efforts as "roguelites") but, for me, not caring too much about gatekeeping, these are the main features which I have tried to convert into the world of casual football that I love.

• Procedural generation. The dungeon, the pitch, is randomly generated. Every pitch you play on will be different with mud patches, puddles, frozen ground, and so on. Get stuck in the mud (or trap defenders in it). Slip in a puddle and the ball might shoot off in an unexpected direction.
• Permadeath. There's no save-and-reload. When the match ends, it ends, and you start a new one. Lose time if you are tackled or the ball goes out of play. Pick up two yellow cards and you're sent off. The Dwarf Fortress crowd call this "losing is fun". We might call it Sunday.
• Unidentified pickups. Pickups appear as coloured "?" marks at first. The colours are shuffled every game, and you only learn what one does by stepping on it. Some will help you: a shield against tackles, a burst of pace, a powerful kick. Some are regular features of Sunday League: cramp, a clash of heads, a yellow card.
• Turn- and grid-based. The pitch is a grid of tiles; each arrow press is a turn, for you and your opponent.

Built with Python and python-tcod. Windows build made with PyInstaller. All code by Eric Kerr.

Creator: Eric Kerr (2019)

OVERVIEW

• Players: best with 2–6, but more can play.
• Cards: one standard 52-card pack, or two packs for five or more players. Remove the Aces before play.
• Object: to collect the most multiples by the end of the game. You collect multiples by declaring them and winning the round. A won multiple is roughly what card games such as whist and bridge call a "trick".

CARD VALUES

Number cards are worth their face value (e.g. an 8 is worth 8). Jack = 11, Queen = 12, King = 13.

KEY TERMS

• Deck: the face-down pile that cards are drawn from.
• Multiplier: the face-up card on top of the multiplier pile. A new one is turned over at the start of each round, and it changes during the round whenever a player calls Divide.
• Multiplier pile: the pile the multiplier sits on. Earlier multipliers, cards taken by Divide and losing declarers' cards all go into it, face down.
• Multiple: one or more cards whose values add up to a number that divides exactly by the multiplier. For example, if the multiplier is 9, a multiple can add up to 9, 18, 27, 36 and so on. A single card can be a multiple (e.g. a 6 when the multiplier is 3), but not a single card that matches the multiplier: if the multiplier is a 6, you cannot play a 6 on its own, though a 6 can be part of a larger multiple. Two or more cards can add up to the multiplier itself (e.g. 3 + 4 when the multiplier is 7).
• Addition: the multiples a player has won. They are kept face down next to the player's hand, each as a separate set, and counted at the end of the game.

SETUP

Choose a dealer. The dealer shuffles the pack, deals three cards to each player, and places the rest face down as the deck. Play always goes clockwise. The player to the dealer's left starts the first round, and the starting player moves one place clockwise each round.

PLAY

The game is played in rounds. Each round has three stages.

1. The multiplier. The starting player turns any face-up card on the multiplier pile face down, then turns over the top card of the deck and places it face up on top of the multiplier pile. This is the multiplier for the round.

2. Decisions. Each player, in turn, chooses one of two actions:

• Draw: take the top card of the deck and add it to your hand.
• Divide: say "Divide!" and turn the multiplier face down. Then take one card, unseen, from each other player who has at least one card, and place them face down on the multiplier pile, in any order you choose. Turn the last card you placed face up: it is the new multiplier for the round. You can call Divide only if at least one other player has a card.

3. Declarations. Once every player has taken an action, each player, in turn, can declare by saying "Multiply!". A player who declares places the cards of each multiple face down on the table as a separate set, and leaves them there until the end of the declarations. Everyone can see how many sets each earlier player has put down. A multiple can be made from some or all of the cards in a player's hand, and a player can declare more than one multiple, but each card can be used in only one multiple. Players can choose not to declare. A declaration cannot be changed once its cards are on the table, and players must not make false declarations. At the end of the declarations, all declared multiples are revealed and checked.

WINNING A ROUND

• Mistakes. Mistakes are dealt with first. If a declared multiple turns out not to be a multiple, the player loses those cards, which are shuffled into the deck, and that multiple does not count.
• If only one player declares, they win the round.
• If more than one player declares, the player who declared the most multiples wins the round. If that is tied, the tied player with the single multiple that has the most cards wins. If that is also tied, nobody wins the round.
• The winner adds all their declared multiples to their addition, each as a separate set, together with the multiplier. The other declarers place their declared cards face down under the multiplier, into the multiplier pile.
• Unwon rounds. If nobody wins, every declarer takes their declared cards back into their hand. The multiplier stays on the multiplier pile, and a new round starts with a new multiplier.
• Nobody declares. The dealer shuffles the whole multiplier pile and deals it out equally to every player. Any cards left over are shuffled into the deck.
• Players keep all their remaining cards in their hands for the next round.

END OF THE GAME

The game ends when either:

• there is no card left in the deck to turn over as the multiplier; or
• immediately after the multiplier is turned over, there are fewer cards left in the deck than there are players. For example, with three players and only two cards left in the deck, the game ends and that round is not played.

SCORING

Each player counts the multiples in their addition. The player with the most multiples wins. Cards in players' hands and the multiplier pile do not count.

If there is a tie, the tied player whose addition has the highest total value (adding up the cards' values) wins. If that is also tied, the game is a draw, known in Multiply as "equals".

STRATEGY TIPS

The game allows for many different strategies including bluff and deception.

Creator: Eric Kerr (2017)

There is a degree of randomness in any justice system. My first degree is in Scots Law and so it makes a kind of biographical sense that the first game I'm uploading is about jurisprudence.

This is not so much a game as a random sentence generator (although there are some gamey-elements, so that there is some kind of replayability). An absurd joke about the strange language that legal systems generate, and preserve, almost automatically. In the Court of Random Crimes, all that we know is the name of the alleged crime. No evidence, no witnesses, no testimony.

Many verdicts are possible, including the now obsolete "not proven" verdict from Scots law. A popular joke in Scotland is that if your judge tells you your case is "not proven" they are just politely omitting the "yet".

I think the game highlights a curious feature of language in relation to criminal justice: Almost for any crime the game generates, if you squint a little, it sort of feels right, once you are prepared to allow the odd creativity of legalese. One can imagine what crime might have been committed, in some strange legal system, where the accused is guilty of "stalemating gains while expiring agents" (to cite the example that randomly generated as I was writing this). The grammatical constructions are very simple each time. This has parallels with Derrida's reading of J.L. Austin in which he points out that it cannot be known, just by the words used, whether a given speech act is sincere or ironic, a joke (Derrida 1977). Here, it is impossible to tell, just by the words used, whether the criminal offence is real or fake.

FEATURES

• millions of possible crimes (and billions of possible events)
• defend your client (and risk angering the Lord Ordinary) or remain silent (and risk the crime being written into the books)
• 11 possible verdicts including Scots Law's now-obsolete "not proven"
• win by freeing 10 verbs from the criminal code before the judge becomes completely unreasonable!

REFERENCES

Derrida, Jacques. 1977. Signature-event-context, Glyph 1: 172-97

SCREENSHOTS

LICENCE

Random Crime Generator v0.1 (2017) is written in Python 3. v0.2 (2017) is written in Twine and Javascript. v0.3 (2026) is a major update that adds some game elements. It is an open source program licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported License.

Git is here.