Saturday, February 20, 2016

Cellular automata

The most conspicuous recurring motive in Ulysses is the urge for a drink, which can only be filled at one of the many, many drinking establishments in Dublin.  One needs money, or friends one can impose on.

A simple model could let thirsty characters wander from bar to bar by very simple rules.

- wander home after reaching preset max
- go on random bender
- get in fights when drunk

Unrelated themes to explore:
- spotting macintosh
- avoiding Boylan
- flirting, pseudonym
- seeking/avoiding law
- tempting females

promising parallel projects

Wednesday, February 17, 2016

Traffic maps

Since we know who lived where, can we experiment with varying average-daily-travel graphs, and see how much foot traffic and/or carriage traffic would be generated and where?

There are lots of unnamed characters-- can we calculate how close they probably are to their homes?

How extraordinary is Patsy Dignam's trek to the butcher's?

How common are Bloom's neighborhood convenience shops?

Monday, February 15, 2016


the pieces offer faint metaphors for the characters:

Bloom = king
Molly = queen

Boylan = black knight

Conmee = white bishop?

pawns = passersby

Tower = rook?

it might be fun to create mini-dramas like knight threatens queen

Friday, February 12, 2016

SMASS et al

SMASS is a PROLOG agent-sim

eg: "Each actor tries to find a neighbour who is physically weaker, and to take away
some part of that person's wealth."

EOS: leader/follower planning (Deasy is Stephen's leader, Bella is Zoe&co's; Molly is Bloom's, at end Bloomis Stephen's and Molly's?)

"Agents decide the resources to target, the number of agents needed to acquire them, and then seek to make an agreed allocation of agents to resources. A single agent and its following may be assigned several resources to harvest. Plans have explicit estimated numerical payoffs"
"Plan execution is also complex. It involves the setting of a rendezvous point and time, and
the pooling and sharing at the rendezvous of all the resources harvested. Further complex-
ity is created by the possible failure of some or all of the agents to perform the tasks they
have agreed to undertake, and the fact that the whole process is partly recursive -- thus a
subleader will execute its part of a plan by creating and recruiting to a subplan, and so on"
"Any inter-agent relationship is expressed in the EOS model by beliefs within the social
models of the agents concerned. Thus, to say that agents X and Y are in a leader/follower
relationship means that Agent X believes that Agent Y is its leader and vice versa. After
multiple instances of cooperation between two agents (an instance of cooperation occurs
when one agent agrees to take part in another’s plan) an instance of a leader/follower rela-
tionship may come into being. This happens if cooperation is consistently ‘one-way’. If
agent X is repeatedly recruited to agent Y’s plans over a limited period of time, then both
X and Y will come to see themselves as in a leader/follower relationship with Y as the
leader. Leaders tend to generate more highly valued plans, and they attempt to recruit their
followers first to their plan. Followers supply information to their leader about their needs
and awareness of resources, and will always wait for a plan to be proposed by their leader
(but may then reject the leader’s proposal in favour of an alternative). A leader/follower
relationship may dissolve if a follower loses contact with its leader or repeatedly rejects
the leader’s plans."
"to say that agents X and Y are in a leader/follower relationship means that Agent X believes that Agent Y is its leader and vice versa"
"agents’ misbeliefs about the existence of unreal ‘pseudo-agents’, and believed
relationships with them, may inhibit agents from attempting to ‘kill’ one another which
they would otherwise do. All other things equal, the effect is a substantially larger average
population size."

PECS: "These selfish agents belong to different neediness classes, i.e. they
have, for example, different levels  of  wealth, they are free to choose partners for
cooperation, and they typically act with their own interest in mind." (Henry and Martha?)

Wednesday, February 10, 2016

Planning paths

create a simple grid of cross streets

place a 'planner' at any point

assign them a random destination, with a schedule of rewards depending on when they reach it

let them calculate a route and realistic timeline
(ignore obstacles for now)

(could there just be a big table of distances between any two points,  and you plan by looking at the points you can move to and choose the one that's closest to your goal?)

let them follow the route, checking occasionally that they're on time

add other planners

if two planners cross paths, let them stop and 'socialise'

let socialising offer substitute rewards
(is balancing different kinds of rewards tricky?)

let them monitor that they're still on time

add other distractions (alcohol, gambling)

let planners allow for different margins of delay

(generate interior monologs for each, see below)

"i have to get to XY by Z o'clock"
"to get to XY by Z i need to be at xy by z"
"i can go via xxyy and still be on time"

simulation software can optimize paths
Defects, Overproduction, Waiting, Not Utilizing Talent, Transportation, Inventory Excess, Motion Waste, Excess Processing [pdf]

spreadsheet people's days [eg]