A Clojure library for exploring personality modeling in simulated populations.
Python Clojure
Switch branches/tags
Nothing to show
Latest commit c6a9116 Nov 29, 2013 @oubiwann Text updates.
Permalink
Failed to load latest commit information.
docs
sandbox
scripts
src/simulacrum
test/simulacrum
.gitignore
LICENSE
Makefile
README.rst
project.clj

README.rst

clj-simulacrum

A Clojure library for exploring perosnality and emotional modeling via software agents in simulated populations.

Installation

clj-simulacrum depends upon incanter, which has a non-standard source build. If you don't have incanter already installed, you will need to install it. We provide a conveient means of doing this with a make target in our Makefile.

If you do have incanter installed, you can skip this "Installation" step and move right to the "Project Inclusion" step below. You project will download clj-simulacrum from Clojars when you run lein deps on your project.

For those that don't have incanter installed:

$ git clone git@github.com:oubiwann/clj-simulacrum.github
$ cd clj-simulacrum
$ make deps

This will download incanter and then build it into your ~/.m2 directory. After that, you will be able to run clj-simulacrum.

Project Inclusion

You can add clj-simulacrum to your project.clj with the following:

(defproject your-project "1.2.3"
  ...
  :dependencies [[org.clojure/clojure "1.5.1"]
                 [...]
                 [clj-simulacrum "0.1.0"]]
  ...)

You can then use it in your project like so:

(ns your-project.client
  (:require [simulacrum :as sim]))

Or from the REPL:

(require '[simulacrum :as sim])

Library Usage

TBD

TBD

Interpretation

The various operations defined by the library (some of which are presented above) are explored in more detail in the :docs/resrouces/compat.rst document.

Running the Scripts

This project provides a small set of scripts that may be useful for you. They are described below.

Big Five Short Inventory

The Big Five short inventory test displays 10 questions for the user to answer and then tallies the results, displaying the five-trait break-down at the end.

To run:

$ ./scripts/bigfive-short.clj

Big Five Long Inventory

The Big Five long inventory test displays 44 questions for the user to answer and then tallies the results, displaying the five-trait break-down at the end.

To run:

$ ./scripts/bigfive-long.clj

History

Versions

  • 0.1 - Dummy release, containing only versioning info and no real code.

Project Background

This project was originally started in 2004 as "Emotional Modeling" or "Emotional Models" (depending on which names were already taken in which hosted code repository service). The purpose of that code was to attempt simulation of agents with minimal personality definitions.

Later work on other code bases (game-related ones, such as Myriad Worlds, Peloid Server) caused interest to be resumed in this particular code base. However, there is a long legacy of code in this project and a fresh start was in order. This was started, although it continued to use the Python programming language.

After creating the "cweþan" project whose ultimate intended use was for NPCs in text-based games, another Old English name seemed appropriate. After some poking around in dictionaries, "innoþ" was chosen.

Innoþ can mean "the inner part of the body", "the inside", "breast, heart, stomach, womb, belly", or in reference to feeling, emotion, etc. It is cognate to Old High and Low German words for viscera: "innethron" and "innod," respectively.

This was appropriate enough, at two levels:

  1. The obvious application of this term is for what the library is trying to model rudimentarily: emotional states (which can then be used to inform behaviours);
  2. At another level, simply to write this code, one must peer into the bowels of human nature, dig into its guts, cut it down to basic principles. This task is rather encompassing and is at the heart of any code that might be written to simulate what this inward-looking process might reveal.

Not much progress was made for many reasons, not the least of which was the relative difficulty to run massively parallel simulations efficiently in Python. This, however, was just the sort of which the Actor model (e.g., the Erlang programming language) has proved quite adept at.

With the addition of Parallel Universe's Pulsar open source project in the spring of 2013, the Clojure ecosystem now has an Actor model library with light-weight thread support (called "fibers" which only use ~400 bytes each). As such, now seems like a perfect opportunity to update this library, switch to Clojure, and take advantage of these features for personality simulation in the large.

Finally, the project was renamed clj-simulacrum, the Python code was moved into the sandbox directory, and a fresh start was made in the Clojure programming language.