The hm Package

The hm package is an open-source Common Lisp package for investigating the chronological structure of archaeological stratigraphy.

With hm you can:

  • represent the Harris matrix as a stratigraphic acyclic directed graph (DAG) as described by Dye and Buck (2015)
  • investigate and analyze the structure of the Harris matrix using graph theoretic functions and characteristics of the archaeological data
  • generate a chronological DAG that can guide data input for Bayesian calibration software such as BCal, OxCal, or Chronomodel
  • create a graphic visualization of a stratigraphic DAG or a chronological DAG in any one of the graphics formats supported by the open-source Graphviz dot software

A central organizing principle of the hm software is the self-contained project. An hm project has its own file directory that includes:

  • two or more input files with information on
    • units of archaeological stratification, or contexts
    • observations of stratigraphic superposition
    • inferences of once-whole contexts
    • periods
    • phases
    • dated events
    • event order
  • one or more initialization files that specify
    • input files
    • classifiers that are mathematical functions on the stratigraphic DAG or archaeological constructs specified by the excavator
    • output files
    • parameters to configure the Graphviz dot software
  • zero or more output files generated during investigation and analysis

The hm software is designed to be collaborative:

  • input files are comma-separated value text files that can be produced by
    • most spreadsheet applications
    • most database software
    • any text editor
  • initialization files are the widely-used .ini file format
  • the text file output from hm is processed by the Graphviz dot software
  • compressed hm project archives can be sent via email to colleagues, who will then have the information needed to replicate a stratigraphic investigation and analysis
  • planning is underway to design an interface from hm to Bayesian calibration software