/statement.ssla.roundtable.caa2019

Statement for the CAA2019 Roundtable Scientific Scripting Languages in Archaeology - Limits and Opportunities of Open Research

(Don't Fear) The Reaper. A SIG as long term software maintenance network?

For about 50 years archaeologists have been using computational methods to answer archaeological questions. A considerable part of the many thousands of resulting papers relied on interpreted languages. Unfortunately, most of this software was never published or distributed in a reproducible way.

I suggest the Special Interest Group (SIG) for Scripting Languages in Archaeology (SSLA) to become a vehicle for software sharing and long-term maintenance. It should provide a collaborative and organizational framework for joint work and responsibility to create new and to prevent further loss of already established methodological knowledge.

There are several organisations trying to provide this service in a research subject agnostic way, most notably the rOpenSci community. rOpenSci members develop research software in R, provide code reviews for their colleagues, promote releases and use cases, and pick up relevant projects that are at risk of getting lost due to -- among other things -- changes in employment situations, illness, or even the death of maintainers.

I admire this project, but I also believe that it has two main weaknesses: Its limitation to R and its interdisciplinarity. Neither is inherently bad, but for the specific applications of archaeology and cultural anthropology, a more thematically focused and technically open community might be more appropriate. The SIG SSLA could close that gap.

How exactly could this work?

  1. SSLA members could register sufficiently generic projects as part of the SSLA software family. These projects would have to meet certain requirements concerning reproducibility and maintainability, but could then enjoy the stewardship of the SSLA community.
  2. SSLA members could enlist in a network of mutual technical support and code review.
  3. A quarterly report could list developments in the Archaeology-Scripting languages world in collaboration with already established software lists, like e.g. Ben Marwicks CRAN Task View: Archaeological Science or Zack Batists Open Archaeology Software list.