6 ppl attended
Open exchange and announcements
- Even more summerschools:
- atRium Brno Training School on computational archaeology
- https://petrpajdla.github.io/atRium
- funding available for participants
- planned for 16-20 September 2024, deadline for applications is May 31st
- Summer School in Digital Palaeography
- Neural Networks for Archaeologists, with Python
- atRium Brno Training School on computational archaeology
- EUROPEAN DS4CH: European Data Space for Cultural Heritage
- Flagship initiative of the European Commission to accelerate the digital transformation of Europe’s cultural sector, and foster the creation and reuse of content in the cultural and creative sectors
- https://www.photoconsortium.net/european-ds4ch
- GitHub & ORCID collaboration
- https://info.orcid.org/orcid-and-github-sign-memorandum-of-understanding
- it is now possible to link GitHub and ORCID accounts
- New software tool for PCA analysis in genetics (triggered a discussion in the group about the use of PCA in archaeogenetics)
- R packages to make PCA and other methods of multivariate statistics available within tidy(verse) workflows
- Online Survey of Cultural Heritage Professionals
- TETRARCHs: Running seminar series on computational archaeology
- Archaeological paper in Nature by Riris et al.
- “Frequent disturbances enhanced the resilience of past human populations”
- https://www.nature.com/articles/s41586-024-07354-8
Project discussion
- Context: Reviving old archaeological research software tools
- Concrete project: Reviving the “Tools for quantitative analysis” (http://tfqa.com)
- James Allison contacted the author Keith Kintigh:
- Prof. Kintigh appreciates the effort of rewriting some of the tools in R; he would even be willing to get involved directly
- TFQA was a commercial project, but there is no significant demand for the old distribution, so making the rewritten code/tools openly available is OK
- Some tools are computationally intensive and optimization may be an issue in R
- There are some potential numerical analysis issues in computing the factorials for binomial probabilities for some of the tools - developers should rely on established numerical libraries and not implement this math from scratch
- Matt Peeples has already implemented some of the tools in R (https://mattpeeples.net/data-and-software), but his focus was on the core algorithms - wrapping them in a convenient, well documented functions is still an open task
- Various TFQA tools are described across different publications. See the discussion in the Google group for some relevant papers
- Potential targets/low-hanging fruits to work on:
- An R package compiling different archaeological diversity measure algorithms
- A package or a set of blogposts/scripts reproducing some of the spatial data analysis tools
- A key feature of TFQA is domain-appropriate resampling (e.g. temporal or spatial resampling), which should be preserved in any rewritten version
- Technical observations about TFQA:
- Multiple of the open-source TFQA tools are written in Delphi, a dialect of Pascal (https://en.wikipedia.org/wiki/Delphi_(software))
- They were usually compiled for MS-DOS or MS-Windows, some compiled executables do not run any more on modern Windows versions and require emulation
- The tools typically have a simple TUI (terminal user interface) that guides the user through the input preparation
- Some of the tools are pretty complex and include multiple subcommands
- Further discussion should happen on GitHub in dedicated issues at https://github.com/sslarch/tfqar
Next SIG Meeting: Wednesday, June 5, 2024
We invited Nicolas Frerebeau to introduce the tesselle R package collection: https://www.tesselle.org
This was originally planned for last session, but had to be postponed
