Announcing diGNU

A new open source software for archaeologists is coming

As posted on and

Hi all, This appears to be, more or less, a quick presentation of the startup and of our main project, so if you know a better way to spend your time today, simply ignore this.

Who runs what?

OIA (Open Ideas for Archaeology) has been started in the past months after a funding we received from the Apulia Region government ("Principi Attivi 2010"), focused on innovative ideas from new young entrepreneurs. We are a team of 4 archaeologists with different backgrounds (osteoarchaeology, ceramics, survey, etc) and a conservation scientist (hey, that's me) interested in open source GIS, FLOSS and GNU/Linux applied in the archaeological field.

What's the mission?

Create and offer open source tools to archaeologists through the web, to easily manage on-field and post-processing works and elaborations.

Be more specific, please!

Our main project is called diGNU, GNU licensed, GeoDjango spatially-enabled server-side archaeological management tool, to easily access, organize, modify and communicate archaeological data during and after the excavation. Follows a quick list of the main features; diGNU will:

  • organize and query all geographic and survey data using PostGIS
  • offer a (hope for) rocking integration with an open source GIS system GUI locally installed (QGIS?), accessible from remote server
  • export all catalogue data using LaTeX, with specific templating support and a near real-time map rendering integration for small areas
  • support the archiving of single objects 3D meshes in the database, and integrate them in the LaTeX generated printed version of catalogue data
  • integrate an easy install process using the Ubuntu/Debian package system (apt, PPA)
  • export cartographic data using Mapnik, in various formats (PDF, SVG, JPG/PNG) and customization possibilities
  • automate matrix creation using GraphViz

Other features we will likely introduce:

  • full support for osteoarchaeology and paleopathology data, comprehensive of detailed data sheets exportation, using LaTeX
  • support real-time collaboration between field and office connecting databases with mobile GIS (gvSIG mobile?)
  • a full-featured quantitative archaeology analysis system, based on a KISS web interface to R and using PostGIS data
  • here we are: a plugin system to easily add features to the whole product

Where we are now?

At the starting point. I'm writing the django basis on which diGNU will run. Yeah, it sounds like a glue-project taking the best of open source tools and offering them to archaeologists on a silver plate, like the "Wordpress" for archaeologists.

Do you have any deadline?

The main features must be completed and rocking in 16 months starting from today.

OK,show us the code...

We will bring out the first public release in December, hopefully.

Why are you writing here?

Because diGNU is a gift to the entire FLOSS-powered archaeological community, and we need your help to define some concepts, starting, for example, from the database model. We need your experience, needs and requests to create a product modelled on the user base, on the concrete needs and on the dreams of every archeo-geek.

I've already read some discussions on the database model, starting from iadb iadb. We need to know your approach, assuming that nearly we don't have any standard, today.

OK, give us contacts

At now, we don't have a website, will be up in some weeks, but we're already working on django, database and PostGIS issues. If you're interested in the project or have any suggestion, please append it to the discussions that will take place in this ML, or simply mail me at