Between the Caves

color reversal: Unknown film. SBA settings neutral SBA off, color SBA onThe project “Between the Caves” started in 1993 when Meg Conkey began a regional survey in the foothills of the French Pyrenees looking for Paleolithic materials. Up through the mid 2000′s Meg and her team surveyed plowed fields and vineyards in that region and conducted some test excavations. Meg’s challenge today is to access the abundant and diverse materials from the survey and make it accessible for publication and for further research.

We of course have data and information in many forms, since 1993 was some time ago in our data collecting lives: starting with paper forms –used all the way throughout up through more than 360 different site locations, and moving to Zip discs, etc. I have a lithic inventory of more than 6000 artifacts, digital terrain maps, some GIS data, detailed hand written field notes, hundreds of first slides, then scanned, and then digital photos. A major portion of the scanned slides were also keyworded with an (earlier version of) Portfolio.                        Meg Conkey

In order to meet Meg’s needs we started the BTC/Codifi project in January 2013. The main goal was to harmonize all data into a coherent and integrated Codifi database, and at the same time build a sustainable methodology for the future.

We kicked off the project creating a workflow and shared tools for the team to proceed fast and efficiently through coordinated and collaborative actions. Following the Codifi methododology, the original data were made available to the team for data auditioning and for modeling the data architecture, but the original files were and are considered “historical sources” and remain untouched.

After the data auditioning phase we identified our project’s goals, located all relevant information, and we built our data model. Data modeling is a process that brings structure to your content and organize your universe of different types of data into structured information that is easier to understand and manage. So, during this particular phase it was critical to work closely with Meg and have her constant feedback on how to better model all the parts together.


We are working on data cleansing and processing as we go. One of the longer but more interesting phases of the project was to convert and export the photo catalogs of the project from the proprietary Kodak format into a usable format and transform the keywords into usable data (more on this in a future post).

Screen shot 2013-03-26 at 12.25.52 PM

Another important step was the extraction of useful information from Meg’s rich and detailed fieldnotes.  Digital scans of the fieldnotes are now available as multipage PDFs. The scope of this phase was to extract relevant information for the surveyed sites, and create entities that could be mapped out and linked to the entities from other sources.

Screen shot 2013-05-07 at 2.23.45 PM

We have completed the identification of entities and relations, and the mapping into Codifi. Main outcomes so far are:

  • all photos and fieldnotes in catalog with rich embedded metadata
  • extracted metadata from fieldnotes and graphics
  • media, fieldnotes, gis and excavation data semantically connected
  • concepts, keywords and site records harmonized into a single, extensible framework

Using Codifi is not just gathering old parts into an archive, it is creating a new enhanced source of information that allows you to think differently and develop all sorts of new ideas!            Meg Conkey

We are presently working on layout and interface and will be posting some updates soon, stay tuned!

Leave a Reply