Epigraphic Digitization and Imagery Annotation

My query about a Hadrianic boundary marker from Bulgaria was occasioned by a demo that Sean Gillies and I (mostly Sean) worked up for online epigraphic image annotation using some free, open-source software called OpenLayers. Sean blogged about the demo, and this has provoked some inquiries from folks in the geospatial computing community, like this one from Paul Ramsey:

What’s the use case for digitized inscriptions? I don’t comprehend.

I thought readers of CurEp might be interested in the demo. I also hope I can encourage a discussion on the potential merits and pitfalls of digitally tracing and annotating inscriptions. Can we answer Paul’s question, both for him and ourselves?

About Tom Elliott

Tom is Associate Director for Digital Programs and Senior Research Scholar at the Institute for the Study of the Ancient World, New York University. Tom holds an undergraduate degree in Computer Science from Duke University and a Ph.D. in Ancient History from UNC. His master's thesis treated a class of late Roman census documents from the Aegean islands and Asia Minor. His dissertation assembled and analyzed the epigraphic evidence for boundary disputes in the early Roman empire.
This entry was posted in methodology. Bookmark the permalink.

2 Responses to Epigraphic Digitization and Imagery Annotation

  1. Well…. many folks working on inscriptions and handwritten texts draw, or trace, their own letter forms over an image of the text, to produce both an interpretation and be able to extract those letter forms to build up an overview of letter forms and hands. If they are doing it anyway, whats not to like about doing it electronically? That way, the resulting captured letter forms can easily be reproduced, repurposed, and used in further analysis, through various image manipulation, image processing, and AI techniques.

    If you are looking at digital images on screen, wouldnt it be nice to be able to annotate those images in the digital domain? and wouldnt it be even nicer to be able to *do something* with those annotations once you had captured them? howabout sharing your dataset with others (here – have my corpus of letter forms of Old Roman Cursive) – or use the data gathered from one document to aid in transcribing damaged and abraded text from another source which uses the same letter forms?

    There’s no doubt it will be a lot of work to make a useful, and used, tool – but what larks….

  2. Tom Elliott says:

    Although Sean and I don’t have bandwidth or funding to do more work on the InscriptOL demo at this time, we have licensed the code and made the Mercurial repository accessible so others can pick it up and run with it if they’re so inclined. See further: http://zcologia.com/news/694/inscriptol-source/

Leave a Reply

Your email address will not be published. Required fields are marked *