Last week Chris Francese led some very constructive discussions about how and why we need to improve the automatically generated morphological analyses of Latin available. This was helpful to those of us working at Perseus because we have been struggling with how to support decentralized improvements and contributions, not just to Perseus, but to the growing number of collections with Creative Commons licenses that support re-use and new derivative works.
We have posted two blog entries to describe some of our thinking.
1) http://sites.tufts.edu/perseusupdates/2012/07/17/sdl-cfp-draft/ -- This is a draft for a possible Call for Participation to the long-planned Scaife Digital Library, essentially the aggregation of CC-licensed materials on the Greco-Roman world.
The problem that we face is that EVERYTHING can be improved or augmented. What is the starting point? The discussion about morphology was helpful not only because the morphology needs work but because it was a strong statement that morphology should be high on the list.
What else needs to be done? We need textual notes, translations, commentaries, and a whole range of machine actionable annotations. What is the logical starting point?
2) http://sites.tufts.edu/perseusupdates/2012/07/17/dighumtec/ -- This describes very current work using the Son of Suda Online (SoSOL) distributed editing platform developed by Papyri.info as a starting point for more general decentralized editing. Work underway includes (1) generalizing to other subsets of the TEI besides epiDoc (e.g., TEI-A developed by the Nebraska Abbott project), (2) integration with the Center for Hellenic Studies' CTS/CITE architecture, and (ultimately) scaling up to many languages and hundreds and thousands of contributors.
|