Ancient Lives: A Final Report

After two iterations of Ancient Lives (two different versions of the site, that is), we feel the time is right to bring the project to a close. As it stands now, we have collected roughly 1,160,000 transcribed characters, and this data comes from the work of 358,000 volunteers.

First and foremost, we want to thank all the Ancient Lives’ citizen scientists/papyrologists for devoting their time and effort to provide us with such a massive amount of data. Of course, we would also like to thank our funders: the National Science Foundation, the Arts and Humanities Research Council, the University of Minnesota’s Minnesota Futures Grant and the National Endowment for the Humanities. One of the fundamental questions the project always received was: How are the transcriptions going to be used?

After the initial launch of Ancient Lives, the work of our volunteer transcribers was used to design algorithms to process those data so that the text of the fragments could be reconstructed and easily visualized by scholars (our consensus and line sequencing algorithms) in their research. In 2016, we even published a selection of fragments that had been identified by the project in volume 81 of The Oxyrhynchus Papyri series. But with the ongoing advances in machine learning and artificial intelligence we then turned our attention to exploring the application of neural networks for automated/object character recognition. Using your work as the basis to train our algorithms, we are getting accuracy in some of our tests that are about 80%.  We still have a lot of work to do, but it would have been impossible without your efforts.

We intend to submit our findings for publication within the next year.  Research in this area will also continue. We have gathered more than enough data to do so thanks to your volunteer efforts! 

Again, we thank you for participating in one of the first handful of Humanities-based transcription projects in the Zooniverse and indeed one of the first projects on the entire Zooniverse. We encourage you to investigate other transcription projects that need your help, like Scribes of the Cairo Geniza [], which invites participants to transcribe fragments of pre-modern and medieval Jewish texts, from everyday receipts to biblical works.