Annotation and Alignment of Texts and Objects (Perseids, Wikimedia Commons, Flickr) (March 9)

Annotation and Alignment of Texts and Objects (Perseids, Wikimedia Commons, Flickr)

March 9, 2016: 17h00-18h15 CET

Marie-Claire Beaulieu (Tufts University), Gabriel Bodard (ICS London), Tim Buckingham (Tufts University), and Pietro Liuzzo (University of Heidelberg)

• Slides
• Live Broadcast

---

Common session plan

1. Overview of annotation and alignment (GB) (10 minutes)
2. Perseids image/text alignment (MCB/TB) (15-20 min)
3. EAGLE Wikimedia Commons identification/annotation (PL) (15-20 min)
4. Flickr image annotation (GB/MCB) (15-20 min)
5. Exercise

Overview

In this session we will look at practices for annotating and aligning text, image and object information. We will see examples of projects where annotating and aligning have been useful, exploring why to annotate images and texts and why to align datasets.

Required readings

• Baumann, R. (2013), “The Son of Suda On-Line.” In Mahony/Dunn (Eds.), The Digital Classicist 2013. The Institute of Classical Studies, University of London. Available: http://ryanfb.github.io/papers-BICS/sosol-bics-draft.pdf
• Bigi, F. (2014), “Towards an EAGLE Standard in Translating Inscriptions.” In: Orlandi/Santucci/Casarosa/Liuzzo (Eds.), Information Technologies for Epigraphy and Cultural Heritage. Proceedings of the First EAGLE International Conference. Roma, pp. 167–178. Available: http://www.eagle-network.eu/wp-content/uploads/2015/01/Paris-Conference-Proceedings.pdf
• Tarte, Ségolène (2011), “Digitizing the act of papyrological interpretation: negotiating spurious exactitude and genuine uncertainty.” Literary and Linguistic Computing 26.3, pp. 349-358. Available: http://llc.oxfordjournals.org/content/26/3/349.full

Further readings

• On Perseids see http://sites.tufts.edu/perseids/publications/publications-and-presentations/

Assignments

EAGLE: Try to do one of the following

1. Identify and align an image
   • create an account in Wikimedia Commons
   • select a photo of an inscription you like in this category
   • read the text and eventually the metadata, search for it in EAGLE portal, using either the text or the image search.
   • if you find it edit the Wikimedia Commons page entering the link to the database where more information can be found (take it from “Original Source” on the EAGLE record), if possible according to these wiki-templates
   • categories for the main references, e.g. AE or CIL identifiers following the format already in use in Wikimedia Commons for CIL and AE
2. Enter a translation and align it to a text
   • Enter a translation of an inscription in the EAGLE Mediawiki, following the Guidelines in the wiki, making sure you enter the identifiers.

Flickr: see https://docs.google.com/document/d/1G_Npie5Pyv4viRFgjPo4ZWQOmgR9iOXTN9-w1qrsRW8

Essay question

• Discuss with reference to at least two projects, the value of annotation or alignment of textual features for research.

Lecture notes

1. Overview of annotation and alignment (GB) (10 minutes)
2. Perseids image/text alignment (MCB/TB) (15-20 min)
   • General Overview of Image annotation
   • Discussion of the imgspec tool on Perseids (visual demo)
   • Magna Carta workflow, Inscription annotation workflow
   • Showing student publications
3. EAGLE Wikimedia Commons identification/annotation (PL) (15-20 min) This part will look at some cases of alignment and identification
   • Unidentified objects: entering categories or templated information to identify an object
   • Translations of ancient inscriptions and alignment with texts (the EAGLE Mediawiki, The EAGLE portal, Wikidata, Perseids)
   • Text-translation alignment via unique identifiers used in EAGLE benefits the portal, scientific workflow, benefits other projects, e.g. Attic Inscriptions Online
   • Text / object contexts: cases in which an edition of the text is not enough information
   • Aligning text and portions of images: a manuscript example page with two different continuous texts
   • Disambiguation and identification of text: plurality of online editions
   • Unidentified photos in Wikimedia Commons
   • Availability via common protocols for reuse in project, e.g. via IIIF for better alignment
4. Flickr image annotation (GB/MCB) (15-20 min)
   • theory and possibilities
   • case study, Mediaeval Latin manuscripts
   • the exercise/practice for remote students