To the main page for the Ontologies SIG. To the meetings page.

Minutes from the meeting in Rome October 3, 2013

The meeting was held in Rome, Italy from 14.30 to 18-00 on October 14, as part of the TEI annual meeting. There were some 20 persons present.

Presentations given at the meeting:

Invitation and programme

  1. Introduction for newcomers to the work and topic of the SIG
  2. A discussion on the introduction of TEI elements for concrete and abstract object
  3. Tentative: Should the TEI have a typed hierarchy of its elements for real world information?


The current issue of the group is the introduction of one (or perhaps two) elements for objects:

  1. An element for physical objects on the same level as person, places
  2. An element for abstract objects comparable to work/expression/manifestation in FRBR and motif, e.g., Mona Lisa, in line with the existing elements for person and places.

There is already an element 'object' in the Manuscript Description module. Can this element be generalized to all physical objects as event was extended from a specialized element for the transcription of oral material? Could this new object element be even more generalized to cover abstract/conceptual objects or do we need another element for conceptual objects like work, expression, motif, etc?


The meeting started with an introduction to the SIG and the previous work of the SIG. The programme for the meeting was outlined and Christian-Emil Ore gave a presentation of the topics which have been central to the work of the SIG the last few years. After that, we moved to the main topic for the meeting, namely, a discussion of a generalised object element.

Object element

In the previous discussion of a generalised object element for abstract and concrete objects, two problems were faced:

  1. Integration with the manuscript description module (TEI P5 chapter 10). This module was speficied in too much detail too soon.
  2. The use base for an object element. Are there any real use cases for it?


Should the object element be for text bearing objects only? What then about a blank manuscript page? This is not a manuscript, but still highly relevant for TEI encoding.

There are several parts of the standard to connect the new element or elements to: The desc element, the elements for text external features such as place, event, the systems for bibliographical information, and the manuscript description module. Christian-Emil Ore gave a presentation of the elements in this module.

It was suggested that we could remove the ms prefix from the manuscript description elements, which would give elements such as:

  • Identifier
  • Part
  • physDesc
  • history
  • Additional

It was noted that the need for a general object element is a real one; it is common to have people coming in with objects which could have been encoded with the manuscript description module, but which are not manuscripts and therefore out of scope.

The elements listed above could be added in addition to the manuscript description ones, as generalisations. This would maintain backward compatibility; however, backward compatibility is not something that should be maintained for any prize.

Would the content model of manuscript description work for all other objects? The only problematic one would be the desc element. Could we add more children to this element for non-manuscript objects where appropriate?

Then the discussion went on to different connections between textual and "ontological" features in general. It was suggested to include for every TEI element an "antimatter" element to make negative statement, such as "There are no illustrations in this ms." Another option would be a policy for negation in specified elements, as such a policy would probably not work in general.


We will work towards adjusting manuscript description elements to cover all objects, text bearing or not. In a previous attempt to do this for paintings, it turned out that only one single element was missing, which could easily be added.

We will make a proposal to the TEI council for this. Sebastian Rahtz will prepare a document to council for their next face to face meeting based on the notes from this meeting, which was sent to him.

This concluded the TEI Ontologies SIG meeting.

2013-11-11 Øyvind Eide