Login Form

To register please get in touch via the "contact us" form in the main menu.

Designed by:
SiteGround web hosting Joomla Templates
eSAD: e-Science and Ancient Documents, Powered by Joomla! and designed by SiteGround web hosting
Subjects discussed in the "Quo vadis" sessions at the workshop PDF Print E-mail
Written by Segolene Tarte   
Thursday, 21 May 2009 16:31

Please note that the workshop talks are available for download here.

1- Morning session

About letter shapes - Is there any way to deal with statistical letter shapes? Melissa’s work did build statistical models of letter shapes for the Vindolanda ink tablets, based on positional probabilities within a character box. Another statistical modeling method would be by Principal Component Analysis, provided that enough samples are available. One question is then: “How much do letter shapes differ between ink tablets and incised tablets?”, and also “How could one ’compensate’ for the shapes not represented in the statistical model?” ’Compensating’ isn't quite the right term; what we want is in fact to inform and feed the database (and the model) with the newly found instances of letter shapes as well as consult it. For example in the case of the Frisian tablet (shown during the presentations), dated as a 1st Century AD tablet, a quite unique palaeography was found; ‘A’ as shaped in the document, would not have been identified as ‘A’s according to our current knowledge of shapes of ‘A’s in 1st Century AD Old Roman Cursive.

About ontology - Percepts were defined as ‘elementary perceptions’ that are already to some level an interpretation. How does that fit into an ontology? An ontology is seen here as a kind of ‘super’-data-structure, that allows to separate content from the domain-specific structure as well as characterize relationships between ’objects’ in the ontology.
About expertise - The problem of expertise was pointed out. Experts are notoriously bad at analysing what they do and they tend to “jump to conclusions”. With that respect, it was suggested that observing non-specialists undertaking the tasks the experts do might be a valuable source of information and help decompose cognitive processes that are ‘built-in black-boxes’ in experts.
About other kinds of imaging - Spectral imaging is a very useful tool when dealing with ink texts. It is the chemical composition of the inks that provokes a reaction (or not) to specific wave-lengths of light. That kind of signal amplification is bound to be unfruitful for incised texts – unless the incising tool left substantial (despite residual) traces of itself in the incisions. Shadow-stereo is a different way to amplify the original signal, in that it accentuates the features of the text by casting shadows and projecting highlights.[...]

2- Afternoon session

About using and building knowledge bases - Perseus was cited as a database that would be useful to link up to the ISS, in addition to the Vindolanda knowledge base, which is being put together. Regarding this knowledge base, lemmata were confirmed as one of the search criteria for word searches; the problem of spelling and how an index represents misspelling was also evoked, as well as evolving (!) Latin grammar.
About analysing the usage of the knowledge bases - It was suggested that we could record the statistical usage of the built knowledge base, which could help gather information, e.g., on word frequency.
About flexible use of databases - Considering to allow optional inclusion or exclusion of a database to search, by project and according to the user’s needs and choices was suggested.
About designing an Interpretation Support System - It was mentioned that the Googlemap model was adopted for collaborative work on and annotations of music manuscripts in Harvard. The tool is called Inventoriana. This layered-type design might prove useful for our own tool.


Last Updated on Wednesday, 05 August 2009 10:19