(By Notis Toufexis, response at the OSCE workshop, King's College London, September 2006)
It is easy to agree with Neel Smith's proposed method of moving from a conceptual model to the analysis of its distinctive features, and to then translate the resulting functional requirements into technical ones. Equally agreable is the simplicity of his proposal: reliance on well-established technical protocols - HTTP as the transport mechanism, XML for service requests and replies - in order to provide basic citational functionality - on different levels of granularity - and indexing services. The crux of his presentation, it seems to me, is, at least for our purposes, his advocacy of an agreement on "the meaning of standard values", i.e. an ontology to serve as the shared matrix of our disciplinary discourse venturing into the digital era.
What I would like to do in my brief response is to briefly raise two of the more complex issues relating to his proposal by simply asking two sets of questions:
Neel Smith recognises the necessity of publications - particularly scholarly publications - to adhere to both "permanence and citability" - the latter requiring the former. Yet, is it not one of the advantages of digital publications to they can be revised easily - corrected, even developed - without the wait for the edition to be out of print? And is this feature of impermanence not even more desirable in a collaborative framework, making the resource richer as it is moulded by subsequent generations of collaborators? How, then, do we refer to a work-in-progress in an authoritative way, how do we point at a moving object?
In addition to the complex web of entities created by the publication of manuscripts and printed works, what ~~new entities~~ is the digital medium supplementing and how should we deal with them?
Beyond a mere transcription of a 'text', we can now have texts intimately interwoven with layers of markup - some encoding the same features, but differing, others adding further features, and still others combining any number of them. Should we distinguish these from the base text and/or from each other?
We can have furthermore, a plurality of 'editions' coming out of the same repository, catering for different interpretive needs. Again: should we - and if so: how? - distinguish archival and presentational layers?
Concerning his concrete proposal a further related question must be asked: does a hierarchical model cater for all these - old and new - categories?