I would like to comment/clarify on David Stone's ODA remark.
(These comments are based on my own understanding of ODA--I do
not claim to be an authority on the subject.)
>From: DSTONE%[hidden email] ...
>Contrast this with the ODA-ODIF standard for interchanging office
>documents, promoted by the CCITT, ISO, ECMA and several major computer
>vendors. It uses ASN.1 to structure the document - and editors that
>understand ASN.1-ODIF are few and very expensive. ASN.1-structured data
>are almost impossible just to read, let alone edit, without special
>tools. (ASN.1 and ODIF have their advantages, too: but that's another
First, the purpose of ODA definition is, as David says, the
*interchange* of documents, between different systems, not the
*long* term storage format, which is, as I believe, the
goal of TEI. [At this point SGML fills this goal better than
Second, ODA defines the document or information in abstract
form, regardless of the actual presentation in a particular
computer. The ASN.1 is *not* used to structure the document,
the ASN.1 is used to define a *transient* interchange format for
the abstract form, so that this form can be transferred between
Editors are not required to understand ODIF normally, they are
supposed to read whatever local presentation they are used to, be
it eiher SGML based or something else. Editor may support "import"
function that is capable of reading ODIF and transforming the
document to the local format.
Totally different thing is, that the abstract structure
defined by ODA might be a good guideline for very sophisticated
document processors. But, I guess we are getting here into that
Markku Savela | [hidden email] Technical Research Centre of Finland |
Telecommunications Laboratory | [hidden email] Otakaari 7 B, SF-02150 ESPOO, Finland | savela%[hidden email]