paul@ecrcvax.UUCP (Paul Martin) (01/21/88)
As there seems to be an increasing interest in the realm of ODA/ODIF protocols I thought a brief introduction to ODA/ODIF may be of interest to readers of this news group. I worked on the PODA ( Piloting Office Document Architecture ) ESPRIT project 1024 and was actively involved in its first public demonstration in Hannover '87 ODA (Office Document Architecture) views a document as consisting of 2 components - a logical structure and a layout structure. Logical components of a document include paragraphs, sections, footnotes etc. Layout components include pages, columns , and content areas of different types. ODIF( Office Document Interchange Format ) is the standard format for the interchange of documents between different manufacturers word processors. The aim of PODA is to demonstrate the use of ODA for the interchange of processable documents between current word processors ( currently ICL, Siemens , BULL and Olivetti ) "Processable" means that the document is transmitted by taking a logical rather than a physical view of how its contents are structured. The implications are : a) when the recipient displays or prints the document its layout is not necessarily identical to the original; it is consistent with certain layout directives but they may be interpreted differently on different equipment, for example the algorithm to determine position of line breaks may differ; b) the recipient is able to reformat or edit the document consistently with the originator's intentions; for example, when adding pages or moving text over page boundaries the correct page headers and footers are maintained, or when inserting new sections the automatic numbering scheme can be retained. In a typical scenerio, a document held in the internal representation of word processor A is converted (EXTERNALISATION) into an ODA data stream before transmission to word processor B. The document is there converted again (INTERNALISATION) into the required internal representation for word processor B. Word processor B can display or edit the transmitted document and send the modified or unmodified document on to another word processor - which may be the original. A more detailed description of the process of EXTERNALISATION (word processor to ODIF) is given below: a) Open a UNIX file to contain the document b) Create C language data structures to represent each of the required ODA constituents - achieved using ODA generator code c) Set values in data structure members corresponding to ODA attributes. d) Close the constituents e) Close the document file f) call ODIF functions to translate the ODA document to ODIF g) send ODIF document ( the transmission medium is unimportant - kermit over an RS232 line was used for the Hannover '87 demo ) the process of INTERNALISATION (ODIF to word processor) a) recieve ODIF document b) call ODIF functions to read ODIF representation and produce from it an ODA representation in store c) Use ODA analyser code to read ODA document and generate the corresponding proprietory word processor document. d) edit word processor document in accordance with originator's intentions This is only a very brief introduction to the purpose of ODA - a more detailed definition can be found from the ISO Draft Proposals 8613/1-6. It may also interest readers that a demonstration of ODA will be given during the week March 16-23 in Hall 16 at the Hannover Fair '88 I am sure participating parties will only be too pleased to offer further information. disclaimer: I do not pretend to be a specialist in the area of ODA/ODIF - the above simply represents my own knowledge of the subject and am passing it on in the hope it will be useful to someone. ------------------------------------------------------------------- PAUL MARTIN ECRC - European Computer-Industry Research Centre GMBH ARABELLA STRASSE 17 D-8000 MUNICH 81 EMAIL : paul%ecrcvax.UUCP@Germany.CSNET --------------------------------------------------------------------