From masonjd@y12.doe.gov Fri May 25 10:39:21 2007 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Date: Fri, 25 May 2007 13:37:16 -0400 From: "Mason, James David (MXM)" Subject: RE: Scope-Durusau-11 In-reply-to: <4654C023.3020706@durusau.net> To: patrick@durusau.net, Dave Welsh , Doug Mahugh , Frank Farance , Jerry Smith , Jon Bosak , "Levine, Leonard F CIV DISA GES-E" , Norm Walsh , "Lynne A. Price" , Rex Jaeschke , Richard Barber , Robert Weir , Steve Carton , Tim Schoechle , Kevin Trainor , Eduardo Gutentag X-MIMEOLE: Produced By Microsoft Exchange V6.5 Content-class: urn:content-classes:message Thread-topic: Scope-Durusau-11 Thread-index: AcedlsOo2ct7UuvGSKSaS5K6D+9DnQBVrqww X-PMX-Version: 5.2.0.264296 X-MS-Has-Attach: X-MS-TNEF-Correlator: X-OriginalArrivalTime: 25 May 2007 17:37:17.0949 (UTC) FILETIME=[5722D2D0:01C79EF3] This message should be read after my comment on NamingDIS29500-Durusau-12. The key part of Patrick's comment is this: It has taken a lot of reading of OpenXLM and working with the documents provided by Ecma to realize that there is no document model offered by DIS 29500 in the sense that the term is understood in the XML community. This, too, takes me back to my encounters with ISO/IEC 8613, ODA, in the time roughly 1981-1997. Like the present ECMA proposal, ODA, which also came largely from ECMA, operated primarily by capturing the appearance of documents. (I limit myself primarily to the WordprocessingML component of the proposal, since ODA never attempted spreadsheets.) ODA talked about a "layout structure" and a "logical structure", but neither of these corresponded clearly to a "document model . . . in the sense that the term is understood in the XML community." (I cannot give detailed citations from 8613 because I long ago gave my documents to an archive.) "Layout structure" dealt most clearly with the capture of formatting artifacts, such as font changes. That much of ODA was fairly clear. Layout might also encompass page areas, such as headers and footers, though some called those logical elements. "Logical structure" caused considerable confusion. Both the ODA and SGML camps initially confused it with the hierarchical structures that appear in document models like DocBook or TEI. It emerged only gradually that logical structures in ODA meant simply recognizable blocks that had identifiable, consistent properties. Typically "logical" components were at the paragraph level, though they might appear inside paragraphs, as in the case of highlighted phrases. In effect, the model was no more complex that of HTML 1.0, in that elements, such as they were, could appear in almost any order. I often compared to document model of ODA to that implicit in my ancient IBM Mag Card I typewriter (ca. 1975; it's in my dungeon): even that device recognized paragraphs. I've been playing with some OOXML instances recently, and I find the same sort of thing. There may be localized hierarchies (e.g., a fraction has a numerator and denominator; I've been looking at math), but overall the structure appears flat. ODA originally was conceived as having a binary interchange format (based, in theory, on ASN.1 but actually unique to itself) called ODIF (Office Document Interchange Format). With the merger of committees in both JTC1 and the national bodies, pressure arose to create a parallel SGML interchange representation of ODA documents in ODIF. The absence of a traditional "document model" in ODA led to the creation of an SGML tool that could be used to represent previously undefined, ad hoc structures: Architectural Forms. (The supposed advantage of the binary representation was compactness, but tests at NIST showed that the SGML was actually more efficient.) Coming back to Patrick's document: The lack of a traditional document model is probably a neutral condition. That the documents in question can be represented in a structure marked up in XML is simply a demonstration of the flexibility of XML, just as the SGML component of ODIF showed in an earlier generation of such things. I second Patrick's recommendation that the scope of the proposed standard be changed to clarify that it is not an attempt to promulgate a new XML document model in the tradition of DocBook and other industry standards. Jim Mason