TITLE: | Disposition of Comments to JTC 1/SC 34 N 801 - Text for CD ballot for ISO/IEC 19757-9: Document Schema Definition Language (DSDL) Part 9 - Namespace- and datatype-aware DTDs |
SOURCE: | Mr. Francis Cave |
PROJECT: | CD 19757-9: Information technology - Document Schema Definition Languages (DSDL) - Part 9: Datatype- and namespace-aware DTDs |
PROJECT EDITOR: | Mr. Francis Cave |
STATUS: | Disposition of comments |
ACTION: | For information |
DATE: | 2007-03-24 |
DISTRIBUTION: | SC34 and Liaisons |
REFER TO: | N0801 - 2006-11-22 - Text for CD ballot for ISO/IEC
19757-9: Document Schema Definition Language
(DSDL) Part 9 - Namespace- and datatype-aware DTDs N0819 - 2007-02-22 - Summary of Voting on JTC 1/SC 34 N 801 - Text for CD ballot for ISO/IEC 19757-9: Document Schema Definition Language (DSDL) Part 9 - Namespace- and datatype-aware DTDs |
REPLY TO: |
Dr. James David Mason (ISO/IEC JTC 1/SC 34 Secretariat - Standards Council of Canada) Crane Softwrights Ltd. Box 266, Kars, ON K0A-2E0 CANADA Telephone: +1 613 489-0999 Facsimile: +1 613 489-0995 Network: jtc1sc34@scc.ca http://www.jtc1sc34.org |
Source: Francis Cave, Project Editor
We believe that it is now too late to provide a DTD extension as an standard, especially because this is still a CD.
Response: Comment not accepted.
Many major document applications around the world continue to be dependent upon DTDs for a number of reasons:
There is a large established pool of expertise in the development and maintenance of DTDs. The cost of retraining to use other schema languages would be high compared with the anticipated cost of additional training to use tools that implement DSDL Part 9.
There is a large installed base of off-the-shelf and bespoke validation tools that rely upon DTDs for their operation. The cost of replacement of existing bespoke validation tools) would be high compared with the anticipated cost of acquiring tools that would perform the additional validation tasks enabled by implementations of DSDL Part 9.
The cost of conversion of DTDs to other schema languages would not always be low, since the task cannot be completely automated. DTDs would in many cases need to be retained for entity management and would therefore have to be maintained in parallel with schemas in other languages. In such a situation it would be a requirement that a DTD should not treat as invalid any instance that is treated as valid by a replacement schema, and it is unlikely that this can be guaranteed without manual checks. The costs of guaranteeing this level of equivalence between DTDs and schemas is anticipated to be high in many cases.
What happened to the accepted Japanese comment (see N0748).
Response: The comment is reasonable. The Japanese comment in the N0748 was accepted and it had been the editor’s intention to include text in 9.2. The editor apologies that this was not included and will rectify this in the next draft.
Supply a processing model as a reference model, and make sure this is implementable on top of XML conformant processors.
Response: Comment accepted. A processing model will be added to the text.