- From: James Fuller <james.fuller.2007@gmail.com>
- Date: Fri, 8 Jun 2007 00:07:42 +0200
- To: public-xml-processing-model-comments@w3.org
Hello, The current draft of the spec alludes to both a standard step library and the possibility of supporting a additional types. 'This specification defines a standard library, Appendix A, Standard Step Library, of steps. Pipeline implementations may support additional types of steps as well.' This wording makes any statement of conformance with respect to optional steps a little unclear. In Appendix A, under A2, there is then a 'laundry' list of possible optional steps. I propose to make it easier to classify a specific XProc implementation by doing the following; * delineate between Standard Step library and all other optional steps by moving A2 to its own appendix * promote pipeline reuse and extension mechanisms by segregating current and future optional steps into separate pipeline-libraries...perhaps called system, test, optional, etc... with these changes it would be possible to fully clarify an Xproc implementation conformance with respect to optional libraries....allowing implementators to pick and choose which optional pipeline libraries they want to implement. In addition, this provides a bit of wiggle room to absorb into the XProc draft, any 'out in the wild' libraries that become wildly useful for XProc without upsetting the core spec (or standard library). cheers, Jim Fuller
Received on Thursday, 7 June 2007 22:07:50 UTC