- From: John Boyer <boyerj@ca.ibm.com>
- Date: Tue, 23 May 2006 16:03:32 -0700
- To: Danno.Ferrin@intelliden.com
- Cc: www-forms-editor@w3.org
- Message-ID: <OFF9621781.36B3A777-ON88257177.007BBB8C-88257177.007EAB9B@ca.ibm.com>
Hi Danno, My apologies that we have not responded sooner to your email. The XForms working group has discussed the issues and asked me to respond based on the discussion. <danno> While reading the section on inserting into empty collections it seemed to me a bit Kludge to have to references a (likely) separate model element as the origin xpath. One idea that crossed my mind was to allow the insert element to contain any content (xsd:any) that will be inserted. This content could not exist if there is any Node Set Binding attributes or origin attribute, or it would need to have precedence explicitly set to above or below the node set binding. This would help to localize the data within the form nearer to it's use. </danno> We agree on the necessity of tightening what we say about Evaluation Context for edge cases such as when a model change occurs. However, the context model is selected by the model attribute and applies to the initial selection of the context node to be used in evaluating the insert context, not just the nodeset binding. Moreover, the insert context is used to initiate the evaluation of the origin. So, we believe the insert action operates over the nodes of a single model, which addresses your concern over having two models being used within an insert action. With regard to the suggestion that we instead allow the insert element to contain xsd:any content to be inserted, it should be noted that it is a good suggestion, though we cannot adopt all good suggestions. In this case, we have specifically avoided this approach as it causes XML data to infest the entire document rather than staying centralized within XML instances. Also, by keeping XML data strictly within the instances of models, the template for new insertion data is closer to the point where it will be inserted. In other words, the template data is used within instance data, not at the point where an insert action appears. We hope this explanation satisfactorily explains why the working group has elected not to pursue further the idea of placing the insertion template into the insert element as content. Best regards, John M. Boyer, Ph.D. Senior Product Architect/Research Scientist Co-Chair, W3C XForms Working Group Workplace, Portal and Collaboration Software IBM Victoria Software Lab E-Mail: boyerj@ca.ibm.com http://www.ibm.com/software/ Blog: http://www.ibm.com/developerworks/blogs/page/JohnBoyer
Received on Tuesday, 23 May 2006 23:03:46 UTC