[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Addition ideas for EDXL Distribution
EM TC Members
Further comments I pondered while creating the current EDXL schema:
1. First I would like to thank Art for catching the misspelling of the elements in the proposed EDXL schema.
2. I would prefer to use strongly typed values rather than general string values like URN. In the example I generated, I used simple letters (e.g. a, b, c, d …) in the “ValueListUrn” elements.
3. Also, our data dictionary for “messageReference” is using the CAP 1.0 versus CAP 1.1 format. I would like to ensure the pattern for the “messageReference” format is “messageID,senderID,dateTimeSent”. This would eliminate some parsing errors for references later in subsequent EDXL message processing and/or Updates or ACKs.
4. Since there is a one-to-one correspondence with Message Element and Content Object, It would be a lot easier to “specify” the content namespace in the “messageElement” element definition to ensure consistent namespace for all subsequent child Message Elements.
<element name="messageElement" minOccurs="0" maxOccurs="unbounded">
<anyAttribute namespace="##any" processContents="skip" />
This would allow uniform namespace for both the “keyXmlContent” and “contentObject” complex types. Currently our schema would allow the definition of elements in “keyXmlContent” from one namespace and elements from “contentObject” from another namespace.
Finally, I believe conducting a distribution “use case” for how various distribution elements are to be used by Directed Distribution, Directory Lookup, “Publish/Subscribe” and “Workflow Control Routing” is critical to finishing this EDXL distribution definition.
I believe these are critical tests for all elements and decisions while we develop the EDXL distribution schema.