The Draft 0.50 schemas specify elementFormDefault="qualified" which results in prefixing of XML elements with a namespace identifier.

Thus our messages would have a format similar to <cmis:ElementName> instead of simply <ElementName>.

Some tools do not allow the user to specify "pretty" prefixes like "cmis", so the actual qualified XML might look like <bons7:ElementName> where bons7 (business-object-namespace-7) was earlier mapped to our CMIS domain namespace via a xmlns:bons7="..." attribute.

This adds complexity and bulk to authoring and reading XML messages and I am unaware of any substantive benefit.

Major vendor XML tooling create schemas that leave the elements unqualified without observable negative impact.

A "Best Practices" site for XML (http://www.xfront.com/HideVersusExpose.html) recommends leaving elements unqualified "when simplicity, readability, and understandability of instance documents is of utmost importance".

Unless there is a compelling argument for using "qualified" elements, I think the group should should similarly favor "simplicity, readability, and understandability" for our XML messages and allow this schema specifier to default to "unqualified".

Could one of the participants from the original draft group provide some insight on the basis for this earlier decision?