This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
We have published an extension architecture which is primarily defined through an XML schema. We would like the ability to deprecate out-dated elements as we move our architecture forward, so our XML validators would have the chance to flag them.
Sorry for the lack of response to the issue you raise. The following comments reflect my individual views; I am not speaking for the WG or any organization. At the moment, you can add <xsd:annotation> <xsd:appinfo> <my:deprecate date="2006-11-02" level="2"/> </xsd:appinfo> </xsd:annotation> to the declaration of each element or type you wish to deprecate. Or more simply, you can add attributes to the declarations: <xsd:element name="unit" my:status="deprecated"> ... </xsd:element> Any processor which provides access to the full PSVI will point from the element to the governing element declaration and governing type definition, and you can check the appinfo to see if the element is deprecated or allowed. (Alternatively, it would not be terribly difficult to write an XSLT stylesheet to read schema documents in which deprecations are marked in this way, and generate an XSLT stylesheet which reads instances and generates messages when it sees deprecated elements or types. That way you don't need access to the full PSVI.) But I assume you already knew that, and that you don't want to do that, but have the deprecation somehow understood and processed by the schema-based validator. What would you like the schema processor to do with deprecated elements? Accept them, since they are after all valid? Reject them, since they are after all deprecated? Support a run-time switch to reject, or to raise a warning when it sees, deprecated elements or types in use? Signaling that something has been deprecated is clearly an important part of managing a vocabulary. But it's not immediately clear, at least to me, what effect such deprecation might or should have on schema-validity assessment. Your thoughts on this topic would be welcome. I don't want to promise much -- XML Schema 1.1 is long overdue and the Working Group may well wish not to undertake any new work. But if you have specific proposals, I'd like to hear them.
On 2007-03-29 the WG discussed this issue at its f2f meeting. The WG determined that MSM gave a good answer to the issue in Comment #1. Accordingly we have classified it as WORKSFORME.
Comment #2 should have added explicitly that the Working Group requests that Derek Johnson, as the originator of the issue, indicate whether you are satisfied with the WG's response and its rationale, either by changing the status of the bug to CLOSED, to indicate that you are content with the disposition of the question (or at least resigned to it), or else by changing it to REOPENED to indicate that you're not satisified and wish the WG to reconsider (or failing that, that you wish to lodge a formal objection to the WG's decision and appeal it to the Director at the next document review). In the latter case, it would be helpful if you respond to the ideas in comment #1. If we don't hear from you in a couple of weeks, we'll assume that silence implies consent, and close the issue on your behalf. Thank you!