At 3:03 PM +0900 7/28/04, Martin Duerst wrote:
At 13:05 04/07/27 -0700, RJ Auburn wrote:
On 07/21/2004 22:03, "Larry Masinter" <LMM@acm.org> wrote:
These comments are as much about the general "IETF MIME type registration from W3C recommendation" as they are about this particular registration:
Martin: Would you be the person to handle/address the general issues?
Yes. For everybody's information, RJ is following the procedure laid out Because he is the first to do so, this is a very good case to see where we have to tweak that description. I have already made two additions: 1) Added a sentence "Make sure that this part of the specification is readable on its own, without the context of the specification." [for further details, a good example is probably better than a lot of explanations] 2) Added a sentence "To make it easier for your WG to track comments on the Media Type section, you may cross-post the comments list for your specification." [I want to leave this to the group for the moment. They have to show that they addressed comments to the IESG, so having that documented in a last call table may have advantages and disadvantages.]
Also, I'm planning to add some pointers to examples to the above description, once we have them. That should make it easier for others to do this.
Your translation from HTML to ASCII left out line breaks before heading lines, which made your template hard to read.
If needed I can resubmit a nicer looking version. Let me know...
I guess that can wait for the next time you send something anyway, but I hope this will be soon.
Published specification:
This media type registration is for CCXML documents as described by this specification.
I'm not 100% sure if this is necessary, but I'd expect if the template were to appear elsewhere to see a bibliographic citation, e.g.,
"Voice Browser Call Control: CCXML Version 1.0", W3C Working Draft, 30 April 2004, W3C,
Is "this specification" (or the whole specification) precise enough? In some other cases, a single W3C recommendation defines many different data types. Perhaps it would be useful to say, somewhere, e.g., that the MIME type refers to XML bodies that conform to the DTD/Schema referenced in Appendix B and C and interpreted by the rules in the cited specification.
Pointing at the schema/dtd sections seems reasonable. How is this for text:
Published specification: This media type registration is for XML bodies that conform to the DTD/Schema referenced in Appendix B and C and interpreted by the rules this specification
'this specification' -> 'of this specification'
Person & email address to contact for further information:
RJ Auburn, <rj@voxeo.com>.
Should there be a W3C contact as well?
Dave/Max/Martin: Thoughts?
Adding the name of a staff contact or so might be a good idea.
Intended usage:
COMMON Author/Change controller:
The CCXML specification is a work product of the World Wide Web Consortium's Voice Browser Working Group. The W3C has change control over these specifications.
Or perhaps the W3C contact address should be listed here.
Dave/Max/Martin: Thoughts?
The W3C is 'on the Web', not at a particular physical location. This kind of wording has been used in some previous registrations, and should be okay.
Actually, Martin, the W3C should come up with something better. This is a detail of the "life after Rec[ommendation status is achieved]" that I am not aware we have laid out adequately. There is a slight problem because the W3C has studiously avoided being a legal entity which could be identified in standard X.500 terms. Do you have a technical comment? W3C technical reports include email addresses where readers should send technical comments. </quote> The Consortium has also assiduously avoided promising to answer questions about its utterances, other than that one may post 'comments' in the above manner. Clearly a 'comment' alleging that there is an inconsistency in the provisions of a specification, if read and found to be accurate, may result in something being added to the errata for a document. I would suggest that the 'comments to' email address be included in the contact information provided in a MIME type registration request. As far as identifying The Consortium it should suffice to use the URI as identification in this context. It would be helpful, but probably not required by the IETF pro-formas, to also allude to the process for change control by a reference to the W3C Process Document identified Al
Regards, Martin.