[Wg-uma] RFC: Specification authoring process
eve at xmlgrrl.com
Mon Aug 24 15:43:53 PDT 2009
This approach seems sound to me. It meets my goals of:
- Authoring in a format that multiple people can contribute to
- Targeting IETF in deed as well as in word
- Producing native HTML in an attractive format that will hopefully
appeal to developers
- At least middling levels of content management, such that diff
redlines can be viewed
I think we're still fine with native Confluence as the source form for
auxiliary non-normative documents. (Note that xml2rfc wouldn't be
suitable for any document that relies on graphics, as it doesn't allow
non-ASCII graphics per IETF.)
Thanks for looking into it, Paul!
On 24 Aug 2009, at 3:16 PM, Paul C. Bryan wrote:
> During our discussion in last week's conference call, we began
> addressing how the specification should be authored, where content is
> hosted, etc.
> I was leaning towards using a tool like xml2rfc , as it provides
> IETF-compliant output in a number of useful formats.
> Before proposing an approach, I first wanted to determine how Kantata
> would be hosting files. After contacting staff, I have learned that
> allow revision control of files via Subversion and allow publication
> raw HTML on the wiki.
> This seems to be a good fit for me to propose the following approach:
> 1. The protocol specification will be authored in the xml2rfc schema.
> 2. The master source file(s) for the protocol specification will be
> in Subversion.
> 3. When changes are made, xml2rfc will be run to generate HTML
> 4. HTML content will be published through the wiki as raw HTML.
> 5. Formal versions of the spec will be tagged in the Subversion
> 6. HTML output of the formal versions be retained separately in the
> Please let me know if you have any questions or objections to this
> proposed approach.
>  http://xml.resource.org/
> Wg-uma mailing list
> Wg-uma at kantarainitiative.org
eve at xmlgrrl.com
More information about the Wg-uma