User Tools

Site Tools


design:20130814-notes

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

design:20130814-notes [2013/12/19 16:44] (current)
rsewikiadmin created
Line 1: Line 1:
 +1. Administrivia
  
 +
 +   * SVN access ok?
 +
 +   * RSE wiki access ok?
 +   * call schedule acceptable? (that means 6 calls between now and Vancouver)
 +
 +2. Expected outputs of group
 +
 +
 +   * examples of Publication outputs (they may be done manually and are designed to show what things will look like)
 +   * a specification or set of specifications on what needs to be done to create the necessary tools for XML to be the canonical format that in turn creates four publication formats
 +   * documentation on what areas we have discussed and resolved internally so when I take this back out to the community, I can point to something that says "we discussed that issue, not discussing it again unless you have some new information that is relevant"
 +   * clear guidelines on how different scripts (see RFC 6365 for terminology clarification) will be allowed in specific sections of an RFC now that the UTF-8 encoding will be allowed in the Author's Address section (I expect this to have an impact on the tools, but I don't know what that impact will look like)
 +
 +
 +3. Discussion topics
 +
 +
 +   * Semantic HTML markup (Joe)
 +   * SVG negative profile
 +
 +
 +
 +Open questions so far:
 +* How to deal with unintentional variations in output?
 +* How does xml2rfc fit in to this work?  What needs to change?
 +* Given the end goals, what tools are entirely missing? (i.e., is what we have now sufficient for diffs?)
 +* If errata are submitted on an error of some kind in a Publication format, should the document be updated? Who would judge whether the bug was in the Canonical format rather than the Publication format?
 +* will we have a "preferred" publication format?
 +
 +-----
 +Notes
 +Administrivia
 +SVN: All could check out; several haven't been asked for credentials to test check in; folks should test and if not working let 
 +RSE wiki access: Ted and Alice to sort stuff out
 +
 +PDF format - 8x11 as well as A-4
 +Get a copy of PrinceXML and create a PDF from that; that will include headers and footers
 +See about a license to run this on a server so the community could use it as well; this would be part of the tools we provide to the community; similar to submitting XML to resource.org; if we get a server site license, we will need to have some kind of archive of the code
 +
 +HTML format - Joe has been in contact with someone at the W3C that might have useful guidance on this
 +
 +EPUB - Ted would be interested in helping create what this could look like for us in the future; current xml.resource.org will create EPUB output though there are known problems with it (according to website); Julian and John Levine would be good resource here
 +
 +Specification or set of specifications
 +* xml2rfcv2 was created as a better software base on which future changes could be built
 +* there are several points of strong divergence from the xml2rfc architecture; we could refuse the back end
 +* changes to the DTD (defines to compliance to XML; and if we add to the DTD that would impact compliance checking among other things) if we want to enrich the xml2rfc language with specific additions, is that part of the v3 work? what new things do we actually want?
 +* if the canonical format is actually XML, then the whole concept of xml2rfc flies out the window; what would be needed in tools would be different; today it takes xml in, validates it, then spits it out in different formats (paginated and unpaginated txt and nroff and html today); if the first step is put out HTML (semantically complete and convertible back and forth from HTML); the DTD differences wouldn't use
 +** benefit of going from XML -> HTML -> others; more tools exist to convert from HTML than from XML and would essentially re-render rather than have all the formats figure out proper breaks, etc; reuse of existing tools rather than write new tools for each format
 +the new tools will be more multi step than the current tool
 +will we require data in the XML for pagination, or can we defer that to the HTML?
 +both aspects of pagination tends to be a trickier programming problem than most people seem to expect; if there are existing tools that can handle this in a nice way, this will be helpful right up until we MUST submit a manual break 
 +
 +
 +For what specs could look like, go see xml.resource.org, RFC 2629 as a sample
 +The RFP for v2
 +A doc that explains how it works and test cases (if possible)
 +
 +We need to maintain a list of test cases and examples to validate the code after the fact
 +
 +
 +Internationalization (certain scripts) would be a great thing for test cases, making sure a verifying catchings inappropriate use or location 
 +
 +note that the format of xml2rfc is not directly the question; Paul (channeling Julian) is thinking more about a revised DTD heavily based on what we have and see what we can do from there - start with current, a relaxNG schema of that, and examine deltas from there; better to start with the one in common use
design/20130814-notes.txt ยท Last modified: 2013/12/19 16:44 by rsewikiadmin