What steps will reproduce the problem?
1. Use interpreted text with the default role (title-reference)
2. run rst2rfc
3. run xml2rfc

What is the expected output? What do you see instead?
rst2rfc produces a <cite> element, which will be stripped (with its content!) 
by xml2rfc. At least 
the text should be preserved, and RFC2629/xml2rfc doesn't provide any 
equivalent, so I propose just 
outputting the text as is, without markup.

What version of the product are you using? On what operating system?
Single version to date (Dec 10, 2009)

Please provide any additional information below.
See 
http://code.google.com/r/tbroyer-rst2rfc/source/detail?r=b185859cc564c8c36809b99
ef7b2ca7fd458e13b

...         
789 789         self.endtag()
790 790     
791 791       def visit_title_reference(self, node):
792     -       self.body.append(self.starttag(node, 'cite', ''))
    792 +       pass
793 793     
794 794       def depart_title_reference(self, node):
795     -       self.body.append('</cite>')
    795 +       pass
796 796     
797 797       def visit_topic(self, node):
798 798         if 'abstract' in node['classes']:
...         

What steps will reproduce the problem?
Include a citation such as [W3C.REC-html401-19991224]_

What is the expected output? What do you see instead?
I'd expect the refs/W3C.REC-html401-19991224.xml file to be "integrated" into 
the resulting XML, just like with RFC citations (such as [RFC2616]_); 
instead, the citation isn't processed in any special way.

What version of the product are you using? On what operating system?
Single version to date (Dec 10, 2009)

RFC2629 allows notes after the abstract in the front matter. This is generally 
used in I-Ds to add an 
"Editorial Note (To be removed by RFC Editor before publication)" note linking 
to the mailing list, 
issues list, etc.

See 
http://code.google.com/r/tbroyer-rst2rfc/source/detail?r=9eb859a867dcb841ae7b331
7846113c376fd239d

What steps will reproduce the problem?
1. process an rst file with rst2rfc
2. send the resulting XML file to the online service at 
http://xml.resource.org

What is the expected output? What do you see instead?
xml.resource.org complains about the lack of ipr= attribute

What version of the product are you using? On what operating system?
Single version to date (Dec 10, 2009)

Please provide any additional information below.
This could be done using attributes such as:
:ipr:full3978

(and similar for docName)

What steps will reproduce the problem?
Use a "emphasis", "strong emphasis" or "inline literal".

What is the expected output? What do you see instead?
While "emphasis" and "inline literal" generate nothing special (text is 
preserved but is not marked 
up specifically), strong emphasis will produce a <strong> element which will be 
stripped (with its 
content!) by xml2rfc.
At least the text that is put in strong emphasis in reST should be 
preserved,but xml2rfc (RFC2629-
bis) provides appropriate markup that'll be converted into _emphasis_, *strong* 
and "literal" 
respectively by xml2rfc.
See 
http://xml.resource.org/authoring/draft-mrose-writing-rfcs.html#rfc.section.2.3.
1.9

What version of the product are you using? On what operating system?
Single version to date (Dec 10, 2009)

Please provide any additional information below.
See 
http://code.google.com/r/tbroyer-rst2rfc/source/detail?r=5bf0b0d4edc411db90ebb8f
6813afec142f464cf

Currently (single version to date: Dec 10, 2009), all citations/references 
go into a single, untitled <references> section in the output XML. It'd be 
good if citations/references could be somehow "categorized" as either 
"normative" or "informative", generating the appropriate sections:
   <references title="Normative References">...</references>
   <references title="Informative References">...</references>

Maybe this could be done by introducing new "interpreted text roles" 
instead of using the existing :citation: one, such as:
   :norm-ref:`RFC2616`
   :info-ref:`W3C.REC-html401-19991224`
It'd have the side-effect advantage that no corresponding "dummy" citation 
be present at the end of the document (the ".. [RFC2616]" thingy), with the 
drawback that other, "standards", docutils tools (such as rfc2html.py) 
could no longer be used.