[xml2rfc] References feature

[fenner at research.att.com (Bill Fenner)]

While I don't think this belongs in the DTD, other than for Julian's
suggestion of consistency checking, I have envisioned implementing
a references mode for xxe - showing all the references you've made,
the ones that might be dangling [offering to insert <?rfc include=?>
for them], and making it easy to move between references sections.

(This would require a fair amount more java programming
than I've done in a while, which is why it's not there yet)

  Bill
>From Dale.Worley at comcast.net  Tue Jul  4 15:55:39 2006
From: Dale.Worley at comcast.net ([e-mail])
Date: Tue Jul  4 11:55:46 2006
Subject: [xml2rfc] Problem with <texttable>
In-Reply-To: <[e-mail]>
	([e-mail])
References: <[e-mail]>
	<[e-mail]>
	<[e-mail]>
	<[e-mail]>
	<[e-mail]>
	<[e-mail]>
Message-ID: <[e-mail]>

   From: Marshall Rose <[e-mail]>

   i'm not sure i follow.

   you sent a file that showed that xml2rfc had a bug.

   i suggested you add the width attribute above as a temporary hack to  
   get around the bug.

   did the fix work for you?

Yes, it works in the document I originally had the problem with.

Dale
>From dbharrington at comcast.net  Tue Jul  4 18:11:44 2006
From: dbharrington at comcast.net (David B Harrington)
Date: Tue Jul  4 14:13:04 2006
Subject: [xml2rfc] References feature
Message-ID: <0d8201c69fae$74646420$[e-mail]>

Hi,

I am not saying that Marshall is incorrect. In fact, I think he is
correct:
> "Z is normative" has meaning only in the context of the referring
document.

However, I apparently did not describe my suggestion adequately.
My suggestion would only affect the referring document.

The attribute I suggested is part of the xref tag.

<xref  target=Y usage=normative> //the usage at this point is a
normative reference; 
                                //to implement this part of the spec,
you must understand Y.
<xref target=Z usage=informative>
<xref target=Y usage=informative> //the usage at this point is an
informative reference; 
                                //Y may be helpful, but is not
required to implement this spec.


<!-- these are sorted into normative or informative by xml2rfc based
on the usage attributes in the xref tags; if any xref for a document
has usage=normative, then the reference is put into the normative
section. If all xrefs to a document have usage=informative then the
reference is put into the informative references section.-->
<references> 
<normative references/>
Y
<informative references/>
Z
</references>

Dbh

p.s. My first post to an IETF mailing list was to argue a point with
Marshall. 
I learned to never say that Marshall is incorrect ;-)

p.p.s. this feature might be better implemented in an editing program
rather than xml2rfc, especially if the program, like XXE, manages the
references.

> -----Original Message-----
> From: Dave Cridland [mailto:[e-mail]] 
> Sent: Friday, June 30, 2006 2:58 PM
> To: Marshall Rose
> Cc: Mailing list for software packages implementing rfc2629; 
> David B Harrington
> Subject: Re: [xml2rfc] References feature
> 
> On Fri Jun 30 19:33:27 2006, Marshall Rose wrote:
> > in other words, my belief (which may be mistaken) is that the 
> > statement
> > 
> > 	Z is normative
> > 
> > has meaning only in the context of the referring document.
> 
> And David is saying that you're incorrect, because the statement "Z 
> is normative" has meaning only in the context of the xref. By 
> collecting references into Normative and Informative piles, we force

> the scope into being document wide, but this can be done quite
easily 
> - if there are any normative xref to Z in the document, then the 
> reference itself is normative.
> 
> In other words, xml2rfc (or whatever) could handle the division for 
> us, given more metadata at the xref itself.
> 
> Dave.
> -- 
> Dave Cridland - mailto:[e-mail] - xmpp:[e-mail]
>   - acap://acap.dave.cridland.net/byowner/user/dwd/bookmarks/
>   - http://dave.cridland.net/
> Infotrope Polymer - ACAP, IMAP, ESMTP, and Lemonade
>