Re: [xml2rfc] xml2rfc online: references to I-Ds and associated dates

[John C Klensin <[e-mail]>]

--On Friday, July 12, 2019 10:09 +1200 Brian E Carpenter
<[e-mail]> wrote:

> On 12-Jul-19 09:51, John C Klensin wrote:
>> Hi.
>> 
>> I happened to look today at the reference produced for an I-D
>> by xml2rfc (at least by the online version at
>> https://xml2rfc.tools.ietf.org/ )
>> 
>> The XML for the relevant cited I-D says (irrelevant material
>> elided and the identity of the I-D changed to avoid
>> distractions):
>> 
>>   <reference <reference anchor="ID.draft-ietf-foo-bar"
>> 	 target="https://datatracker.ietf.org/doc/draft-ietf-foo-bar
>> 	 /"> ...
>>     <date month="June" day="12" year="2019" />
>>     </front>
>>   </reference>
>> 
>> The output is "June 2019".
>> 
>> Now, independent of what the RFC Editor decides to do with the
>> reference entry when/if the document gets to them [1], while
>> the document in which this is embedded is still an I-D, it
>> can be very important to the reader to know which version of
>> the I-D was cited and, of course, two or more versions of an
>> I-D in the same month is not uncommon.
> 
> Two versions on the same day is not impossible. For an actual
> case, I once did this:
> 
> <!-- ********** This reference is specifically to version 07
> of the draft and is therefore      ********** hand-crafted
> instead of using the I-D bibliography. -->
> 
> <reference anchor='RPL-07'>
> <front>
> <title>RPL: IPv6 Routing Protocol for Low power and Lossy
> Networks</title> <author initials="T." surname="Winter"
> fullname="T. Winter"/> <author initials="P." surname="Thubert"
> fullname="P. Thubert"/>  <date month='March' year='2010'/>
> </front>
> <seriesInfo name="Internet-Draft"
> value="draft-ietf-roll-rpl-07"/>  </reference> 
> 
> You can see the eventual result, which was not 100%
> satisfactory, in https://tools.ietf.org/html/rfc6294

Well, I just looked at the RFC and someone, probably the RFC
Editor, discarded your seriesInfo material entirely, leaving
just the authors, title, month and year, and the notorious "work
in progress".  If my guess about the relationship between [RPL]
and [RPL-7] is correct, the latter is certainly not a work in
progress either.  And, if those are the two versions in
question, they are in the same month but a year apart.
However, that is all about the RFC and other rules, including
the insertion of "work in progress" apply.

For the I-D case, there is no question that the information can
be forced in.  One can use your strategy, one can use something
a little more like I wrote, with the link in the target
attribute of <reference>, and then use 
  <seriesInfo name="Internet-Draft" version="07"/>
or one can use an annotation element and describe what is going
on.  Probably there are other options too.

But, as a long-term fan of the principle of generic markup, I
wasn't looking for a workaround, I was looking for the
information that appears in the markup to be reflected in the
output unless there is a substantive reason to not do so.  

At least some years ago, if a complete date appeared in markup,
a complete date also appeared in the output.  Either its
disappearance is an unintended side-effect of some other change
or someone decided to take it out.   Either way, I suggest the
result is not what we want and that it would make sense to
restore the prior behavior.

best,
   john