[rfc-i] Number of submission formats

[ynir at checkpoint.com (Yoav Nir)]

On Jan 18, 2013, at 11:30 PM, Nico Williams <nico at cryptonector.com> wrote:

> On Fri, Jan 18, 2013 at 12:45 PM, John Levine <johnl at taugh.com> wrote:
>> 
>> Heck, no.  There is a canonical output format, the one to which we
>> point if someone wants to know what is "the RFC."
> 
> You misunderstood.
> 
> Today's I-D upload tool has several input formats allowed, one is
> required.  The .xml input is only really useful when a) you want
> others to be able to see it (and contribute XML changes) or b) the I-D
> is about to get sent to the RFC-Editor queue, in which case the editor
> can just... find it there, saving a step in the process.

or c) you want the reflow-able html to be generated, like so:
http://tools.ietf.org/id/draft-williams-websec-session-continue-prob-00.html

> But the RFC-Editor will also take nroff, so being able to submit
> either or both of those is useful.
> 
>> There are acceptable input formats.  Currently there are, as I
>> understand it, 2 1/2 input formats, line printer format similar to the
>> current output format, xml2rfc, and maybe nroff if it uses the same
>> coding the production people use.
> 
> I can't imagine the RFC-Editor being happy to work with the formatted
> .txt only as an input.  Say I edited an I-D that way, formatting and
> paginating by hand (don't laugh, I used to, though I had a script to
> do the pagination)?

They'll take just the txt version, although they'll grumble. RFC 4478 had no nroff or xml source.

>> I don't see any reason to change that structure.  [...]
> 
> Paul is the one that proposed a change.  I was trying to get that
> change narrowed and clarified.

Currently, Internet Drafts look very similar to RFCs. I think we'll want to keep that similarity.

Yoav