[rfc-i] Update on the plain text thread(s)

Heather Flanagan (RFC Series Editor) rse at rfc-editor.org
Thu Jun 26 11:21:42 PDT 2014

Hello all,

I’ve pulled several useful points out of the wealth of responses this
week, and thank the people who offered them.  For those of you just
coming into the RFC format conversation, I strongly suggest reviewing
the mailing list archives starting back in early 2012.  That’s when the
most recent round of conversation around the format for RFCs began;
there have been discussion about RFC format for over a decade, but the
last two years or so have pretty much rehashed most of the older
arguments, so you don’t need to go back to the very beginning.  If you
choose to do so, I suggest meeting with folks like John Klensin or Brian
Carpenter and offering them a drink or two at the bar.  In the meantime,
I will not respond (again) to the points or concerns that have already
been addressed unless new information is provided that was not already
in those previous discussions.

I will continue to announce the progress around this work at IETF
Plenaries, on this list, at IAB sessions, in the hallway, at other
conferences, and even in an upcoming article in the IETF Journal.  If
you have suggestions for other forums (or fora, if you prefer Latin),
please let me know.

First, regarding the draft itself.  I agree, better justification within
the draft is necessary.  I will work on that text.  And I also agree
that the non-ASCII draft (draft-flanagan-nonascii) should be normatively

Second, regarding line lengths.  I appreciate the thought that went into
the original line length choice.  They certainly make sense, and yet
people have strongly suggested that those limitations, particularly when
it comes to ASCII drawings, are problematic in terms of creating useful
diagrams.  I proposed 85 as a possible compromise.  I shall review that
recommendation, and include the justification in the draft.

Third, regarding artwork as normative, I have discussed the issue of
text-as-normative/artwork-as-informative with the IESG, and they had a
very valid concern that some things used as artwork do not transfer to a
text description particularly well.  The general consensus out of that
discussion was that text-as-normative should be recommended but not

Fourth, regarding pagination.  The discussions on pagination, pros and
cons, have been hashed out at great length, over several years.  I have
not seen any new information in this most recent debate on the
merits--or lack thereof--of pagination.  Note that there is nothing
prohibiting the community from creating paginated plain text if
individuals find that particularly useful, and the design team has
talked about the possibility of making such a tool available on the
tools page.  The current tools create paginated text from unpaginated,
not the other way around.  Also, there will be a PDF format, and that
cannot help but be paginated.  Yes, having some things paginated and
others not may be problematic for people who cannot be certain what
other format a reader may be reviewing; that’s one of the reasons the
RFC Editor has recommended for years to point to section numbers instead
of page numbers.

Fifth, paragraph length and markup.  The concern about pagination
quickly leads into a concern about paragraph length and possible markup
for each paragraph.  During one of the earlier rev’s of the revised
style guide (which is in RFC-EDITOR state, for those of you tracking
that particular effort), we tried to find language stating that the RFC
Editor recommends limiting paragraph and even section lengths but found
that we had difficulty getting consensus on the language.  I think the
challenge of very long paragraphs is a very valid concern, and hopefully
enough others will agree during the authoring and review processes that
this will be dealt with before it gets to the RFC Editor.  I also agree
that providing something to number each paragraph may be very helpful. 
It is easy enough to do and represent in the XML, HTML, and PDF, but I’m
not clear on how this will look in a text file. 

Action items for me: revise the plain text draft, update the RFC Format
FAQ, and discuss with the design team how to implement paragraph numbers.

Action items for you: review the other drafts associated with this
effort, search the archives for related discussions (in case we have
already had them), and watch my five minutes of fame at Plenary for a
short, short update on the effort so far.

Thanks for your enthusiasm; there are few things more frustrating than


More information about the rfc-interest mailing list