Re: [rfc-i] [Rsoc] RFCs with page numbers (pretty please) ?

[Tony Finch <[e-mail]>]

John R. Levine <[e-mail]> wrote:
>
> My point was that if you want a mechanical reference to a section
> in an RFC, the RFCs we publish now have them.

I think that when you have neither page numbers nor paragraph numbers then
document authors ought to be much more keen to have lots of numbered
subheadings to make it easier for people to refer to a particular point.
(Actually I think more subheadings are good regardless of numbering!)

Numbered paragraphs would make references even easier. The C standard has
paragraph numbers but they are not written in a way that suggests exactly
how a reader should refer to them
<http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1570.pdf>. There have
been a few RFCs with paragraph numbers, e.g. RFC 2671, RFC 2845 (I don't
know if this is a Vixie-ism or if other authors did it too) and they
helpfully include the full dotted section number so they make it clear how
to refer to them, and they give you more context if you are lost in a wall
of text.

I had a look for recent RFCs with problematic lack of numbering and the
first I found was RFC 8923. There's a figure in section 6.1 without a
number. There's a huge list of protocol features in appendix A which a
reader can only refer to by name (if they don't want to depend on a
particular web site), and which can only be located by searching (awkward
if you are reading on paper). These problems can be avoided within the
current technical style but it needs a change in editorial style to ensure
the table of contents has a high enough density.

Tony.
-- 
f.anthony.n.finch  <[e-mail]>  http://dotat.at/
Sole: East 5 or 6. Rough. Fair. Good.
_______________________________________________
rfc-interest mailing list
[e-mail]
https://www.rfc-editor.org/mailman/listinfo/rfc-interest