Re: [Tools-discuss] Interest in title-less sub-sub-sections?

[Julian Reschke <[e-mail]>]

Am 13.04.2021 um 01:29 schrieb Nico Williams:
> Consider the x.680 and x.690 series documents.  They typically have many
> tiny sub-sub-sections, with many sub-sub-sections having just one or two
> paragraphs and no title.
>
> Presumably those tiny sub-sub-sections lack a title because a title
> would make the documents less readable.  For tiny sub-sub-sections, the
> lack of a title almost makes the sub-sub-section number akin to a
> paragraph number.
>
> E.g., x.690, where sections 8, 8.1, 8.1.1, ..., 8.19 have titles, but
> 8.1.1.1, ..., 8.19.1, 8.19.2, ..., do not:
>
>     8 Basic encoding rules
>
>     8.1 General rules for encoding
>
>     8.1.1 Structure of an encoding
>
>     8.1.1.1 The encoding of a data value shall consist of four components
>     which shall appear in the following order:
>
>     ...
>
>     8.19 Encoding of an object identifier value
>
>     8.19.1 The encoding of an object identifier value shall be primitive.
>
>     8.19.2 The contents octets shall be an (ordered) list of encodings of
>     subidentifiers (see 8.19.3 and 8.19.4) concatenated together.
>
>     Each subidentifier is represented as a series of (one or more)
>     octets. ...
>
>     8.19.3 The number of subidentifiers ...
>
> What's really nice about this is how easy it makes it to make very
> specific references.  Also, because most sub-sub-sections are so short,
> it is easy to refer to most every normative requirement/recommendation/
> option.

As I just sid in the other thread: we already number paragraphs; we just
do not display these numbers.

In the example above, "8.1.1.1" would get an anchor like
"section-8.1.1-1". That's totally fine for references, the anchor is
easily discoverable from the HTML format.

> Would we need any changes to the xml2rfc schema or tooling to make this
> style possible for Internet-Drafts and RFCs?

It would need to start with a change in the Style Guide. I would prefer
to use what we already have (and maybe tune *that* instead).

Best regards, Julian