Sectioning gone wild

[scholar_vit]

Scott Pakin in comp.text.tex writes:

I suppose you haven't read my favorite example of a badly structured document: the Universal 3D File Format (a.k.a. U3D or ECMA-363) specification, which is available from http://www.3dif.org/. In that document one can actually find a section 9.6.1.3.4.9.3.2.1.3! That represents a subsubsubsubsubsubsubsubsection. I defy anyone to find a deeper section nesting in a "real" document (i.e., one not created just to demonstrate how deep section nesting can go).

I've checked this. Indeed, there are sections 9.6.1.3.4.9.3.2.1.1, 9.6.1.3.4.9.3.2.1.2 and 9.6.1.3.4.9.3.2.1.3. An amazing feat of bureaucratic writing!

Comments

[observarius.livejournal.com]

Кодеры, фуле.

[xgrbml.livejournal.com]

Вот если бы еще на каждый уровень вложенности по новому размеру шрифта -- то-то текстик был бы!

[scholar-vit.livejournal.com]

Ну, они там что-то с отступами делали... как это будет по-русски - lame?

[ygam.livejournal.com]

Wow!

[dimrub.livejournal.com]

Take any document describing a MIB (any MIB), you'll find nesting twice as deep.

[scholar-vit.livejournal.com]

I think this is a different matter. "Sections" in MIB are actually addresses of SNMP variables. The structure is intended for computer parsing. Most (all?) MIB browsers represent it as a clickable expandable hierarchy rather than a running text. This is like a hierarchic file system: my computer probably has even deeper nesting of directories and files.

Standards, on the other hand, are intended to be read and referenced by humans. However, a human cannot easily spot the difference between 9.6.1.3.4.9.3.2.1.1 and 9.6.1.3.4.9.2.1.1, which defies the purpose of section numbering.

[maroussia.livejournal.com]

Well, in my PhD... :-(

[netch]

Это ничуть не хуже традиции европейско-советского стиля сквозной нумерации. Был бы пункт 4313:)

[scholar-vit.livejournal.com]

А правильно - аккуратно выбирать уровни разбиения, чтобы получалось что-то вроде 6.24.30...