W3C home > Mailing lists > Public > public-html@w3.org > March 2008

Re: Marking Up Poetry

From: Dr. Olaf Hoffmann <Dr.O.Hoffmann@gmx.de>
Date: Sat, 1 Mar 2008 12:36:01 +0100
To: public-html@w3.org
Message-Id: <200803011236.01839.Dr.O.Hoffmann@gmx.de>

Sam Kuper wrote:
> It strikes me that what Olaf is trying to get across here is something
> rather fundamental along the following lines. Please, Olaf, correct me if
> I'm wrong, and forgive the crude paraphrase.
> "There are basically two forms natural language comes in:
> - prose
> - poetry
> All texts, regardless of domain (law, cookery, transportation, literature,
> science, lyrics, whatever), are composed of some amount of prose, poetry or
> both.

This is a rough simplification, but already a useful approach to understand,
how written text appears.
From history, I think, there was first conversation - aural interactivity only
with only a relation to the present situation.
But then some culture came up and rules and experiences of groups.
There was a need to memorise this information somehow. Because
tar and gzip were not invented yet, they did it with poetry, rythm and
songs, this was the origin of poetry as a method to archive information
for a group of people. Much later is was possible to write down information
somehow and often it was not very important anymore to memorise 
information in someones head. Strophes and strophe lines began to
degenerate. This can be still seen in old texts, fairy tales or for example
the bible, full of stories, originally stored in something like strophes and
simple rules to memorise - ten fingers - Ten Commandments. 
Written law texts derived from strophes too, when the rules started to
get more complex, that they have to be written (and the people got
indifferent about them, because they are too complex, but this is
another problem).
A typical written law paragraph has a more complex structure than an 
ordinary paragraph and typically slightly different form a poetry strophe,
but it is list like content too.
An ordinary paragraph can be seen as a degenerate strophe derived 
from poetry without any further substructure (this was well defined
in HTML4, XHTML 1.x and is currently unfortunately corrupted in
the HTML5 drafts, p is degenerated to an alias of div.
Therefore for prose content in HTML5 a paragraph element is 
required now to cover this new gap - but this is another problem,
not directly related to poetry content.

To resume, poetry and prose can be seen as two poles. 
You know, ice bears are only at the north pole, penguins at the
south pole. With ice bears at the south pole soon there would
be no penguins anymore. They are well separated species,
but there is much life between north and south pole ;o)

> HTML is good at marking up prose text, but not poetry text. Support poetry
> and you cover all natural language text.
> An implicit point is that all text that is not poetry is prose, so things
> like legalese, footers and list items are typically prose. (One could
> conceivably have a poetry footer, or a country might decide that all its
> legal documents are to be written in verse, but by current norms these
> would be atypical cases to say the least.)"

See above, historically law text and poetry is closely related and law
text has some minor specific requirements to the meaning of a
paragraph or a dl/dt/dd list too, this is already covered too by the
proposal in the wiki about a refined definition of dl/dt/dd with 
an additional attribute, to cover such specific requirements.


Ian Hickson wrote:
> Note that "paragraph" is a defined term in HTML5, which specifically says
> that it covers poetry:
> | A paragraph is typically a block of text with one or more sentences that
> | discuss a particular topic, as in typography, but can also be used for
> | more general thematic grouping. For instance, an address is also a
> | paragraph, as is a part of a form, a byline, or a stanza in a poem.
>  -- http://www.whatwg.org/specs/web-apps/current-work/#paragraphs

In HTML4, XHTML1.x this is excluded, because p cannot contain something
like da dl/dt/dd list to markup poetry in a useful way.
In the current HTML5 drafts, indeed the p is redefined mainly as an alias for
div. But because it still has no defined substructure like a list, one has to
put dl/dt/dd into it, which are currently are not compatible the current
use of them to markup poetry, this is not available in HTML5. Anyway, 
even if it would be, the p element would be quite superfluous for the markup
of a strophe, because this is already covered by the used list like elements.
And because p is specified as a paragraph in HTML4, XHTML1.x it would be
a backwards incompatible use as a containter for a complete poem or
song text, because the HTML4, XHTML1.x is from the point of poetry a
degenerate descendant of a strophe without any further substructure,
therefore it can never contain itself strophes and strophe lines.


Sam Kuper wrote:
> On my current understanding of Olaf's ontology, the term "paragraph" isn't
> neutral: it applies only to one side of the prose/poetry distinction.
> Therefore, <p> is not neutrally defined insofar as it is stated to
> represent a paragraph.

It is well defined in HTML4, XHTML 1.x. For the current HTML5 draft there
is obviously a requirement to define a new element for a simple text
paragraph to be able to markup any prose content in a somehow useful
way ;o)


Leif Halvard Silli:

> I also would like to mention that historically, I don't think there is a
> difference between a paragraph and a heading - headers were just bits of
> the text that was "highlighted" by being placed in a paragraph of their
> own.

Now, historically a heading or a title is something like a short summary,
abstract or caption of the following text, therefore it has a very specific
semantical meaning. This is almost the same for prose and poetry, very
generic. It is always well separated from the other content and on top of
the text fragments, it is related to.
Sometimes now it is used slightly different in fiction, drama and newsletters,
but this is mainly only a more abstract and complex but at least similar
relation to the following text as in the historical/classical approach. 

Received on Saturday, 1 March 2008 11:47:12 UTC

This archive was generated by hypermail 2.4.0 : Saturday, 9 October 2021 18:44:27 UTC