Re: Definition lists

Holger Wahlen (wahlen@ph-cip.Uni-Koeln.DE)
Wed, 30 Jul 1997 02:24:39 +0200 

Date: Wed, 30 Jul 1997 02:24:39 +0200
Message-Id: <199707300024.AA02146@jupiter.ph-cip.Uni-Koeln.DE>
To: www-html@w3.org
From: wahlen@ph-cip.Uni-Koeln.DE (Holger Wahlen)
Subject: Re: Definition lists

First, thanks to Albert Lunde for digging out the old
messages from the html-wg list on the topic and quoting some
excerpts.

Responding to my suggestion about the markup for his multiple
DD example, Jordan Reiter wrote:

| ><DT>bad
| ><DD><UL>
| ><LI CLASS="formal">Something that isn't good.
| ><LI CLASS="slang">Something that is good.
| ></UL>
| 
| Except that it seems to me that if you *are* dealing with 
| definitions, there's no need to complicate their meaning by 
| adding an unordered list to the fray.

I don't quite understand why this "complicates their
meaning". It's more complicated to type, sure, but how does
it affect the meaning of a definition when I say, "there's a
list of several definitions, this is one of them"? Is the
following bad - formal sense :-) - as well then?
	...
	<P>&quot;Bad&quot; can mean two things:</P>
	<UL>
	<LI CLASS="formal">Something that isn't good.
	<LI CLASS="slang">Something that is good.
	</UL>
	<P>Watch out, foreigners.</P>
	...


| >Regardless of the exact implementation, I agree that there
| >should be the restrictions listed by E. Stephen Mack
| ><estephen@emf.net>:
| >
| >| making it impossible to start with a definition, or to have
| >| only one definition without any terms or only one term
| >| without any definitions.
| 
| Notice, though, that my example is alright with his 
| restrictions.

Yes, of course; I didn't mean to imply anything else. I just
wanted to sum up these points as the ones that we apparently
all agree about (at least I haven't seen any other opinions
until now), so that we can consider them as definite
requirements for a new definition, while the question of
multiple terms or definitions is still under discussion. As
you put it:

| I think some consideration should be made to conforming the 
| DTD to these restrictions, rather than saying "well, these 
| restrictions fit into the current DTD anyway."

Agreed. Unless I have overlooked something, there are four
possibilites: If we want to keep the present content models,
which don't allow lists within DT, there are
	(1) (DT+, DD)+
(multiple DTs are allowed, multiple DDs aren't, plus the
restrictions given above) and
	(2) (DT+, DD+)+
(only those restrictions, arbitrary order otherwise); if the
DT content model is changed, "DT+" can be replaced by "DT" in
these two. To me, such a change doesn't seem appropriate,
though, and I'm still not convinced as to why the UL solution
shouldn't be the most logical one for a case with multiple
definitions; so my favorite remains (1).