Re: Tags inside themselves?

newtonjs@char.vnet.net (Stan Newton)
Date: Tue, 13 Sep 94 09:10:36 EDT
Message-id: <199409131309.AA14961@char.vnet.net>
Reply-To: newtonjs@char.vnet.net
Originator: html-wg@oclc.org
Sender: html-wg@oclc.org
Precedence: bulk
From: newtonjs@char.vnet.net (Stan Newton)
To: Multiple recipients of list <html-wg@oclc.org>
Subject: Re: Tags inside themselves?
X-Listprocessor-Version: 6.0c -- ListProcessor by Anastasios Kotsikonas
X-Comment: HTML Working Group (Private)
Dear Dan,

I think I understand now how nested highlighting would be processed and 
Murray makes a good point about the intent of highlighting (i.e. italics in 
italics should really be plain).

I'd like to take this one step further. If I understand the present tagging 
rules, the following example would be proper.

<B>text and more text and <B> and then some bolded text</B></B>

Before today, I would have said this is meaningless tagging and that the 
internal set of B tags should be ignored. Now I have to ask, "Does this mean 
the internal set of B tags should be presented as plain?"

Hopefully not. It would seem useful to distinguish highlighting over a 
paragraph level font from the highlighting-on-highlighting case.

By the way, B is just used as an example. All highlighting tags appear to 
defined this way. The DTD now says:

<!ELEMENT (%font;) --(%phrase.content)+>

where %font is defined as a member of %phrase.content.

This seem to say that B is allowed directly inside B. The HTML DTD 
Reference, (pages 6.x), in fact reports it this way. Shouldn't the 
definition somehow exclude the specific tag from appearing inside itself?

NOTE 1: I don't see a problem with an alternating sequence such as a B tag 
inside an I tag inside a B tag. Just the B..B, CITE..CITE, etc. combinations.

NOTE 2: I really like the presentation format you use in the HTML DTD 
Reference. This helps me a lot in understanding the rules in the DTD itself 
which can be dense. (It was in the Reference that I noticed this feature of 
B allowed in B, etc. which I am raising here.)





Stan Newton
Newton Computing Solutions