W3C home > Mailing lists > Public > www-html@w3.org > April 2004

Re: <NOBR> - Returning to the question....

From: Jukka K. Korpela <jkorpela@cs.tut.fi>
Date: Sun, 4 Apr 2004 09:38:51 +0300 (EEST)
To: www-html@w3.org
Message-ID: <Pine.GSO.4.58.0404040910570.27115@korppi.cs.tut.fi>

On Sat, 3 Apr 2004, Ernest Cline wrote:

> I've noticed that in all of the examples you have put forth, Jukka,
> have the following thing in common, they contain content you want
> to indicate as being a single syntactic unit.

That's a general property of logical text-level markup, is it not?
In this case, it's a matter of a special kind of being a single unit.
For example, the text in a link (<a href="...">...</a>) should
preferably be kept on one line, partly to avoid confusion (does the blue
underlined text at the end of a line and at the start of a new line
constitute one link, or two links?). But <nobr> would say that
unconditionally.

> However, prevention
> of line breaking is only one presentational way of indicating that.

I would rather say that line breaking is just one, though the most common,
feature of presentation that should be affected. In speech synthesis,
there might be longish pauses for various reasons, remotely corresponding
the line breaks caused by the nature of a visual presentation media; and
no long pause should be introduced by a browser within <nobr> a element.

> Presentationally, that could be indicated in other ways such as
> by giving the content a different background, font, and/or border.

But that's a completely different issue. The <nobr> markup does not
indicate any kind of emphasis or being different from normal text,
except that a string of characters constitutes a group that shall not
be broken.

> Just as <i> is not the same as <em>, <nobr> is not equivalent
> to the semantic element you want in HTML.

If the difference is as small as between <i> and <em> (which is usually
relevant to theorists only) or between <b> and <strong>, there's little
reason to introduce _two_ elements (<nobr> and <syntacticunit>, perhaps,
following the tradition of naming the logical element with a longer name
so that few people want to use it) at this point. I'd compare this to
<hr>: as its name suggests, many people take it just as a way of drawing a
horizontal line (and keep asking why there's no vertical counterpart); and
we theorists can happily explain it as a poorly named logical element for
a change of topic, which is actually why people usually want to draw a
line, they just don't quite think it in such terms.

> The questions thus become,
> 1) Is it something that can be distinguished from <code>?

It has nothing to do with <code>. If <nobr> happens to contain computer
code, as it often does, that's incidental; in that case, you would use
<nobr><code>...</code></nobr> of course.

> 2) If so, is it common enough to warrant having as an element
> distinct from the generic <span>?

It has nothing to do with <span> either. The <span> element is a pure
container with no meaning (logical or presentational).

> There are all sorts of semantic elements that could be
> added but aren't.

This isn't really about adding an element. It's about recognizing and
standardizing an element (or, actually, two elements - <wbr> should be
added too, and it should not be any harder problem than <br>, since
<wbr> is simply an allowed breaking point, <br> is unconditional)
that has existed in HTML from rather early days.

The <nobr> element is partly an inline counterpart of <pre>, except that
<nobr> says less about the presentation - it does not enforce white space
preservation. Logically, <pre> says that its content is a string of
characters grouped into lines so that the exact line structure as well as
the use of spaces (and tabs) is significant. Compared to this, <nobr> is
much simpler - and more logical.

-- 
Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/
Received on Sunday, 4 April 2004 01:38:56 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:00 GMT