On Mon, 12 Apr 2004, Ian Hickson wrote:

> On Mon, 12 Apr 2004, Jukka K. Korpela wrote:
> > On Mon, 12 Apr 2004, Ian Hickson wrote:
> >> If you have *blocks* of text in which the newlines are important, then it
> >> is preformatted text, and the <pre> element is relevant.
> >
> > Maybe, but <pre> has _four_ components in its meaning:
> > - preserve newlines
> > - do not wrap
> > - preserve spaces
> > - use monospace font.
>
> No, it has one meaning. "The contents have already been formatted and
> shouldn't be formatted further".

If you say so. This one meaning still has four components, so it is not
quite correct to recommend it if you only know that _one_ of those
components ("newlines are important") is desired.

> In effect it is an "escape" from HTML's
> user agent stylesheet.

Really? I thought <pre> existed long before style sheets were a reality.

> Python and JavaScript are both languages where newlines can affect the
> semantics of the code.

Python, yes. JavaScript, in part. In any case, <nobr> can be used just for
that part of the content where line breaks shall be prevented.

> I have lost track of whether you are arguing in favour of <code> and
> saying that <nobr> should be deprecated or arguing in favour of <nobr> and
> saying that it is not purely presentational (or some other argument). Both
> seem to be supported by the paragraph I quoted above, depending on how you
> read it.

Really? Anyway, I have argued that <nobr> is semantic, and so is <code>,
and they are orthogonal.

-- 
Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/