Re: Comments on HTMl 4 draft (9/Nov/1997)

Jukka Körpelä wrote:
> 
> On Tue, 16 Dec 1997, Sue Jordan wrote:
> 
> > Neil St.Laurent wrote:
> > ...
> > > B and I seem so natural and so often used to get rid of them would be
> > > dangerous.  Italics / Bold isn't always meant to draw special
> > > attention, or to put emphasis on, but often meant to just be
> > > different from the other text.
> >
> > Why not deprecate all of them? (read: relegate to CSS where
> > they belong)
> 
> Because CSS gives, at best, a suggestion about a possible rendering
> for some browsing situations.

Granted, given the usual CSS caveats.  It does seem,
however, that deprecating presentational elements would be
an incentive for wider deployment and better implementation
of CSS, without causing harm to existing documents.
 
> My favorite examples have been the use of the I element for taxonomic
> names, like "<I>Homo sapiens</I> L.", and for words used to refer to words
> themselves, not to those things to which they normally refers, as in
> "the plural of <I>ox</I> is <I>oxen</I>". They are not _essentially_
> presentational but structural. Using italics is definitely the
> _preferable_ presentation due to common practice of doing so in printed
> media.

Taking the liberty of combining your sentences results in a
paradox for me: " Using italics is definitely the
_preferable_ presentation... [sic. although]...not
_essentially_ presentational but structural."  I just don't
seem able to discern the difference between preferably
presentational and essentially presentational, even given
the taxonomic example. 

> So a revised specification of HTML should first give us structured ways
> of expressing such things before it could deprecate the I element.

Why ever would this order be important?  It isn't as if a
deprecated <I> or <B> is suddenly _not_ going to be
rendered, is it?  There is no finite length of life assigned
to deprecated elements, is there? 

> Please notice that things like structured markup for things like the one
> in my second example will soon gain importance. 

Having just re-read RFC 2070, I don't understand why use of
the LANG attribute would not be a better answer for your
second example than keeping or adding presentational
elements.  
 
Sue Jordan
[agreeing to disagree, and retiring from this thread]

Received on Thursday, 18 December 1997 11:40:36 UTC