Re: Structure and Style revisited

Jelks Cabaniss wrote:
> 
> There is one HTML element in particular that jumps out as needing to be moved to
> Style -- BR.  When you have something like ...
> 
>         <h1>The Old Man<br>
>             and<br>
>             the Sea</h1>
> 
> ... the <br>s are there just to break the lines; they are for visual effect, not
> for sematic markup. 

yes

> Wouldn't it be better to have ...
> 
>         <h1>The Old Man and the Sea</h1>
> 
> ... with something like the following applied to it ...
> 
>         h1
> 
>                 break-after: words(3,4);

yuk, not a whole lot better. It means that if I edit the content, i need
to edit the style too; it means the style is only useful for thstone
instance.
It might be better to see what the <br> is being used for:

- to produce a short measure
- to prevent semantically jarring breaks:

The Old
Man and the
Sea

The former can be helped by setting margins and padding appropriately;
the latter by judicious use of non-breaking spaces or by incorporating a
line breaking dictionary into the formatter.


> ... where "words" could take a single value or a list?  Maybe even have
> "letters(...)" as well as "words(...)"?

where word is defined as a sequence of letters separated by spaces? does
that incluude ideographic spaces? Does that include languages which do
not put spaces between words?

What we are seeing here is a tip of an iceberg: to get the desired
formatting, reqyuires either a complex style specification or some hard
coded tweaks. Since the ful lrange of stylistic control is not there,
the tweaks get used.

> Isn't <BR> part of the legacy of presentational HTML?  How would one express the
> equivalent of <BR> in XML? 

<br/>

> I really haven't thought much about the syntax, but
> I do think there is a need to somehow express this in CSS.  As a precedent, we
> do have "page-break-before", etc. for paged media.

By moving away from "putting the breaks in" and towards "why did the
author want hard breaks" we are more likely to get a general, robust and
declarative solution, it seems to me.

--
Chris

Received on Monday, 22 February 1999 19:24:51 UTC