Re: What about  ? (fwd)

Martin J Duerst wrote:
>If these characters appear singly between other characters, e.g.
>words, this certainly makes sense. But the situation could be a little
>bit different e.g. for an NBSP at the end of a line. 

The teatment of such like sick cases in purely an implementation issue. The specs 
tell us that at the NBSP a line break is to be prevented. They also tell us that 
at a space or any whitespace a line break is allowed. If an NBSP is adjacent to 
whitespace there seems to be a conflict between the two.

>                                            General processing
>would replace the end-of-line by a single space, and thus allow
>breaking after the NBSP, whereas the user's intention most
>probably is to not allow a break.
>
>So the desired treatment might look as follows:
>- Keep these "special" spaces.
>- Completely eliminate surrounding whitespace (as opposed to
>	other occurrences of whitespace, where it is just collapsed
>	to a single space).

It is not productive to even try to guess what the author intended 
with such an obvious error. It's like asking the browser to correct 
capitalization errors. I tend to say that the NBSP should be at the end of the 
line - i.e. it should be treated just like any other graphic (non-whitespace) 
character.

>There are other cases where a special treatment of whitespace
>is necessary, 

These special spaces are not normally considered whitespace.

>... This is of course not a very good argument;
>the difference between space and NBSP is only the non-breaking
>property of the later, and if both are treated equally, the
>"raison d'etre" of the later is gone.

Hear, hear.

Jonathan Rosenne

Received on Tuesday, 30 July 1996 09:58:26 UTC