Re: default charset broken

Martin Duerst <duerst@w3.org> wrote in
news:4.2.0.58.J.20030608155637.07447af8@localhost:

> They could always use numeric character references
>> (which work better anyway).
>
> Would you ever want to try to edit a Japanese Web
> page with numeric character references?

I would of course use an authoring tool and write the Japanese
characters directly. The authoring tool would be responsible for
encoding the characters as bytes.

> Why would they work better in such a case?

There are in my experience tons of search engines and other
software (e.g. blogging software and browsers) which have problems
displaying/handling UTF-8 encoded text (they just handle the bytes
*as if* it was in their default encoding, usually ISO-8859-1).
Using numeric character references solve this problem.

-- 
Karl Ove Hufthammer