Re: [CSS21] response to issue 115 (and 44)

* Bert Bos wrote:
>Is 0xDB in MacThai a "zero-width no-break space," or rather a U+2060
>"word joiner"?

This depends on whom you ask, this changed in December 2002.

>> You did not address what to do if the processor encounters an encoding
>> error.
>
>Do I have to? CSS gives rules for parsing things that might someday
>cease to be errors. But I don't see what these non-characters can ever
>be useful for, so why should I say how to parse them?

No, you are not required to address it, currently the specification is
not either, but I as an implementer need to address it and I want the
specification to tell me and my competitors what to do in order to ease
and back my implementation.

This is not about non-characters, this is about digital representations
of stylesheets that can't be properly decoded because the representation
is not encoded in the determined encoding or because the determined
encoding is not supported. These are error conditions and I want to know
how to handle them.

Received on Sunday, 22 February 2004 21:27:29 UTC