>In message <199506031505.LAA11148@ebt-inc.ebt.com>, Gavin Nicol writes:
>>I don't think it is non-conforming SGML (the standard is somewhat
>>vague). I think the best way is not to say anything, other than it is
>>an error, and leave it to browser implementors to decide upon the
>>behaviour.
>
>This is the argument I've been waiting to hear. (except that the SGML
>standard is not vague on conformance.) It sounds good to me, but I
>thought we agreed that it was a good thing if all browsers handled
>this type of error the same way. The June 2 verbage is motivated by
>the fact that several widely deployed browsers behave that way.
It's just the simplest way, if you want to show the user that there
is something you can't handle. But it is by no way the best way.
You can't, as suggested by somebody else, go to a book with
*decimal* values. Most coding tables, especially ISO 10646/
Unicode, use *hexadecimal* values.
So even if several browsers behave that way, there are better
ways to behave, and there is no need for us to decide because
in fact user agents are not our business.
Regards, Martin.