I would argue that XML, in practice, has many more. Everything in XML is a string by default, and only an XSD can provide actual type information. Lots of things don't, and those that do often define their own boolean types.
> I would argue that XML, in practice, has many more. Everything in XML is a string by default, and only an XSD can provide actual type information. Lots of things don't, and those that do often define their own boolean types.
I think the problem happens when there are 22 ways to write them in the standard and you only know 16 of them (or 27 of them). You're going to be surprised (by the standard) at some point.
If the standard doesn't define anything, then you at least know the responsibility for the definition is between you and your integrations, so at least you won't be surprised by the standard itself.
As far as standard representations go I think the ideal numbers are 0 and 1: either it should define nothing or pick one unambigous standard representation.