Re: Proposed resolution for #21: duplicateAttributes

[James Clark <jjc@jclark.com>]

> Here is my propsal: This is not an error.  Implementations
> MAY issue error messages.

If the spec is going to say "Implementations may issue error messages for
duplicate attributes", then the spec has to define exactly what constitutes
a duplicate attribute.  How should it do this?

I am also concerned that this will create interoperability problems (like
the ambiguity constraint in XML).  Either it should be an error and
implementations should be required to detect it, or it should not be an
error and implementations should be required to accept it.

Could one not impose some simple constraint that would not require GNF
normalization to implement? For example, something like saying that in

<group> p1 p2 </group>

the set of possible attribute names occurring in p1 must be disjoint from
those occurring in p2.

James