proposal for ISSUE-109 ill-typed literals

This is a proposal to finish off the work done for ISSUE-109, ill-typed
literals, in the core WG documents.

The WG has discussed the status of ill-typed literals, and has been leaning
towards having an ill-typed literal being inconsistent.   The Semantics
document reflects this understanding, changing the L2V mapping so that
ill-typed literals do not denote so a triple containing an ill-typed
literal is false in every D-interpretation.

The Concepts document says, in Section 5.5
    Otherwise, the literal is ill-typed, and no literal value can be
    associated with the literal. Such a case, while in error, is not
    syntactically ill-formed.
This should be changed to
    Otherwise, the literal is ill-typed, and no literal value can be
    associated with the literal. Such a case produces a semantic
    inconsistency but is not syntactically ill-formed and
    implementations MUST accept ill-typed literals and produce RDF
    graphs from them.  Implementations MAY produce warnings when
    encountering ill-typed literals.

I hesitate to include the last sentence, which would be the replacement for
the Issue note in Section 5.5.

The WG should close the substance of ISSUE-109 by voting on a resolution.

Proposed:  Close ISSUE-109 by making ill-typed literals not denote and
requiring that RDF implementations accept ill-typed literals and produce RDF
graphs from them, possibly producing a warning.

Received on Wednesday, 17 April 2013 17:09:07 UTC