The term "Unicode string" is unspecified and, from what the term suggests, the examples are confusing. A "Unicode string" is presumably a string of Unicode characters, possibly with "/" characters and therefore intended to be interpreted as a file system path. It is, however not logical, to assume a Unicode string is already -escaped (otherwise, "Unicode string" and "IRI" would be identical concepts). Consider, eg, the string "/test/50_chance.xml". Thus, an original Unicode string of "/41/61.xml" (a sequence of the Unicode characters "/", "", "4", "1", "/" …) must be represented as an IRI or URI "/2541/2561.xml". Also if a Unicode string is intended to be interpreteted as a unified file system path with national characters, a backslash "\" character in the original string should be mapped to a "/" character in the IRI, and not -escaped.

Verify the intended encoding rules and examples and specify the term "Unicode string".

Part 2, section :A.4

ed

Proposed Disposition of DIS 29500 Comment DE-0158 (Modified: 2007-12-07) Agreed; the following change will be made to Part 2, §Annex A, page 64, line 3: Package clients might use Unicode strings of Unicode characters to represent relative references to for referencing parts in a package. Further in this annex such strings will be refered to as Unicode strings. [Example: Values of xsd:anyURI data type within XML markup are Unicode strings. end example]