5384 2008-01-16 18:44:11 +0000 Uniqueness of alias URIs 2008-06-05 19:10:11 +0000 1 1 1 Unclassified SML Interchange Format unspecified PC Windows XP RESOLVED FIXED P2 normal LC 1 kirk.wilson kumarp virginia.smith public-sml oldest_to_newest 18364 0 kirk.wilson 2008-01-16 18:44:11 +0000 The following statement (actually three) statements in section 5.3.2 of SML-IF seem to be equivalent: Each member of the set of all alias URIs in an SML-IF document MUST be unique. That is, no two alias URIs in a given SML-IF document may be equivalent. All alias URIs in an SML-IF document MUST be unique. Is there a difference? (If so, it should be made more explicit.) 18381 1 johnarwe 2008-01-17 13:17:21 +0000 I think they are expressing the same concept in different words. fwiw, "equivalent" might be confusing here. I usually see that in the context of how to normalize/compare URIs, e.g. RFC 3986 section 6.1: 6. Normalization and Comparison . . . . . . . . . . . . . . . . . 38 6.1. Equivalence . . . . . . . . . . . . . . . . . . . . . . 38 Given our current rules around URI comparison, I don't think we need to invoke "equivalent" or any variation. We already define what equality (the strongest form of equivalence) means to SMLIF, and don't need anything weaker IIRC. Net, I'd strike all phrases that appear to relate URIs and any form of equivalence other than straight equality. 18396 2 virginia.smith 2008-01-17 19:32:56 +0000 Editors to deal with both comments. Remove 'equivalent' and all its variants. Reference the section in IF that defines equality of 2 URIs. 18536 3 kumarp 2008-01-25 02:16:57 +0000 changed the section title, from: 5.3.1 URI Equivalence to : 5.3.1 URI Equality changed all references to URI equivalence to use the term URI equality instead. for example, from: If the URI is equivalent to the URI in an alias (see 5.3.1 URI equivalence), ... to: If the URI is equal to the URI in an alias (see 5.3.1 URI Equality), ...