Fwd: Excerpt from I18N f2f minutes on XML Schema

>From: "Martin J. Duerst" <duerst@w3.org>
>Subject: Excerpt from I18N f2f minutes on XML Schema

>Dear XML Schema WG,
>
>Below please find the excerpts from the minutes of our last meeting
>relating to your work. Please consider them as input to the
>last call resolition process. Please feel free to contact us at
>any time if you have any comments or questions.
>
>I will also send the new/changed comments to your public comments list.
>
>http://www.w3.org/International/Group/2000/06/ftf10/minutes
>
>Regards,   Martin.
>
>
>     [21]XML Schema 0
>
>      [21] http://www.w3.org/TR/xmlschema-0
>
>    DECISION: we reviewed and confirmed the comments sent to the IG list
>    by Martin and Misha (2000-05-25 and 26).
>
>     [22]XML Schema 1
>
>      [22] http://www.w3.org/TR/xmlschema-1
>
>    DECISION: we reviewed and confirmed the comments sent to the IG list
>    by Martin (2000-05-30).
>
>    In addition:
>
>    DECISION: we request that the xml:lang attribute be allowed on the
>    top-level element of a schema.
>
>     [23]XML Schema 2
>
>      [23] http://www.w3.org/TR/xmlschema-2
>
>    DECISION: we reviewed and confirmed the comments sent to the list by
>    Martin (2000-05-29), with the exception of the solution offered for
>    [43] which needs further refinement:
>
>    [43]: the basic i18n requirements are that new versions of Unicode
>    that introduce additional characters (or blocks) have to be supported
>    in a timely manner. Such new versions of Unicode may also include
>    corrections to the properties of existing characters. Both of these
>    can conflict with the requirement that an instance, once validated
>    against a particular schema in a particular version of the Schema
>    language, should remain valid forever (under that schema and version
>    of the Schema language). Consensus is that the Schema language 1.0
>    should be tied to a precise version of the Unicode Character Database
>    (namely the version current when Schema goes to Rec) and that there
>    should be a strong commitment, written in the spec, to update the
>    language for each new version of Unicode that adds new characters.
>
>    In addition:
>
>    DECISION: In 2.4.2.12, the encoding facet is badly named, especially
>    since "encoding" has a very different meaning in the XML 1.0 spec. It
>    should be named "transfer-encoding" or similar. Further, the sentence:
>
>      For example, "20" is the hex encoding for the US-ASCII space
>      character.
>
>    is a bad choice because since the facet is for binary data, not text,
>    using a character as an example is very confusing. The sentence must
>    be changed to something like:
>
>      For example, "01" is the hex transfer-encoding for a byte with the
>      value 1.
>
>    Clarification on [31] (3.3.1 language): there are 4 options: 1) XML
>    Schema does nothing (drop the current language); 2) constrain only to
>    something like ALPHA ["-" ALPHA]*; 3) have option 2 as basetype,
>    derive current 1766 (and its eventual successor) as derived type; 4)
>    constrain to current 1766 plus initial ISO 639-2 language codes.
>    Consensus is for 3.
>
>      [31] http://www.w3.org/TR/xmlquery-req

Received on Monday, 26 June 2000 04:16:13 UTC