[osis-core] Dallas TC meeting minutes for May 22

Kirk Lowery osis-core@bibletechnologieswg.org
Thu, 22 May 2003 18:50:13 -0400

Also at: http://whi.wts.edu/OSIS/admin
Kirk E. Lowery, Ph.D.
Director, Westminster Hebrew Institute
Adjunct Professor of Old Testament
Westminster Theological Seminary, Philadelphia

Theorie ist, wenn man alles weiss und nichts klappt.
Praxis ist, wenn alles klappt und keiner weiss warum.
Bei uns sind Theorie und Praxis vereint:
nichts klappt und keiner weiss warum!


                               May 22, 2003

TC members present:
Steve DeRose, chair
Kees Deblois, vice-chair
Patrick Durusau
Troy Griffits
Kirk Lowery, secretary pro tem

Mike Perez
Todd Tillinghast

Issues and Decisions

   [1.1.1] Remove dead element <cell>


   [1.1.2] Milestones

   <milestone> marks special breaks. <milestoneSE> encapsules <closer>,
   <abbv>, <div>, etc. It is intended to be a stand-in for crossing
   elements. It's the overlapping hierarchy solution.

   Two views of markup hierarchy: (1) BSP (Book, Section, Paragraph)
   approach: viewing the biblical text hierarchy as pericopes, chapter and
   verse secondary; (2)  BCV (Book, Chapter, Verse) approach: chapter and
   verse primary, pericopes secondary.

   From software point of view: BCV preferred. Academics like it. BFES
   likes it. UBS/ABS/SIL like BSP. One way is better than two. Can satisfy
   the BCV camp using osisRefs using BSV.


        Applications must support BSP (BCV optional)

        Documents, including Bible texts must include osisIDs

        Add "chapter" to <mileStoneSE> type attribute

        Keep <verse>container, which is canonical container for Bibles, but
        other texts (e.g., Church fathers) have canonical containers, too.
        So we need to specify "what container(s) are canonical?"See point

        An attribute called canonical is globally available and always
        optional, inherits like lang attribute.

        The default for <verse> is canonical true, for <note> is canonical
        false and no other element type specifies a default for the
        canonical attribute; if canonical is true, require osisID; create
        <chapter> element; add  chapter to mileStoneSe type; remove chapter
        from div type; <chapter> is optional splitID attribute; <div> does
        not have a splitID

        Elements that cannot be splittable






        If <div>crosses chapter, <chapter>must break (either splitID or

        milestoneStart, milestoneEnd, should be globalWithoutType,
        attribute value should be milestoneSE which now does not combine
        with osisMilestoneSE to allow extension.

        [Document how to mark up book/chapter/verse when there is no
        section/paragraph information. ]

        So one can have a <book><chapter><verse></verse></chapter></book>
        container hierarchy, if there are no sections or paragraphs. If
        there are sections and/or paragraphs, then at least the cross-over
        chapters and verses must be milestones.

       ACCEPTED AS "OSIS conforming"


   Is it permissable to have chapter/verse containers and milestones in the
   same <osisText>?

   Best Practice: Recommend that one avoids mixing chapters as containers
   and chapters as milestones within a single document.


   Simple <mileStone> can have the canonical attribute.<mileStoneEnd> does
   not get the canonical attribute because it is error-prone and for data


   <mileStoneStart> may have canonical attribute, but <mileStoneEnd> may
   not have the attribute.


   [2.1] Add section, front, body, back, titlePage, introduction, index,
   preface, afterword, colophon, entry, dedication, acknowledgments,
   majorsec, subsec, coverPage, gazetter, commentary, devotional, map,
   imprematur to the type attribute on <div>.


   [2.2] Add all children of <verse> to <catchWord>.


   We want an example for the documentation for <note> in <catchWord>.

   Issue 2.3 "lang/script/ews" deferred until tomorrow.


   [2.4] Allow <table> and <list> in <p> and <speech>. Subject to Patrick's
   objections, elimination any differences of content model and attributes
   between  <q> and <speech>, i.e., they become synonyms or aliases of each

   [2.5] osisID as a list, then it points @ "at" with osisRef with grain

   We need to say in the documentation, that applications are expected to
   discard or map the grain when they cannot get a hold of the right

   DEFERRED until tomorrow[Todd is writing up a summary for approval.]

   [2.6] Should osisRefbe allowed as a <list>?

   Are there discontinuous references?

   DEFERRED until tomorrow

OSIS Technical Committee Minutes ? Dallas          1