[osis-core] <identifier type="OSIS"> - Lets go with Troy's position
Todd Tillinghast
todd at contentframeworks.com
Wed Nov 10 09:07:51 MST 2004
Troy,
The order of identifiers is better or worse depending on the work and
type of work.
1) For the case of KJV putting the name and date prior to the publisher
works well for fall back.
2) For the large number of Bible translations the name is not important
because the language and publisher (and possibly the date) uniquely
identify the translation. In these cases the Name is simply the
language code repeated. So Bible.DUG.WBTI.2004 would be better than
Bible.DUG.DUG.2004.WBTI and would be better than
Bible.DUG.WBTI.DUG.2004.
3) For sermons putting the person delivering/publishing the sermon,
followed by the date delivered, followed by the name would work better
for fall back.
4) For commentaries there will be cases where the publisher or the name
would be better placed first.
The attractive thing about putting the publisher prior to the name is
that we can say that aside from some reserved Names all of the values
after the publisher are up to the publisher (similar to sub domains
aside from the reserved Names).
In the interest of simplicity, agreement, and clean fall back in the
majority of cases where fallback will occur I am suggesting that we go
with Troy's preferred syntax
(Bible.en.KJV.1611.Crosswire.OtherIdentifers.If.Desired).
Todd
> -----Original Message-----
> From: osis-core-bounces at crosswire.org [mailto:osis-core-
> bounces at crosswire.org] On Behalf Of Troy A. Griffitts
> Sent: Tuesday, November 09, 2004 11:23 PM
> To: osis-core at crosswire.org
> Subject: Re: [osis-core] <identifier type="OSIS">
>
> Todd,
> I still see no functional advantage to your suggestion. What
don't
> you
> like about:
> Bible.en.KJV.1769.CrossWire.2004.WordLevelMarkup
>
> which makes you want to arrange it to:
> Bible.en.OSIS.KJV.1769.Crosswire.WordLevelMarkup.2004
>
> I don't understand what the confusing and extraneous OSIS is for, and
> why you'd want our publish date after any misc tags like
WordLevelMarkup.
>
>
> These things are not intuitive and unclear and seem to be to
facilitate
> a _possible_ inconvenience of publishers who want their name first in
> the tag.
>
> I believe my suggested structure is concise and utilitarian.
>
>
>
> Todd Tillinghast wrote:
>
> > Troy,
> >
> > Perhaps a compromise for common works would be something like:
> > Bible.en.OSIS.KJV.1769.Crosswire.WordLevelMarkup.2004
> >
> > Where for OSISReservedNames the Publishers is OSIS and then
following
> > the date the actual publisher/encoder of a specific instance would
> > follow.
> >
> > This maintains the structure for the less common cases and still
> > provides full fallback behavior.
> >
> > So a reference to the KJV in general would be
Bible.en.OSIS.KJV.1769.
> >
> > The trick could possibly be something like
Bible.en.OSIS.NIV.1984.IBS
> > rather than Bible.en.IBS.NIV.1984.
> >
> > Thoughts?
> >
> > Todd
> >
> >
> >>-----Original Message-----
> >>From: osis-core-bounces at crosswire.org [mailto:osis-core-
> >>bounces at crosswire.org] On Behalf Of Troy A. Griffitts
> >>Sent: Tuesday, November 09, 2004 12:19 PM
> >>To: osis-core at crosswire.org
> >>Subject: Re: [osis-core] <identifier type="OSIS">
> >>
> >>Just a few quick notes.
> >>
> >>s/Crosswire/CrossWire/
> >>
> >>I am still in favour of:
> >>
> >>OSISWorkType '.' Language '.' Name ('.' PublisherOrAuthor ('.' Date
(
> >>'.' PublisherExtension )* )? )?
> >>
> >>I think if we are programmatically declining the ID from order of
> >>specificality to generality from right to left, the above order
makes
> >>more sense.
> >>
> >>e.g.
> >>
> >>Bible.en.CEV.CrossWire.2004.b
> >>If the book with the exact above ID does not exist, I would feel
> >>confident trimming from right to left to find a suitable match.
> >>
> >>Debatable in my mind: transposing [date] and [publisher].
> >>
> >>OSISWorkType
> >>+GeneralBook (e.g. Josephus, That Hideous Strength, etc.)
> >>
> >> -Troy.
> >>
> >>
> >>
> >>
> >>
> >>Todd Tillinghast wrote:
> >>
> >>
> >>>We have been over this many times but never reached a conclusion on
> >
> > the
> >
> >>>structure for the value of <identifier type="OSIS">, Patrick and I
> >
> > are
> >
> >>>working on getting the OSIS User Guide completed and we need to
> >
> > finalize
> >
> >>>this piece.
> >>>
> >>>Proposal:
> >>>
> >>>WorkSelfIdentifier ::= OSISWorkType '.' Language '.'
> >
> > PublisherOrAuthor
> >
> >>>'.' Name '.' Date ( '.' PublisherExtension )*
> >>>
> >>>WorkRefIdentifier ::= OSISWorkType '.' Language (
> >
> > '.'PublisherOrAuthor
> >
> >>>('.' Name ('.' Date ( '.' PublisherExtension )* )? )? )?
> >>>
> >>>OSISWorkType ::= 'Annotation'
> >>> | 'Bible'
> >>> | 'BibleStudy'
> >>> | 'Concordance'
> >>> | 'Commentary'
> >>> | 'Devotional'
> >>> | 'Dictionary'
> >>> | 'ReferenceSystem'
> >>> | 'Sermon'
> >>>
> >>>Language ::= '_'
> >>> | ISOLanguage
> >>> | EthnologueLangauge
> >>>
> >>>ISOLangauge ::= ISO two or three letter language code lower
> >
> > case
> >
> >>>EthnologueLanguage ::= Ethnologue three letter language code in
upper
> >>>case
> >>>
> >>>PublisherOrAuthor ::= '*'
> >>> | OSISReservedPublishers
> >>> | NCName
> >>>
> >>>OSISReservedPublishers ::= 'ABS'
> >>> | 'BL'
> >>> | 'CCEL'
> >>> | 'Crosswire'
> >>> | 'IBS'
> >>> | 'NTM'
> >>> | 'SBL'
> >>> | 'SIL'
> >>> | 'SF'
> >>> | 'UBS'
> >>> | 'WBT'
> >>> | 'WBTC'
> >>> | 'WBTI'
> >>> | 'VS'
> >>> (fill in the rest of the values in the
current
> >>>doc and also for members of the Forum of Bible Agencies)
> >>>
> >>>Name ::= OSISReservedNames
> >>> | NCName
> >>>
> >>>OSISReservedNames ::= 'CEV'
> >>> | 'GNT'
> >>> | 'KJV'
> >>> | 'NASB'
> >>> | 'NIV'
> >>> | 'RVR'
> >>> (fill in the rest of the values in the
current
> >>>doc)
> >>>
> >>>Date ::=
> >>>Year(Month(DayOfMonth(Hour(Minute(Second)?)?)?)?)?
> >>>
> >>>Year ::= [0-9][0-9][0-9][0-9]
> >>>
> >>>Month ::= [0-9][0-9]
> >>>
> >>>DayOfMonth ::= [0-9][0-9]
> >>>
> >>>Hour ::= [0-9][0-9]
> >>>
> >>>Minute ::= [0-9][0-9]
> >>>
> >>>Second ::= [0-9][0-9]
> >>>
> >>>PublisherExtension ::= NCName
> >>>
> >>>
> >>>The '_' value of language would be used in the case where the work
is
> >>>language independent as would be the case for a ReferenceSystem.
> >>>
> >>>The '*' value for PublisherOrAuthor would be used for cases where
a
> >>>reference is being make _to_ a work with the specified Name by any
> >>>PublisherOrAuthor. (I am hopeful that this reasonably addresses
> >
> > Troy's
> >
> >>>concern with having Name follow PublisherOrAuthor for cases like
the
> >>>KJV).
> >>>
> >>>The idea of fallback is intended.
> >>>
> >>>The plan is to put out a version of the OSIS User Guide this week,
so
> >>>please voice any comments or objections before that time.
> >>>
> >>>Todd
> >>>
> >>>_______________________________________________
> >>>osis-core mailing list
> >>>osis-core at bibletechnologieswg.org
> >>>http://www.bibletechnologieswg.org/mailman/listinfo/osis-core
> >>>
> >>>
> >>
> >>_______________________________________________
> >>osis-core mailing list
> >>osis-core at bibletechnologieswg.org
> >>http://www.bibletechnologieswg.org/mailman/listinfo/osis-core
> >
> >
> >
> > _______________________________________________
> > osis-core mailing list
> > osis-core at bibletechnologieswg.org
> > http://www.bibletechnologieswg.org/mailman/listinfo/osis-core
> _______________________________________________
> osis-core mailing list
> osis-core at bibletechnologieswg.org
> http://www.bibletechnologieswg.org/mailman/listinfo/osis-core
More information about the osis-core
mailing list