[osis-core] osisCore_Candiate_1.1_003 - 9 Tricky point with default work and refSystem

[Harry Plantinga]

Maybe we should do the minimum with reference systems now
and hash out these additional ideas when reference systems
are fleshed out more fully.

For now, perhaps we could get away with not saying anything
about the reference system in use--only specify a default 
reference system.

-Harry

> -----Original Message-----
> From: owner-osis-core@bibletechnologieswg.org 
> [mailto:owner-osis-core@bibletechnologieswg.org] On Behalf Of 
> Todd Tillinghast
> Sent: Wednesday, August 21, 2002 1:52 PM
> To: osis-core@bibletechnologieswg.org
> Subject: RE: [osis-core] osisCore_Candiate_1.1_003 - 9 Tricky 
> point with default work and refSystem
> 
> 
> > Todd,
> > 
> > I think I see your point. The osisIDs such as 
> Bible.KJV:Matt.1.1 have 
> > the reference system identified anyway, and IDs without the
> > osisWork: part use a default value, so what's the point of 
> saying that 
> > a work uses reference system Bible.KJV?
> > 
> > It seems to me that a document could have a number of different 
> > relationships with a reference system:
> > 
> >   - define it (if your earlier proposal is accepted)
> This seems like a good idea.  The only trouble is that we 
> have not hashed out what is required to define a reference system.
> >   - implement it entirely
> 
> 
> This is an interesting distinction between declaring that a 
> reference system is the default and declaring that a document 
> fully supports a reference system.  (Does fully support mean 
> that all identifiers in the reference system OR that all 
> identified elements are identified using the reference system 
> where possible?)  Also it would seem that the default 
> reference system need not be fully supported and that more 
> than one reference system could be fully supported.  For this 
> behavior it seems that we need add an 
> "entirellyImplementedRefSystems" attribute that is a list or 
> refSystem IDREFs to <osisText>.
> 
> >   - use the occasional osisID from it
> > 
> 
> 
> 
> > It might be worth knowing that a document entirely implements a 
> > reference system. Would that be worth specifying?
> 
> It might be worth specifying.  In the future the issues of 
> mapping will be hashed out.  If a document does not implement 
> any reference system then an external mapping to the document 
> would have to be created for at least one reference system.  
> I would think that the external document would basically be a 
> basically an XLink document that matches up XPointer 
> expressions with identifiers from the reference system.  It 
> might also be more meaningful (if possible) to specify the 
> document that provides the mapping between a reference system 
> and the document that does not implement a reference system.  
> I not sure how much of this information needs to be put in 
> the <work> element and how much needs to just exist in the 
> document being pointed to.
> 
> Todd
> > 
> > -Harry
> > 
> > > -----Original Message-----
> > > From: owner-osis-core@bibletechnologieswg.org
> > > [mailto:owner-osis-core@bibletechnologieswg.org] On 
> Behalf Of Todd 
> > > Tillinghast
> > > Sent: Wednesday, August 21, 2002 1:22 PM
> > > To: osis-core@bibletechnologieswg.org
> > > Subject: RE: [osis-core] osisCore_Candiate_1.1_003 - 9 
> Tricky point 
> > > with default work and refSystem
> > >
> > >
> > > I don't have problem with declaring a "preferred" 
> reference system 
> > > for a work.  I am not sure what value is gained by a work 
> having a 
> > > "preferred" reference system though, since all references 
> have to be 
> > > made using a reference system or a reference system and a 
> work and 
> > > not simply a work.
> 
>