[osis-core] More on pointers and references

[Patrick Durusau]

Harry,

Consider the following illustration using the current proposed syntax:

<work workID="bible.nrsv">
<title>The New Revised Standard Version Bible</title>
<creator role="aut">National Council of Churches of Christ in the United 
States of America</creator>
<date>1989</date>
<rights>copyright held by Division of Christian Education of the 
National Council of Churches of Christ in the United States of 
American</rights>
<coverage>Any reference that is valid for the Standard, Catholic or 
Common Editions</coverage>
</work>

<work workID="bible.nrsv.standard">
<title>The New Revised Standard Version Bible</title>
<creator role="aut">National Council of Churches of Christ in the United 
States of America</creator>
<date>1989</date>
<rights>copyright held by Division of Christian Education of the 
National Council of Churches of Christ in the United States of 
American</rights>
<coverage>Standard Edition</coverage>
</work>

<work workID="bible.nrsv.catholic">
<title>The New Revised Standard Version Bible</title>
<creator role="aut">National Council of Churches of Christ in the United 
States of America</creator>
<date>1989</date>
<rights>copyright held by Division of Christian Education of the 
National Council of Churches of Christ in the United States of 
American</rights>
<coverage>Roman Catholic Edition</coverage>
</work>

<work workID="bible.nrsv.common">
<title>The New Revised Standard Version Bible</title>
<creator role="aut">National Council of Churches of Christ in the United 
States of America</creator>
<date>1989</date>
<rights>copyright held by Division of Christian Education of the 
National Council of Churches of Christ in the United States of 
American</rights>
<coverage>The Common Bible</coverage>
</work>

(Note  that the National Council of Churches of Christ indicates that 
the work is known as NRSV and has different "ecumennical formats" (their 
wording, not mine)

> * a standard edition with or without the Apocrypha, a /Roman Catholic 
> Edition/, which has the so-called "Apocryphal" or "Deuterocanonical" 
> books in the Roman Catholic canonical order, and /The Common Bible/, 
> which includes all books that belong to the Protestant, Roman 
> Catholic, and Orthodox canons.

(Note that <rights> should be added to the current <work> element. 
Entered as corrected here.)

<work workID="bible.nrsv.catholic.oxford">
<title>The New Oxford Annotated Bible</title>
<creator role="aut">National Council of Churches of Christ in the United 
States of America</creator>
<publisher>Oxford University Press</publisher>
<rights>copyright held by Oxford University Press</rights>
<identifier type="ISBN">0-19-528356-2 </identifier>
<coverage>Roman Catholic Edition</coverage>
</work>


The first four, bible.nrsv, bible.nrsv.standard, bible.nrsv.catholic, 
bible.nrsv.common, identify reference systems and a particular 
translation. These are system identifiers that refer to works beyond 
OSIS to define a reference system.

For texts that do not differ in terms of the reference system, I think 
that these would suffice for Harry's osisIDScheme. They do also identify 
a particular translation, but it shares a reference system with other 
translations.

The last example, bible.nrsv.catholic.oxford, is different because it 
identifies a particular edition, and would join with Harry's 
osisIDElement to form a unique osisID (using Harry's terminology below).

>5.  osisIDElement -- like a name token, but can start with a digit. Must be
>unique within an osisIDScheme. Pagebreak milestones (if present) are
>conventionally identified as Page_32, Page_xii, etc. [previously called an
>osisID]
>
>6.  osisID -- [osisIDScheme:]osisIDElement. If osisIDScheme is omitted, a
>default value specified in an <osisText> attribute is assumed.  Thus, an
>osisID says "this element contains the osisIDElement part of osisIDScheme".
>
 
What I am saying quite poorly is that the reference system used by a 
work and identifying the work to satisfy the "who am I" function of the 
osisID, seems to me to be different tasks. All the reference system need 
do is identify the reference system, but an element should not be able 
to say "I am any Mark 5:1"  without reference to a particular version, 
edition, etc., would not be very meaningful. On the other hand, a 
pointer could say, "any Mark 5:1 in the bible.nrsv reference system", 
assuming it used bible.nrsv as its osisIDSchema.

In other cases, such as using bible.nrsv.catholic.oxford for the 
osisIDScheme, then Mark5:1 is a meaningful reference for the 
osisIDElement, since it satisfies the "who am I" requirement for the 
element. The same osisIDScheme (bible.nrsv.catholic.oxford) would also 
be valid for a pointer, although it would only point to Mark 5:1 in that 
particular edition.

In other words, osisElementID should always refer to a particular 
instance of a text (Mark 5.1 from a particular translation, edition, 
etc.) while a pointer may, but does not have to point to a particular 
translation or edition?

Note that the proposed workIDs (or whatever their names, I am just 
trying to be consistent with Harry's post) would be declared inside the 
OSIS document and I am suggesting that we might want to prepare a set of 
such declarations for people to use with OSIS documents. They would then 
pick one of these to form the osisIDScheme for the osisID (as defined in 
Harry's post). Pointers/references could use any of the declared workIDs 
to form a complete osisID.

(We could also declare these within the OSIS schema and give them values 
as fixed as well. Might not be a bad idea for the more common ones.)


Patrick




-- 
Patrick Durusau
Director of Research and Development
Society of Biblical Literature
pdurusau@emory.edu