[osis-core] Proposal on the type="OSIS" proposal
Patrick Durusau
osis-core@bibletechnologieswg.org
Wed, 29 Oct 2003 08:17:47 -0500
Guys,
Here is my suggestion for a resolution to this issue:
<identifier osisTextID="true/false" type="enumerated in manual/defined
in a work element/private">string_that_is_the_identifier</identifier>
Note that moves the type="OSIS" into a new attribute, since that is
different (IMHO) from ISBN, LCCN, etc. It also avoids having to be
inconsistent and sometimes having the prefix when I want to say
type="OSIS and have the other type, for lack of a better name,
associated with the value of the string.
Actually, you could have more than one identifier with osisTextID since
I might want to have multiple osisTextIDs depending upon which type of
system someone wanted to use. Taking Chris's example:
<identifier osisTextID="true" type="LCCN">73173713</identifier>
<identifier osisTextID="true" type="ISBN">3487046881</identifier>
<identifier osisTextID="true" type="LCCall">BR60.A62 L5213 1973</identifier>
Would mean that I could look in three separate reference systems to see
if I have a work in my OSIS catalog that matches one of these identifiers.
Up to you which one your software wants to pick as the "unique"
identifier for the text. Probably should store the others as well, so
that you have a better chance of determining if you have a text when
someone uses a different type of identifier.
Note that omission of the type attribute on identifier will be defined
as = private.
This has the follow impact:
1. Avoids inconsistent use of prefix, that is sometimes use it,
sometimes don't (which Troy wants to avoid at all, inconsistent or not)
2. Separates out a priviledged type (of sorts) from other types
3. Allows texts to declare multiple identifiers of different types for
use in an OSIS catalog
4. Allows documentation of non-enumerated types by use of a work element.
5. Sets for the enumerated strings in the users manual for those who
don't like work elements
6. Type defaults to private if omitted so if all you want is:
<identifier osisTextID="true">blah, blah</identifier> knock yourself out.
7. Your software can choose any or all of these for identifying OSIS
texts. It can even select those without a type or private, if that is
really what you want.
Looks to me like that meets all the various objections and goals.
Hope everyone is having a great day!
Patrick
--
Patrick Durusau
Director of Research and Development
Society of Biblical Literature
Patrick.Durusau@sbl-site.org
Chair, V1 - Text Processing: Office and Publishing Systems Interface
Co-Editor, ISO 13250, Topic Maps -- Reference Model
Topic Maps: Human, not artificial, intelligence at work!