Matt Stephenson expressed the dilemna:
> As someone who has recently been introduced to the vast and varied wonders
> of ISAD (G) I am in a quandary regarding the coding of the elements.
>
> Some colleagues inform me that the codes should be listed as 3.1.1, 3.1.2,
> 3.1.3 etc and others inform me that it should be 1.1, 1.2, 1.3 etc.
Although I can't find anything which states this explicitly in the
current version of the ISAD(G) standard, I have been advised in the
past by some of the Higher Authorities on whom you call that,
strictly speaking, the numbers used in ISAD(G) should be used
*only* to cite that document, and *not* as a shorthand for
designating elements of description.
I have to confess that this is not a rule which I have adhered to
myself at all times! In the EAD sphere (and, I imagine, in other
automated processing environments too), for example, it is all too
tempting to use something like:
<unitid encodinganalog="isadg311"....>
or
<unitid encodinganalog="isadg11"....>
But I suspect that the authors of the standard would argue that
there is absolutely no guarantee that any future version of the
standard will preserve the association between the number 3.1.1
(or 1.1) and the semantics of the "reference code" descriptive
element. This contrasts with, say, MARC field numbers (as I
understand their use), where new versions of the standard may add
new fields, but won't change the semantics associated with
existing numbers (or at least not fundamentally).
So really, if I choose to employ something like the above in my
EAD documents, I have to accept that the meaning will only be
interpreted consistently by my own local processing application,
where I can insist that "isadg311" will always carry the semantic
association of "reference code", in which case this specificity
should perhaps be made explicit as something like:
<unitid encodinganalog="glaisadg311"....>
or
<unitid encodinganalog="glaisadg11"....>
and I define explicitly (in a separate document) the semantics I
wish to associate with my "glaisadgxxx" values, or state explicitly
their correspondences to the elements of a specific version of
ISAD(G).
Just my interpretation of things....
Pete
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|