Jeffrey posted:

>1. Cataloging visual materials proves to be no more difficult using
>MARC21 than some xml markup language.

That is our experience as well.  Despite MARC21's problems (coding
state university press publications as government documents, the
nonsensical order of notes in 5XX, too many fixed fields which differ
from genre to genre) we find that MARC functions well.  We find MARC,
like AACR2, functions as well for nonprint materials as well as for
print materials.  There is also the matter of legacy records as you
point out.

There is MARc'S great advantage of the language neutrality of numbered
fields, and the difficulty of expressing element distinctions in
verbal labels, e.g., 130 vs. 240.

>2.  The underlying costs of conversion from MARC21 to something else
>will impede the majority of libraries from undertaking this move.

There are MARC to Bibframe crosswalks.  SLC could probably adapt one
and could do the crosswalks for about $.10 per record.  But there are
free crosswalks online aren't there?  Whether smaller libraries would
have the IT skills to use them is another question.

>Can anyone tell me what my cost is and what my returns are going to

I suspect your major cost would be for an ILS which can utilize
Bibframe.  Even current "free" ILS have costs associated with them.

Sorry, I see no return for local library patrons.

>3.  The catalog is something at the local level.  We have thousands
>of records that are suppressed on purpose to guard materials from
>being borrowed ...

This is your first comment with which I have difficulty.  Why have a
resource which can not be located?  Rare and delicate materials would
be in a secure location, and perhaps only used on site.  But hiding
their existence seems strange to me.
My hesitancy over linked data differs from yours.  Here, whenever a
logging truck hits a power pole, we suffer the loss of Internet
connectivity; we wonder at the wisdom of having ones local catalogue
dependent on connectivity.  Some answer that data would be stored
locally.  If so, how would it be kept current with the Internet
version?  What about differences in form of entries, such as between
LC, LAC, VIAF, etc.?  I assume local information (such as notation of
an autograph) would be part of a local holdings record, as would
whether an item may circulate or not, be loaned on ILL or not.

The only reason I see to suppress a record is for a lost item which
may reappear or be replaced, and thus the record would be needed

>I would rather see some time spent on further development of MARC21

Agreed.  But even more needed is attention to ILS development.  We
have not fully utilized the the riches contained in our MARC records.  
We are fiddling with the building blocks when we should be paying
attention to the structure.

I wonder if one of SLC's niche markets in the future will be
crosswalks from Bibframe to MARC for ILS still based on MARC?  We have
a major investment in MARC based systems.

   __       __   J. McRee (Mac) Elrod ([log in to unmask])
  {__  |   /     Special Libraries Cataloguing   HTTP://
  ___} |__ \__________________________________________________________