I will contribute one particularly heartbreaking bit from my own
current metadata saga - I'm in one of these hybrid museum/research
library institutions where the library side has a aging MARC catalog
with its own issues that I won't go into at the moment. The museum
side has a commercial collection management database that recently
changed names from ReDiscovery to Proficio. The good news about this
database is that after some digging I uncovered an export method that
is fairly free-form and allows me to write a template to export
directly to MODS xml which is my intended middle ground between
library and museum (the only trick is getting your hands on the Top
Sekrit database field names). The bad - actually painful news was
discovering how data that had been painstakingly entered by hand over
15 years into separated fields was being munged together as free text
within the database. Nobody knew this was happening until I started
trying to export data. So, for example, a name and its associated role
and dates would have been entered into appropriate separate authority
controlled fields in a data-entry form but then would be stuffed into
a single field in the database. The only consolation is that they do
stuff in some text delimiters that are (mostly) uncommon characters
(pipes and underscores) so it is possible to break the fields back
out, just very time consuming and prone to introducing errors.
Lesson learned: vigorously test how well the data comes out of any
system before investing any time putting data into it. Also invest in
time travel to go back and apply this lesson at the beginning...
-Derek
@dmer

On Wed, Jan 25, 2012 at 10:06 AM, Becky Yoose <[log in to unmask]> wrote:
> Hi all,
>
> For our preconference, “Digging into Metadata,” we’d like to get a little
> discussion going to build on once the preconference rolls around.
>
> A good part of our discussion will focus on metadata issues and how folks
> have worked through said issues or have utilized metadata in an unique way
> while keeping the metadata’s context in mind. Some example include:
>
> - Dirty data issues when switching discovery layers or using legacy/vendor
> metadata (ex. HathiTrust)
> - Dealing with free text in MARC records and how to parse them w/o too much
> heartache
> - batch creating and editing metadata
>
> Some of you have already touched on this in the last preconference email
> thread, but we'd like to get some more examples to focus on. What are your
> metadata war stories?
>
> Thanks,
> Becky
>
> ---------------------
> Becky Yoose
> Systems Librarian
> Grinnell College