Regarding the LCCN requirement, you can get around that for more "local" institutional holdings by creating your own MarcXML file for the series you have and placing it in a folder that is set in the settings.py file using the BIB_STORAGE variable (line 127
in this file:
and then editing the batch_loader.py file to point to those files instead of the loc website.
In the current version of batch_loader.py that is on line 260 (https://github.com/LibraryOfCongress/chronam/blob/master/core/batch_loader.py
You would need to make sure you name the files something consistent so it can find it using the lccn variable. For example, if your fake "lccn" number for your local holdings is 'testlccl' then name the file "testlccn.xml", set the lccn in your marc and alto
metadata files to be "testlccn". You would alter the line (line 260) to then point to the BIB_STORAGE directory instead like this:
management.call_command('load_titles', BIB_STORAGE + '%s.xml' % lccn)
Something like that. Hope that gets you started.
First of all, thank you very much for the suggestions last week--we're up and running with an evaluation instance of chronam using only the built-in TIFF support. It's been really great to work with.
The first question that has arisen from our tests involves technical metadata; does anyone know what the minimal amount of technical metadata required is (for the software, not NDNP guidelines)? We've been experimenting with creating our own METS-ALTO
upload batches; so far they're working, but the software seems to need image dimensions from the technical metadata in order to build parts of the interface. I plan to experiment with adding this information but wondered if I might run into other 'gotchas.'
Additionally, it seems the software very much depends on the presence of an LCCN. We're working with several papers (some would be described as "newsletters," some student papers) that don't currently have LCCNs. Otherwise, they fit the profile of things
that could be browsed/managed the same way. We're investigating getting LCCNs assigned to them, but we're not sure what that will involve yet. We were thinking for these items we might try to adapt the software to query some sort of local database of descriptive
data instead of the LC databases that use LCCNs as identifiers--does this sound insane?
Thanks very very much!