BHL
Archive
This is a read-only archive of the BHL Staff Wiki as it appeared on Sept 21, 2018. This archive is searchable using the search box on the left, but the search may be limited in the results it can provide.

TechCall_09jan2017

Dial: 1-877-820-7829
Enter the passcode: 407326

BHL Egypt Work Plan Ideas:
https://docs.google.com/document/d/1-12MJCYcWBPcR-NzBri8Lx60whTMHNYQO1TEkhTpJWo/edit?usp=sharing

Agenda
Introducing Pamela McClanahan, NDSR Resident

BHL Egypt - discuss a discrete package of work they might do for BHL
At Staff Meeting 2015, proposal technical development in lieue of Dues
EC has approved as a concept.
Plan that meets our needs.

Assistance with Metadata App, Creating MARC records. Something more user friendly. Build off CSV.
Build off tools from UIUC, for MARC creation

Bib Alex - used to large scale systems; optimization, data storage.


OCLC and MARCXML
Mike saw Bianca’s notes on wiki regarding OCLC conversations. Turns out OCLC is not able to use our MODS and they want MARCXML. We don’t have MARCXML, we USE it but do not keep it. Would we want to grab MARCXML records that are in IA, and store them someplace to be harvested by people? Note that those will not contain edits or merges that we do in BHL. What do we want to do about this?

MARCXML in admin dash are just the records that are passed at time of digitization. we just need title records in BHL that communicate we have holdings in BHL but not necessary details of edits we've made.
A lot libraries look to OCLC to look at what different libraries hold.
Just need to reference that BHL has that title

Would need to add an 856 or pointer in BHL. For the IA copy.

they've proposed this ability to take MARCXML
Supposed to hear by end of January the status of process

We don't have MARCXML for titles; we have it for items.
We have 1500 titles with no MARC records whatever
The 856 we would add, there's 1/3 of content that didn't come from Partners - is it ok to modify

It seems parsing the MODS would be the easier thing for them to do

Difference between MARC describes print material and one that describes digital material
even if from IA, describes print copy
Description of digital copy can be stripped down

We'll revisit in early February

Year Metadata
Rules that Bianca proposes:
BHL_year_data.pdf

Data used in creating these rules: https://www.dropbox.com/s/vaksez3xc33ht2x/BHL_year_data.xlsx?dl=0. (based on a report that Mike sent on August 18 (http://beta.biodiversitylibrary.org/data/BHLItems20160818.zip).
Bianca analyzed those items (thru Aug 2016), was looking at formats for years. Looked pretty good but definitely some that need to be modified.
PDF of rules outline which formats need to be modified and how. So essentially a script that would find a replace with the formatting
Mike did similar analyses and came up with almost identical rules.

Convention in cataloging for undetermined dates. Things like those would need to be looked at.
190u = 1900-1909

ElasticSearch for now, setting up as 'no date' for faceting

So goal is one-time sweep to clean up
Then put in ingest processes and item editing screen
And in Macaw
when editing an item, four digits
If it's specified in item, it can be 'ruled' but if not specified, would be blank

Could be January

Looks like would impact roughly 14,000
Do we push back to IA and re-ingest?

Does METS file play a part in digital preservation strategy?
Do we need to keep doing this?

Push it to IA, let it happen as normal; It will take longer but not ridiculously longer
Save Mike's time

May want to consider how we want to handle with large changes in general
When we do a mass update, set a flag to ignore the item
Might end up being even more work though
Would be easier to turn off as trigger when we need to

Problem with periods inconsistencies
noticed in lots of fields, like series
Do we want to approach this as a data cleanup project, too

Duplication issues are coming from MARC
stopped tacking c on to titles
before that, more instances of trailing subfields and periods and so on
Not sure if it's worth examining what's coming from original MARC records
If it's an artifact of BHL processing than it should be addressed
not enough info yet

AMNH - Field Notes - Notes field with a period and one without
Macaw was taking bits of MARC, and making description and IA was making descriptions with same, so duplicated

Search and retrieval - should be prioritized

Full Text Search Server
On order at Dell
Joel will check on estimated delivery date