Rebooting ZLOD Project


September 11th, 2022 | Topic: Librarianship, Metadata, Uncategorized, Zen | Author: | no responses |


The ZLOD (Zen Lineage/Linked Open Data) project is something I first began as a final project in a linked data course I took during grad school. It is something I wanted to finish, and keep revisiting, but find that it might make more sense to do a full reboot.

My desire initially was to solely run SPARQL queries against Wikidata. There is a lot of great information there, but cleaning the data for my needs is not working out as I had hoped. What I am going to do now is restart from scratch, and hand build a test lineage chart in JSON-LD. This will help me determine what the application profile needs (i.e., will schema.org be enough or will I need more). It will also give me an opportunity to experiment visualizing solid data in the form that I will need.

I will try to keep posting updates here as I make progress with the project. Sketched out, it so far looks like Schema.org can handle all of my needs:

…and where it makes sense to, I can then link to Wikidata URIs from objects in the triples, such as Person (subject) -> affiliation (predicate) -> soto zen (object) then soto zen (subject) -> URL (predicate) -> Wikidata URI (object)

Fight the system!(?): Inclusivity, discoverability, and straying from authorities in the library catalog


December 16th, 2021 | Topic: Librarianship, Metadata | Author: | no responses |


Libraries are more complex than many realize. This is why discussions about inclusivity, discoverability, and critical librarianship are so important for the realm of the library catalog. Rather than build upon the backbone of antiquated systems already in place, we need to move toward a more agile system. Our resource description methodologies and tools must be able to respond more efficiently with social change that they are in the current state of things.

User experience

Let’s focus on a library catalog’s user experience impact in two primary areas: discoverability and inclusivity (both of which overlap in many ways).

Whether from aging authority files, the accumulation of poor and/or inconsistent key terms, or incomplete records, the negative impact on discoverability rises as metadata quality declines. Ross mentions that a list of 180 “outdated and offensive” terms has been compiled through the African American Subject Funnel Project in an effort to improve inclusivity in the Library of Congress Subject Headings (LCSH) (American Libraries, 2021). That is a long list that would not only improve inclusivity, but also impact discoverability. Inclusive language ensures the catalog accurately and fairly represents the diversity of a library’s user base without “othering” them in ways that would exclude or discourage them from library services. Additionally, proper term use is crucial for discoverability, and changes to make language more inclusive will modernize vocabulary in ways that will improve discoverability.  

Description of resources

Hobart shares how Angelou’s I Know Why the Caged Bird Sings lacks terms that would make it discoverable via a library catalog when searching for race or racism; the topic has to be inferred from the description, which such a catalog search cannot generally do (American Libraries, 2021). The description of resources does more than just explain what the information object is once discovered — it ensures that it can be found based on need.

Authority control & access points

So why not just fix it and be done with it? Well, with this talk of updating catalogs, it is important to recognize that authority files can tie the cataloger’s hands in ways that stunt both discoverability and inclusivity. Yet, authority files are very important for access points. The general effectiveness of a catalog’s search capabilities depends on them. 

A classic example of the importance of authority control is how a user can search for either George Sand or the author’s real name of Amandine Aurore Lucie Dupin and still find works by the author — which is possible in catalogs that follow authority files, such as the Library of Congress Name Authority File (Library of Congress, 2005). Granted, names are quite straightforward when it comes to access points. The trouble creeps in when we start looking at authority control that requires interpretation, such as subject headings. 

Deviation from authority control can have a big impact, especially when catalogs are using authority control that has been in use since the 19th century such as LCSH, which is interconnected in ways that make changes quite laborious (American Libraries, 2021). 

One commonly cited issue and ‘resolution’ is the change of LCSH Illegal Aliens to Noncitizens / Undocumented Immigration. This change to the authority file took over two years, after being first denied, and LC still did not accept the proposal of Undocumented Immigrants that ALA’s governing body was urging for (Freedman, 2016). It is still a good change, but makes LCSH feel sluggish and inflexible. 

Question

My question then becomes: Should catalogers continue to strictly adhere to LCSH even though change has been very slow to take root? 

Granted, this is not an all or nothing question. Consider high-level contributors to WorldCat such as Jenna Freedman who has over 4,000 zine entries that might be copy cataloged by others (2016, Freedman). Her approach is to use MARC 653 fields to incorporate custom, uncontrolled index terms. Otherwise, LCSH will not cover the topic of her zines well enough. The negative to this? As she points out, if LCSH is updated it will be updated on her zines by others through typical workflows. Her 653 entries will be there forever unless manually discovered and fixed. It is likely that such considerations are why authorities such as OCLC instruct catalogers to prefer the use of subject headings from established subject schemes (OCLC, 2020).  

In another example, the William Andrews Clark Memorial Library Statement on Cataloging acknowledges that long-standing professional biases and issues stemming from national and internationally standardized metadata exist, and that they seek suggestions and will fix problems (Schneider et al., 2020). This method of crowdsourcing fixes for inclusive language issues and other errors, while making a concerted effort internally, could be quite effective. Does deviating from LCSH cause more harm than good? 

A non-answer

In my opinion, it is time to seriously think about moving toward a new way of doing things. In one example, advances toward a new system such as BIBFRAME could remove the onerous authority file change / MARC record update that make some hesitant to update terms. BIBFRAME’s use of unique identifiers that will never change, even if the subject heading, etc. that it relates to does, can make for a nimble cataloging system (Librarianship Studies & Information Technology, 2021). Removing the MARC update cascade effect that results from authority control changes is long overdue. Not saying that this is the only thing that makes changes to authority files so sluggish, but I think it could help.

I see this as an important step for the future of libraries. Perhaps not “the answer,” but certainly a step in the right direction.

References

American Libraries. (2021, November 1). Decolonizing the catalog: RUSA webinar explores avenues for antiracist description. https://americanlibrariesmagazine.org/2021/11/01/decolonizing-the-catalog/

Freedman, J. (2016, March 27). Can I quit you, LC? Lower east side librarian. https://lowereastsidelibrarian.info/lcsh/quityou

Librarianship Studies & Information Technology. (2021, April 23). BIBFRAME (bibliographic framework). https://www.librarianshipstudies.com/2017/12/bibframe.html

Library of Congress. (2005, October 28). What is a MARC record, and why is it important? https://www.loc.gov/marc/uma/pt1-7.html#pt5

OCLC. (2020, August 29). 6xx fields. OCLC Support and Training. https://www.oclc.org/bibformats/en/6xx.html   

Schneider, N. M., Fenning Marschall, R., Sanchez-Nunez, A., & Riley, M. (2020, June 1). William Andrews Clark Memorial Library statement on cataloging. UCLA William Andrews Clark Memorial Library. https://clarklibrary.ucla.edu/research/statementoncataloging/ 

Fact Sheet: Dublin Core Metadata Initiative


November 28th, 2021 | Topic: Librarianship, Metadata | Author: | no responses |


This is a fact sheet that explores the Dublin Core Metadata Initiative (DCMI). DCMI’s work is in wide use on a global scale. It is simple to employ and easy to interpret, yet flexible and fully capable of becoming part of metadata application profiles that meet a wide range of use cases.