Getting ready to join my #ld4 Rare Materials Affinity Group colleagues for A community developed BIBFRAME profile for the description of Rare Books. The session (11am ET) will be live-streamed and recorded on LD4’s YouTube channel! #LinkedData #ld4_2023 #LD4_23 #RareBooks #bibframe

Today's #LD4_2023 sessions start with Yinlin Chen discussing using a #LLM to generate SPARQL queries on DBpedia based on supplied questions. There is some evidence that suitably trained large language models can generate half-decent boilerplate code (though they have attribution and copyright issues), though their accuracy may vary widely. I'll be curious to see how well it works in this case. (SPARQL queries typically include a fair bit of boilerplate). #LD4

In Richard Wallis's #LD4_2023 talk on a linked data discovery & management system for the Singapore national library & archives, I'm particularly impressed by the entity reconciliation process he's discussing. With messy data, you often have many different text strings in different records describing the same entity, so it's vital that those be normalized. He describes some interesting work designing similarity assessments between different strings in the bibliographic & authority data. #LD4

Today's #LD4_2023 sessions are starting now. Here's the schedule: https://2023ld4conferenceonlinkedda.sched.com/ #LD4

Peggy Greisinger discusses the chapter she co-authored on privacy of personal information (specifically of creators in name authority records). Many authority and linked data knowledge bases have lots of personal information, and it's usually not clear whether the people want all that info widely available It's good to have well-established ways to managec and keep track of information on consent for providing that info. #LD4_2023 #LD4 (Right now it's very ad hoc on different platforms.)

Erin Canning's ethics chapter talks about replatforming (taking data in one place and putting it on another). The functional, contextual, and audience changes involved have both potential benefits and potential risks; the chapter discusses three main ethical questions to consider. (If you've seen discussions in the fediverse about visibility of posts, whether to allow search and quote-embed, and when to federate or defederate, I suspect they raise many of the same issues.) #LD4_2023 #LD4

The examples in the linked data ethics chapter by @platypus and @Salo all have to do with maintainers of a relied-upon data service not maintaining it in a timely way (either due to temporary issues, or losing the ability or interest to keep it up). Nearly every project will suspend or end its maintenance at some point, but there are important ethical concerns in how you handle that: #LD4_2023 #LD4