Strategies for merging the JSONLD of Linked Art

9 views
Skip to first unread message

Shiloh Alleyne

unread,
May 20, 2026, 8:52:35 AMMay 20
to Linked Art
Hi All!

I've been working on a JSONLD compiler/validator in Haskell, as a means of getting myself more familiar with the format of linked art. The compiler works by parsing the JSON of a piece of Linked art into a statically typed Abstract Syntax Tree (AST).  The types of the AST themselves ensure popper JSON construction.

Since we are using a tree based representation for JSON, it is relatively easy to implement a merging operation between the trees. Currently the strategy for merging is to merge on the most nested Context with a highest Similarity Score. If a less nested Context has a significantly (more than 10%) higher similarity score, then it is chosen for merging instead. If there are no Context in common then an array of the separate JSON is returned.

My question is simply, if this a good approach and reflective of how you handle linked art or not. 

Also if you have any particularly complex pieces of linked art JSON I could stress test against then I would appreciate you sharing them with me!

Cheers,

Shiloh

Robert Sanderson

unread,
May 20, 2026, 8:58:46 AMMay 20
to Shiloh Alleyne, Linked Art

Hi Shiloh,

By "Context" do you mean the JSON-LD @context document, to expand from the JSON keys and values into the full RDF graph?  If so, then there is a specific algorithm with many (many!) test cases that you could compare with your approach:  


Otherwise, a simple example of what you're working on would be great :)

Rob



--
You received this message because you are subscribed to the Google Groups "Linked Art" group.
To unsubscribe from this group and stop receiving emails from it, send an email to linked-art+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/linked-art/6c44a426-9eed-40ef-9be0-779c5a970860n%40googlegroups.com.


--
Rob Sanderson
Senior Director for Digital Cultural Heritage
Yale University
Reply all
Reply to author
Forward
0 new messages