We need to investigate the the structure and content of RISM RDF file before proceeding with reconciliation

[candlecao]

With help of @ahankinson, (Thank you!) we obtained the whole RISM files (2024Nov29) rendered in RDF. It's so enormous, up to 13.03GB, making it not feasible to be displayed with VSC. However, it can still be displayed in textEditor.

It's in n-triple format, right? As far as I know, there shouldn't be name space prefix abbreviation in n-triple file.
So I suggest to segment the file evenly then upload the separated files to Virtuoso, using Virtuoso's BulkLoader toolkit.

While scanning the rows through textEditor, I noticed some blank rows intermittently appear. Will that interfere with the segmentation process?

[ahankinson]

We can load the whole thing without problems in Qlever, so I would be surprised if Virtuoso didn’t let you load it.

[ahankinson]

It’s a good question about blank nodes though. If I get a moment I will investigate.

[candlecao]

Qlever

Thank you for letting me know that there is a "QLever" (https://github.com/ad-freiburg/QLever). Does your company use Qlever for managing LinkedData? As it says from the upper gitHub page: "QLever is fast for queries that involve large intermediate or final results, which are notoriously hard for engines like Blazegraph or Virtuoso." We have to segment the files before uploading them to Virtuoso with its toolkit BulkLoader.