mrshll1001
commented
1 year ago Hi @pleasantone, thanks for this and apologies for not getting back to you sooner.
We've done some exploratory thinking around JSON-LD because you're correct, it's a very powerful technology that would make publishing and scraping easier for some organisations if they've already got their infrastructure set up. The potential of having 360Giving data both distributed and integrated into the LinkedData ecosystem is also quite exciting.
At the moment, the 360Giving publisher base is a mixture of small organisations and large ones, with lots of heterogenous software systems. Some of the people responsible for collating and publishing the grants data would likely be described by most as "non-technical", and are doing the work as part of an administrative role and using technologies familiar to them ie spreadsheets and wordpress installations. This is why the 360Giving tools ecosystem looks like it does now – publishing via a spreadsheet is a very low barrier to entry for a new 360Giving publisher who may be running on an inflexible or basic system. Once their data is up, 360Giving can collate it and convert it to the canonical JSON (or a data user can use tools such as flatten-tool to do this themselves) before sending it to the datastore.
For this reason, at the moment JSON-LD doesn't solve a problem for 360Giving and might actually give rise to a lot of challenges for adoption. From the standard governance level, using definitions from schema.org presents additional challenges. As you note the @MonetaryGrant object is woefully underspecified, and it might take a lot of lift to get schema.org definitions to the point where they're useful for the 360Giving publisher base and compatible with the standard.
What I'd find very interesting to investigate is whether it's reasonable to export JSON-LD as a by-product of some other 360Giving tool, or the collected dataset. The challenges would remain around using JSON-LD properly; what goes into a grant @context, which URLs do we use to identify organisations and the relationships, etc. But this would be a step towards undestanding what 360Giving data would look like as something ready for JSON-LD, and scaffold it for easy integration with other datasets.
pleasantone
Have you folks thought at all about the pros and cons of per-grant website page storing grant information JSON-LD? It would be fairly easy for us to make our grant info machine readable. It might make publishing much easier, especially if a crawler like google does all your indexing already.
schema.org already defines a @MonetaryGrant object, but it is woefully underspecified.
https://shockwave.org/portfolio/tpw/