[Application-profiles-ig] MODS-2-RDF example

Mon Nov 2 16:54:11 GMT 2020

Thanks for the comments, Tom and Karen. >>Thanks, John. These are impressive. It's interesting that the "simple"option is mighty big on its own. However, this shows the difference between what are essentially key/value pairs and more structured metadata with arcs that lead to other shapes. I think one reason two options were offered (simple and complex) was that, even though the Fedora 4 repository was all RDF under the hood (LDP was implemented, for instance, although it was also, frustratingly, not a triplestore), my (non-developer) impression was that it was either not possible or not easy to handle anything other than property/value pairs. I'm not sure if this was due to Fedora or the Hydra stack. Some Hydra/Samvera adopters may have had developer resources (and buy-in) to locally adapt the code to accommodate more complex "shape" structures, but many did not. There was a sense of being between a rock and a hard place amongst metadata people. We at U of A were part of the Samvera community at the time, though we aren't now. These are just my personal impressions. Regards, John -- *John Huck*, MLIS Metadata Librarian / University of Alberta Library / he/him/his ᐊᒥᐢᑲᐧᒋᐋᐧᐢᑲᐦᐃᑲᐣ / Amiskwaciwâskahikan / Edmonton *The University of Alberta respectfully acknowledges that we are situated on Treaty 6 territory, traditional lands of First Nations and Métis people.* On Mon, 2 Nov 2020 at 07:26, Thomas Baker <tbaker at tombaker.org> wrote: >On 2020-11-01 03:32, Karen Coyle wrote:>> We really do need a solution to the "OR"ing of values in a cell.>>We have already discussed supporting picklists as a>valueConstraintType, picklists天生about>"OR".>>The more general question is whether cells in other>columns can hold multiple values.>>In Phil's example [1], the two values in J12 could be>defined explicitly as a picklist by adding a>valueConstraintType of IriPicklist.>>I'd argue for using whitespace (not commas) for such>picklists and accepting that the values of such picklists>(whether strings or URIs) cannot themselves contain>whitespace.>>If two values were given under valueNodeType, as in>John's example at D18 [2], we'd HAVE to explain that this>means "OR" because "AND" would make no sense. (For>example: "must be a Literal AND a BNode" makes no sense).>>Besides handling AND and OR with a sensible default for>each column, I would not want us to try to invent>syntactic conventions for indicating "either AND or OR">in a flat spreadsheet format in any machine-parsable way.>We need to emphasize that the spreadsheet format is not>itself a full-blown validation language and cannot be>used to support validation unless it is translated into a>real validation language. The DCAP CSV format, on its>own, is only good for documentation. To be usable for>validation, it must first be translated into a language>such as ShEx.>>Tom>>[1]>https://docs.google.com/spreadsheets/d/1Das7kQRdaQgrDlYwFsIx7xjG1cyVRE5fXnaIHQ4lyZY/edit>[2]>https://docs.google.com/spreadsheets/d/1rI85OxMNLg2C0cd0zDv_tKxJaOIjMfgaUeFCkWZdnF8/edit>>-->Tom Baker <tom at tombaker.org>>-------------- next part -------------- An HTML attachment was scrubbed... URL: <https://lists.www.voudr.com/pipermail/application-profiles-ig/attachments/20201102/5781656f/attachment-0001.htm>