Closed campmlc closed 1 year ago
Your URL and your example do not align.
In general, I'm not convinced that most users understand the structure of ARKs in sufficient detail to deal with partial identifiers, and I can see no reason that we'd want to type them, deal in partial identifiers, or add unnecessary complexity in any way. Copy the whole thing, paste it into https://github.com/ArctosDB/arctos/issues/6005, and be done.
The same is true of all resolvable identifiers, and much of the recent cleanup is a direct result of users not understanding our mostly-unnecessary complexity and being forced to deal with partial identifiers that cannot DO stuff. This of course does not stop at the boundaries of Arctos: https://github.com/ArctosDB/arctos/discussions/5310
Issuers are much less important on resolvable IDs (or I haven't seen the light yet, maybe) - just like they type themselves (by doing stuff), you'll have a pretty good idea of who issued them when you get to where they go. These however are not ambiguous: USNM isn't going to let anyone other than USNM issue USNM identifiers.
There is in fact an ARK ID type, but I wouldn't suggest anyone use it and I hope it has a very short life expectancy: https://arctos.database.museum/info/ctDocumentation.cfm?table=CTCOLL_OTHER_ID_TYPE#ark
I just found the ARK id in use in Arctos for exactly this purpose, so someone beat me to to it: https://arctos.database.museum/guid/MSB:Para:28915
I can't imagine why we wouldn't want to use this if it is available to link to USNM records. Maybe it isn't perfect, but it is better than linking to a aggregator - this is the primary data source. And until recently they had no way at all to link to a record. ????
And apologies if I didn't submit in the correct format. This is not my area of expertise. That is why we have this forum. But I stand by my question - we have a resolvable identifier in use by USNM. I do not understand why we should reject linking to it.
can't imagine why we wouldn't want to use this
That is not what I said. Absolutely, use it. It's a great identifier, and good on them for having it.
not my area of expertise.
THAT is my point! You (and every user to come after) can
The first has you dealing with 65665/3777ecb64-7edc-4479-8486-a0b584092bd0 (is that the right thing? Are you sure? How would you know?).
The second has you dealing with http://n2t.net/ark:/65665/3777ecb64-7edc-4479-8486-a0b584092bd0 - your browser (and a bunch of other tools, including github as soon as I submit this) will happily tell you if that's the right thing by DOING STUFF.
IF we can find a way to get through some small changes, we can make some otherwise-complex thing usable by most anyone, expert or not.
I definitely did not intend for the original ask to require leaving off the http:// etc prefix. I'm not sure if that is how the current ARK is formatted. I am actually very happy to copy paste the full url. But I think we have solved this issue with using other identifier and issued by the Smithsonian Division of XXX, with the http// ark address as the url? https://github.com/ArctosDB/arctos/issues/5995 Closing
Just found arkIDs in the USNM database that actually link to a record. I would like to be able to create a relationship to this arkID via Arctos, but arkID is not yet in the identifiers code table. Since this is specific to an individual catalog record, it cannot be added to an agent, unless we want to create agents for every USNM record . . .
proposed identifier type: ark url: https://arks.org/ definiton: ?Archival Resource Keys (ARKs) serve as persistent identifiers, or stable, trusted references for information objects. See for example: http://n2t.net/ark:/65665/32dc62b24-383f-4466-b0ff-427b6eb6444f
Not to be confused with "ark Ids" used for creatures created in some gaming platforms . . . ?