1 Topic by DaveSlade

  • DaveSlade
  • Veteran
  • Offline
  • Registered: 15-05-2006
  • Posts: 332

Topic: Darwin Core Record ID

With the old Recorder Model for data sharing, the Taxon Occurrence Key was a universal unique identifier that allowed everyone to keep track of where records came from and went. I presume that there is no direct relationship between between this and the new RecordID that Darwin Core uses, but is the intention for the new RecordID to be a universal constant for a record on Atlas? Or will it change every time a dataset is updated?

Essentially I want to match this key with records previously (and legitimately) downloaded from Gateway and keep track of them so that I can report of data usage. I don't really want to go through that process every time I get a download of open-licensed data.

Dave

Dave Slade
Senior IT & Records Officer,
15 Talbot Road, Talbot Green, Pontyclun, CF72 8AD
www.sewbrec.org.uk, www.sewbrecord.org.uk

2 Reply by DaveSlade

  • DaveSlade
  • Veteran
  • Offline
  • Registered: 15-05-2006
  • Posts: 332

Re: Darwin Core Record ID

It would also be useful to know if any of the Record ID related to the data provider in the same way that the old taxon occurrence keys did, so that I can decide which 36 characters to push into my 16 character hole (and not have to re-design all my SQL Server databases). First 16? Last 16? Or should I just create a whole new TOK based on my own formula?
Dave

Dave Slade
Senior IT & Records Officer,
15 Talbot Road, Talbot Green, Pontyclun, CF72 8AD
www.sewbrec.org.uk, www.sewbrecord.org.uk

3 Reply by charliebarnes

  • charliebarnes
  • Veteran
  • Offline
  • From: Greater Lincolnshire
  • Registered: 07-09-2008
  • Posts: 1,385

Re: Darwin Core Record ID

AFAIK, RecordIDs are (genuine, unlike taxon occurrence keys) universally unique identifiers (UUIDs) - any attempt to shorten them (the 36 characters) may result in a duplicate value.

The data provider would have to use UUIDs as their internal record key (and send them to the NBN Atlas) for them to be persistent across updates on the Atlas, otherwise I expect they are created at import (in fact, they may just be created at import looking at the metadata for a download?). I.e. if it's from e.g. a Recorder 6 database then the answer is no, it will change every time the dataset is updated without some extra work.

If you're after something a bit more useful, some data providers fill in the record _ number field (or the NBN team map this to it) - this contains, among other things, taxon occurrence keys from Recorder 6.

Charlie Barnes
Information Officer
Greater Lincolnshire Nature Partnership

charliebarnes's Website

4 Reply by DaveSlade

  • DaveSlade
  • Veteran
  • Offline
  • Registered: 15-05-2006
  • Posts: 332

Re: Darwin Core Record ID

Thanks Charlie - that is a shame as the Record_number field does not come out in the download, unless there is a way of customising this that I have not yet found. Looks like I'll either have to reconfigure my databases or create my own taxon occurrence style keys.
Dave

Dave Slade
Senior IT & Records Officer,
15 Talbot Road, Talbot Green, Pontyclun, CF72 8AD
www.sewbrec.org.uk, www.sewbrecord.org.uk

5 Reply by charliebarnes

  • charliebarnes
  • Veteran
  • Offline
  • From: Greater Lincolnshire
  • Registered: 07-09-2008
  • Posts: 1,385

Re: Darwin Core Record ID

It does in mine - I wonder if it's being omitted because it's empty?

I think I hit download on this page to test it: https://records.nbnatlas.org/occurrence … 0000530481

Charlie Barnes
Information Officer
Greater Lincolnshire Nature Partnership

charliebarnes's Website