Having only a single external link field isn't adequate

[zygoballus]

Right now, the BCDM schema only has a single field, specimen_linkout, for linking to external resources related to the specimen record. For most of my records, I need at least two external link fields, and sometimes more. Almost all of my records need external links for observation record (e.g. iNaturalist URL) and museum record (URL in museum database). It would also be nice to be able to link to corresponding specimen records in GBIF, MorphoBank, and specialized databases like SCAN and GloBI.

[DNAdiversity]

this could potentially be addressed by have this be an array field to support multiple links. However, the field is primarily meant to link to the source record.

[rdmpage]

I guess part of the problem is that there may be multiple links for the "same" specimen. A specimen may have a record in GBIF (may in fact have multiple GBIF links), may also have an institutional URL (e.g., a "persistent URL" or an ARK), and, as DiSSCo takes off, there may also be DOIs. Perhaps either have as m any links as user wants, or say that field is for the "definitive" link and let someone else handle all the "sameAs" links?

[zygoballus]

the field is primarily meant to link to the source record.

I'm not sure what you mean by "the source record"? All of the links I want to make are to records for the DNA source, i.e. the specimen. Can you elaborate? Ideally I would like to be able to create as many links as needed. An array field sounds like it might be a good solution for this.

[DNAdiversity]

Happy to clarify, I meant the leader record.  For example, if the barcode is associated with a museum specimen that happens to be digitized and placed online by the museum. If you’re looking to enter multiple URLs, please provide the urls in comma separated values format. Cheers Sujeevan Sujeevan Ratnasingham | Director, IT & Informatics | Adjunct Professor Centre for Biodiversity Genomics | University of Guelph CBG Room 207 | 50 Stone Road E | Guelph, Ontario | N1G2W1 519-824-4120 Ext. 53831 ***@***.*** http://boldsystems.org | http://mbrave.net | http://lifescanner.net

…

On Jun 13, 2025 at 3:02 PM -0400, zygoballus ***@***.***>, wrote: zygoballus left a comment (DNAdiversity/BCDM#3) > the field is primarily meant to link to the source record. I'm not sure what you mean by "the source record"? All of the links I want to make are to records for the DNA source, i.e. the specimen. Can you elaborate? Ideally I would like to be able to create as many links as needed. An array field sounds like it might be a good solution for this. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: ***@***.***>

[zygoballus]

Using commas to separate URLs is problematic. Commas are allowed characters in URLs, so this creates confusion about whether commas at the end of URLs are supposed to be part of the URL or not. Also, it's just bad data hygiene to concatenate multiple pieces of data into one field. It's much cleaner to handle these as an array, both for input and output. For example, if I'm retrieving the specimen record from the BOLD API, I don't want to have to implement a special parser just for parsing out the URLs. It would be much cleaner if they were stored and returned as an array.

Alternatively, you could implement different link types as different fields. For example, you could have separate fields for "Museum record link", "Field record link", "Lab record link", etc.

[DNAdiversity]

Use quotes as per CSV spec for strings. Sujeevan Ratnasingham | Director, IT & Informatics | Adjunct Professor Centre for Biodiversity Genomics | University of Guelph CBG Room 207 | 50 Stone Road E | Guelph, Ontario | N1G2W1 519-824-4120 Ext. 53831 ***@***.*** http://boldsystems.org | http://mbrave.net | http://lifescanner.net

…

On Jun 13, 2025 at 5:48 PM -0400, zygoballus ***@***.***>, wrote: zygoballus left a comment (DNAdiversity/BCDM#3) Using commas to separate URLs is problematic. Commas are allowed characters in URLs, so this creates confusion about whether commas at the end of URLs are supposed to be part of the URL or not. Also, it's just bad data hygiene to concatenate multiple pieces of data into one field. It's much cleaner to handle these as an array, both for input and output. For example, if I'm retrieving the specimen record from the BOLD API, I don't want to have to implement a special parser just for parsing out the URLs. It would be much cleaner if they were stored and returned as an array. Alternatively, you could implement different link types as different fields. For example, you could have separate fields for "Museum record link", "Field record link", "Lab record link", etc. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: ***@***.***>

[DNAdiversity]

CSV format typically uses quotation marks to enclose strings.  For example: “https://www.mysite.net”,”https://www.myothersite.net”.  Moreover, while it is technically possible to include commas in URLs, this is not a widely accepted practice. Sujeevan Ratnasingham | Director, IT & Informatics | Adjunct Professor Centre for Biodiversity Genomics | University of Guelph CBG Room 207 | 50 Stone Road E | Guelph, Ontario | N1G2W1 519-824-4120 Ext. 53831 ***@***.*** http://boldsystems.org | http://mbrave.net | http://lifescanner.net

…

On Jun 13, 2025 at 8:03 PM -0400, Sujeevan Ratnasingham ***@***.***>, wrote: Use quotes as per CSV spec for strings. Sujeevan Ratnasingham | Director, IT & Informatics | Adjunct Professor Centre for Biodiversity Genomics | University of Guelph CBG Room 207 | 50 Stone Road E | Guelph, Ontario | N1G2W1 519-824-4120 Ext. 53831 ***@***.*** http://boldsystems.org | http://mbrave.net | http://lifescanner.net On Jun 13, 2025 at 5:48 PM -0400, zygoballus ***@***.***>, wrote: > zygoballus left a comment (DNAdiversity/BCDM#3) > Using commas to separate URLs is problematic. Commas are allowed characters in URLs, so this creates confusion about whether commas at the end of URLs are supposed to be part of the URL or not. Also, it's just bad data hygiene to concatenate multiple pieces of data into one field. It's much cleaner to handle these as an array, both for input and output. For example, if I'm retrieving the specimen record from the BOLD API, I don't want to have to implement a special parser just for parsing out the URLs. It would be much cleaner if they were stored and returned as an array. > Alternatively, you could implement different link types as different fields. For example, you could have separate fields for "Museum record link", "Field record link", "Lab record link", etc. > — > Reply to this email directly, view it on GitHub, or unsubscribe. > You are receiving this because you commented.Message ID: ***@***.***>

[kaldari]

I agree that it would be helpful to support multiple link fields. Perhaps this could be handled in the same way as associated specimens and associated taxa.

[DNAdiversity]

That’s a good recommendation, a delimiter would allow multiple links. Sujeevan Ratnasingham | Director, IT & Informatics | Adjunct Professor Centre for Biodiversity Genomics | University of Guelph CBG Room 207 | 50 Stone Road E | Guelph, Ontario | N1G2W1 519-824-4120 Ext. 53831 ***@***.*** http://boldsystems.org | http://mbrave.net | http://lifescanner.net On Sep 12, 2025 at 12:11 PM -0400, kaldari ***@***.***>, wrote: kaldari left a comment (DNAdiversity/BCDM#3) I agree that it would be helpful to support multiple link fields. Perhaps this could be handled in the same way as associated specimens and associated taxa. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: ***@***.***>