add "familiar name" option

[annamorphism]

I noticed in investigating #704 that the spreadsheet has a "familiar name" field--for some manuscripts this is the provenance ("Salzinnes"), for others it's related to the siglum ("Karlsruhe 60") and for others it's something totally different ( "Hartker", "Riesencodex".) Would it be possible to have these familiar names in CU (maybe in quotation marks?), and use them in some of the places we currently use provenance (e.g. the "Manuscripts" landing page, the banner at the top of the manuscript pages, the tab title text)? Most of these "familiar names" are pretty widely used, while something like "Dominican, McGill MS 073" is not.

[dchiller]

Nice!

My concern is that in doing this we are departing from CantusDB. Does CantusDB have an option like this? I think it would be nicest if it did, and that was just something we pulled from there.

[annamorphism]

that's a fair concern--CantusDB does not currently have this, really. It does tend to include familiar names somewhere, if inconsistently, either in the source information. (e.g. St-Gall 391 has in the summary "Late tenth-century manuscript from the abbey of St. Gall, also known as the Winter volume of the Hartker Antiphoner" so it turns up if people search "Hartker") or in brackets in the shelfmark ("Wiesbaden, Hochschul- und Landesbibliothek RheinMain, 2 (Riesencodex)").

That said, calling things "Rupertsberg, B-DEa 9" is already a departure from CantusDB in its own way--you can refer to it as "St-Pieter-en-Paulusabdij Ms 9" or "the Dendermonde codex" or "B-DEa 9," but just "Rupertsberg" is not how it's referred to in CantusDB (or anywhere, really.) But you're right that since "familiar name" is not a field in CantusDB right now, it would have to be something added from CU uploaders (and what if eventually it needs to be harmonized with some newly-made CantusDB field?)

I guess what bugs me most is that the current provenance+siglum combination looks like a name, but doesn't follow usual naming conventions. Siglum followed by provenance and date (if present), on the other hand, might be another good solution. So instead of "Rupertsberg, B-DEa 9" we get "B-DEa 9 (Rupertsberg, c.1175)." Still tricky to find if you are looking for "the Dendermonde codex," but would work until "familiar name" is something in CantusDB.

[ahankinson]

In DIAMM we simply call this a "name", and it is an optional field on our model: https://github.com/diamm/diamm/blob/master/diamm/models/data/source.py#L52

This is distinct from the siglum + shelfmark. You can get a good idea of how it can be used in the list of MSS from the British Library, where the 'name' is in parentheses, e.g., "GB-Lbl Add. MS 25031 (Worcester fragments)" or "GB-Lbl Add. MS 57950 (Old Hall Manuscript)". Since not all MSS will have a name, it's important to structure your display so that it doesn't look weird if this bit is missing.

I wouldn't necessarily call it "provenance" because it's only that in passing -- to truly track provenance you would need to know what the role of the namesake was: Former owner, or creator, or dedicatee, or copyist, etc. The "Henry VIII" MS is so named because it was commissioned by Henry VIII. The "Anne Boleyn Music Book", however, is named because she owned it. Some have very little to do with people at all, and are instead scholarly abbreviations such as "PadA" (https://www.diamm.ac.uk/sources/505/#/) which is so named to identify it among a set of MSS copied at a monastery in Padua. (https://www.diamm.ac.uk/organizations/20/)

In DIAMM we have a separate model for tracking Provenance: https://github.com/diamm/diamm/blob/master/diamm/models/data/source_provenance.py

This allows you to also associate dates, people, roles, and geographic places to a source.

[ahankinson]

Somewhat related: one of the things that I realized when building DIAMM is that a "common" name for scholars can also be the city where the archive that holds it is located. "Oxford 213" or "Bologna Q15" were my go-to examples for this. These are not tracked as a "familiar name", but since quite a few people actually search DIAMM with these terms, I add the city name of the archive to the search index, specifically to allow sources to be found this way.

[dchiller]

Thanks @annamorphism and @ahankinson!

A few thoughts:

I guess what bugs me most is that the current provenance+siglum combination looks like a name, but doesn't follow usual naming conventions.

Ok, we definitely need to display names according to some wider convention. We could certainly use your suggestion of Siglum (Provenance, Date). The other thing is we have a field called name that we get from CantusDB...you can see them most clearly on the page that lists all of the manuscripts (https://cantus.simssa.ca/manuscripts/) in the small text below the Provenance, Siglum title:

Based on the fact that we name the Provenance, Siglum combination short_name in the code, I suspect that at some point it was thought that this name field was too long. But I'm not really sure that it is? The only three places the short_name is used is in the list of Manuscripts (where we also use name so the concern is not relevant), in the title of the manuscript detail view (there are probably some names that would be too long to display here as we currently do, but we could definitely change how we are styling to make the name fit and look nice), and in the <title> element (controlling what displays in your browser tab, but at least for me the full Provenance, Siglum short name doesn't show fully in the tab anyway, so I think just the naturally truncated name field would work fine here).

So that's another option.

In DIAMM we simply call this a "name", and it is an optional field on our model.

This would be a nice way to do it.

We could add an additional. optional user-generated field for this in Cantus Ultimus -- I am trying to be sticklery in thinking about Cantus Ultimus as not getting out of step with CantusDB (ie. if there is a chant text that needs altering, it should be done in CantusDB and then updated from there in Cantus Ultimus) but if there was a thought of adding this information to CantusDB eventually, I think it would make sense to go ahead and add it now in Cantus Ultimus.

so it turns up if people search "Hartker"

This comment from @annamorphism has me realizing that Cantus Ultimus has no way to search manuscripts...that's not such a big thing when you can see all 17 manuscripts in a browser window, but of course our hope is to surpass that! I think being able to search would be useful and (eventually) critical.

We currently pull the Manuscript description from CantusDB as well, so it's sitting in the Cantus Ultimus database and could be used in this search process, even though it's not currently displayed anywhere in the UI.

[ahankinson]

I am trying to be sticklery in thinking about Cantus Ultimus as not getting out of step with CantusDB

Personally, I think this is a very good idea. I think the "familiar name" should go in Cantus, not CU.