Classical Tagging & the Bryston UI…
Your interest in tackling the problem of how classical music metadata is presented is laudable, as it’s a source of great frustration for serious listeners. There are a number of large problems affecting how classical music is handled… allow me to ramble/rant on a bit.
The system of tags was developed with popular music in mind, and there is no standard way of tagging classical works. Proprietary solutions can work well, as long as you stay within a provider’s ecosystem (e.g. iTunes) and the way the provider has chosen to display/list works suits you, or you manually manage all of your own tags. However, anyone that amasses a large collection of recordings from diverse sources is going to have to deal with missing or erroneous metadata, and matching whatever scheme the user has chosen will require editing the metadata for almost every recording - and classical music has a LOT of metadata.
There are often many different versions of the same work within a collection, potentially hundreds (!), especially when you consider different artists, re-releases, remastering, inclusion in collections etc. The same recording might show up as disc 20 in a large box, as an individual disc in multiple releases, often under different label names etc. How do you differentiate between duplicates (with possibly significant differences)? Some recordings will get paired with different works on different releases too.
Partly due to the lack of tagging standards, individuals have varying requirements as to what information is critical to have in the database. A given work might be associated with a specific conductor, orchestra, recording date (e.g. the same conductor and orchestra recorded it more than once), different soloists, etc. A single disc might have tracks with a different orchestra or conductor for each, with each being a different work, or even the same work in different versions. Individuals have shoe-horned data into tags unrelated to the actual data, merely to have some way of sorting their collections that works for them.
The state of classical metadata is completely atrocious - there are no really good sources (i.e. authoritative) for classical metadata. I’ve often thought that the only hope is for the major labels and distributors to agree on a new common standard for classical metadata and to extend the current tagging scheme to accommodate it. Think Warner, Universal, Sony, Naxos (as a distributor), but don’t get your hopes up. I don’t think they are interested at all in facilitating ease of access to file-based collections. Ideally, there would be an app that would scan barcodes, or use other unique identifiers to definitively identify a recording, and sync authoritative metadata with our recorded files.
Classical music is presented with any number of linking themes… a recording, or group of recordings (e.g. a box set) might be based on the work of a specific composer, a group of related composers, music of a given time (period), music from a given locale, a specific orchestra, form of music (e.g. string quartets), performance style (HIP, avant-garde, baroque, romantic, etc.), a specific soloist, specific instrumentation, etc. etc. How a database manages the disparate ways of sorting this stuff will determine how effective it will be for the widest range of users.
There are a few “quality” streaming providers of classical music, but I haven’t seen how effective their databases are. I have dabbled with Tidal, and while there is lots of classical music available via Tidal, the service is definitely NOT aimed at serving the needs of the serious classical listener. Tidal is like Twitter for audio, for the most part… the only thing that makes it at all usable is searching, and that is far from great.
Many users have elected to use limited tags or simple tree-based file systems for practical reasons, but a large collection overwhelms those approaches quite quickly. The objectives with file-based music should be first, to present the music with best quality possible, and second, to enable quickly and easily locating a specific recording or group of related recordings. We are getting there with the first objective, but the second is nowhere in sight - at least as far as a standardized approach goes.
I have a moderately large classical collection, encompassing recordings from the last 70 years, and while I have a BDP-2 and appreciate its virtues and potential, I have not attempted to rip my full collection because the thought of manually finding/entering/correcting the metadata for many thousands of “tracks” is horrifying, especially given that whatever scheme I choose to use might end up needing to be completely re-done as “standards” for software and hardware evolve. I am very wary of the issues that will arise out of the normal lifecycle of digital replay systems. The more practical, but still daunting solution is a separate external and extensible database that includes a reference to individual files or physical media. That brings us back to the idea of an app, linking to an authoritative third-party database, likely with an ongoing subscription cost.
So, what’s practical for you to implement as part of the next major software release? A basic structure, such as suggested by a few posters above, ideally with some flexibility to edit/associate individual tags in order to accommodate the various schemes used by individuals. Easy to use and powerful searching, using multiple variables. Researching and providing the best-quality metadata lookup possible (which is still going to require huge amounts of manual massaging by users, so it needs to be easily editable.) The effortless ability to present the full range of accented/diacritical characters and common music notation, as used in titles of works, would be much appreciated. Sorting that preserves correct track order for movements and other ordered works. Gapless playback, while retaining functional "track" markers. I'm sure others will provide more core functions that are vital to the enjoyment of classical music.
There is no way you’ll be able to “solve” this problem for everyone. Running MinimServer or something similar customised for Bryston may be attractive, but I'm assuming you're looking to improve the native interface, and many users won't be prepared to delve into complex solutions. Perhaps Bryston can partner with a classical streaming provider, tagging software company, or other entity to gain access to better quality metadata and tag management, but I’m not going to hold my breath - this is classical music, which is seen as a niche market in the industry. As the world shifts to streaming, there will be even less interest in accommodating the needs of people with large collections of physical or file-based media.
Enough for now...
P.S. - I may dig out some worthwhile links to discussions about classical tagging & library management...