Indeed, setting-up a work type structure is going to be tough …
Some of the types that you have listed may be considered as “movement type” rather than work type.
Yes, that’s true. For that list I tried to cover every type of composition.
Be it a top level work, such as Opera or Requiem, or be it a self-contained composition, or a type that can be both.
My objective with that list is that it should contain all types of compositions that you would need to label either a high-level or a low-level work/composition with.
And (very important) not have it containing anything that is only a genre.
Especially that seems to be a big problem when labeling classical compositions.
Mixing up genres and work/composition types is pretty much everywhere I looked.
For this forum post I chose to post it as a flat list, but for constructing it I did take the different levels of classical compositions into account. But I didn’t want to complicate things too much at this first attempt to see if this ‘resonated’, and if something could be done with this.
This seems a pretty subjective area to me. For example, you list “Chamber Music” as a work type whereas I would call it a genre. One suggestion would be to put all your genres into the permissible genres box and all your work types into the sub-genres box. As for different levels, that is a 2nd order consideration - perfectly doable, but needs some thought to make it useful and not just a clumsy addition.
No argument there.
Also understand that this is my list v0.1 alpha.
If anybody has a link to a better list, please let me know?
I don’t want to hack your thread here, so maybe I should create a new topic for getting feedback and possibly improving on this list?
About Chamber Music:
Chamber music is a form of classical music that is composed for a small group of instruments—traditionally a group that could fit in a palace chamber or a large room.
So it is composed with a specific objective, and a rather specific restriction on musicians and environment.
That goes beyond ‘genre’. (I think)
Also, as I said, my objective for this list is that if you have a work or a recording and you want to tag it with a label, the label should be in this list.
And when doing the research, I got the notion that there are quite a few ‘chamber music’ works/compositions that you could not label with anything other than ‘Chamber Music’.
(and I want to be able to label everything 
Thinking out loud:
I think I will be going to setup two different ‘type’ tags:
One for the top-level work, and one for for lower-level and/or independent recordings/tracks.
Maybe call them something like ‘Work type’ and ‘Composition form’.
The first could be: Opera, Symphony, Requiem mass, etc.
The second: Nocturne, Pavane, Etude, etc.
To my knowledge, there is no commonly agreed taxonomy of work types.
Setting up any such form of classification would be a huge undertaking, and subject to changes …
I would take a different approach, less structured, but which I find more practical.
I would use tagging, to flag works with attributes useful to select them.
For example, you could have “sonata”, “piano” tags for a piano sonata, “sonata”, “violin” for violin sonata, etc …
1 Like
I have local MB working. Does the throttling also applies to the local instances ? If yes, is there a way to bypass this ? I’m not beyond editing the code if you point me into right direction 
Hi @nadl40. In theory, the throttling should not apply. I would be very interested to hear how it performs.
Edit. The throttling is in Picard not the plugin.
To the best of your knowledge, is the Picard throttling auto disabled when using local server ?
I really can’t answer as I am away from my PC and don’t know off the top of my head. Perhaps @outsidecontext can say?
Hi @MetaTunes. I’v done a quick test. 899 flac tracks, 30 or so albums.
MB run for 14 minutes, on my local for 12 minutes, so not a big difference.
Have you seen/tried this?:
1 Like
1 Like
Is that ever working. On my local MB it went from 12 minutes to 6 minutes, limiting factor was understandably fingerprinting. On the refresh, no cache, it took under 2 minutes. Great improvement 
1 Like
I’m curious: how ‘powerful’ is your local server? (cpu/memory)
I am asking because it was mentioned on this forum that you would need a really powerful computer to get some noticeable speed gains from running a local database.
Hi @hiccup. Not powerful at all, just a Lenovo ThinkCentre i7 with 8GB and SSD for linux host, 4 threads on 2 cores, local network, off lease machine for about $300 USD.
I’m working on my own scripts and if I did not have a local server, it would be very slow as I’m doing lots of calls for Classical.
That would have my (file) server for breakfast.
(Intel Atom based, 2GB)
1 Like
well, my file server is PI4…, databases need more cycles.
In trying to figure something out, I am running into something I don’t understand:
I thought that the count of cwp_work-part_levels would match the number of populated _cwp_work_n entries.
So if the level count is 2, you would have entries for: _cwp_work_0 and 1
If 3, _cwp_work_0 ,1 and 2
etc.
But looking at these three examples, there doesn’t seem to be a relation.
(only the second one displays what I would expect)
Am I misunderstanding how this works? Or is it a result of inconsistent ways in how these releases were entered in the database?
Is that one release or several? Can you post the link(s)?