Remember this not a billion dollar company we are talking about here, this is volunteer OSS project. If no one wants to maintain the X11 support then that's up to them.
My issue is that the quality of the macos UI is degrading over time. They can't even get rounding consistent, not quite at windows levels of mismatching yet though.
Also no one bothers making the beautiful native apps now, everything is electron, which is equally inconsistent everywhere.
So I think the advantage over time Vs a Linux system is diminishing... Slowly.
"Pelican on bicycle" is one special case, but the problem (and the interesting point) is that with LLMs, they are always generalising. If a lab focussed specially on pelicans on bicycles, they would as a by-product improve performance on, say, tigers on rollercoasters. This is new and counter-intuitive to most ML/AI people.
The gold standard for cheating on a benchmark is SFT and ignoring memorization. That's why the standard for quickly testing for benchmark contamination has always been to switch out specifics of the task.
Like replacing named concepts with nonsense words in reasoning benchmarks.
I have tried combinations of hard to draw vehicle and animals (crocodile, frog, pterodactly, riding a hand glider, tricycle, skydiving), and it did a rather good job in every cases (compared to previous tests). Whatever they have done to improve on that point, they did it in a way that generalise.
The -C option to avoid copying files and the -W option to prevent writing tags (or set it in the config file)
It will still attempt to match and store the results in its DB for later writing to files, or moving, or querying.
For the data to be most useful, you will need a client that connect to Beets to get the data though, Im not sure how prevelant those are. You can use the web-ui plugin though.
A few people commenting that some of their collection "doesn't exist in any DB", the best way to fix it is to add it to Musicbrainz[0] yourself!
I have found that adding things to Musicbrainz is actually pretty easy (and if you are so inclined like me, pretty rewarding and fun).
Streaming releases (and Bandcamp) you simply drop the release URL into Harmony[1] and it does most of the work for you.
Musicbrainz can represent nearly everything musically related and its all freely licensed, a very cool thing to exist.
Most (non-destructive) edits are auto-applied, whilst the rest go through a 7 day voting period (they are still applied by default unless someone votes against). The barrier to entry is very low.
You're still assuming that all of a person's music metadata _belongs_ in a global database. Some of my collection includes:
-"frankenstein" musical soundtracks where I've assembled my favorite version of each song from all the various published recordings. Sometimes I've even edited pieces of different recordings together into a hybrid track.
-a soundboard recording of my friend's high school talent show performance
-Music I've personally recorded from video games or other random sources
-Songs where I've edited out parts I don't like (such as overly-long drum solos) leaving just the parts I do.
I've organized these things in a way that makes sense for me, not for the world.
This is like a bookshelf that can only store books with an ISBN that can be classified in the Dewey decimal system. Too bad about your family photo scrapbooks or your personal sketchbooks!
What I think they're doing (and what I've done) is add music that wasn't already in the global database to it. For example, a promotional CD, or releases from a small publisher, etc.
In these cases, you're not adding personal metadata or mixtapes.
That said, definitely do self organization if it works for you. Most of your examples seem like they're suitable for custom playlists.
I have similar reservations about custom tags. My collection makes heavy use of non-standard tags, such as performer:<instrument>, opus, key, subtitle, or style. I can't find anything in the documentation about which tags are supported, or for which formats (I keep A/V registrations in Matroska containers, using the same tagging convention -- like using performer:<role> to save actor credits).
There is value for me in having a central database for this data, for example to find misspellings of the same name. But the fear of having 20 years of custom data entry destroyed by an overzealous tool makes me very hesitant to even try a solution like this.
Wow I've needed harmony for years, thanks for sharing! My dumb ass was filling out the Musicbrainz by hand for like two months before I just gave up on beets.
You'll want to bookmark Harmony even if you're using Picard since it's driven by the same database. Happily, once you add an item using Harmony, Picard can find your just-entered release almost immediately via the Release ID of the release you just created.
I find myself needing to create releases for ~10% of the albums I tag, and Harmony is a game-changer for that.
Material Design have been overhauled twice since the theme has been released. That was not the reason why they succeeded, just a solid foundation which they happened to built from.
In the end I just need more available PCIe lanes (so I can chuck more disks in there) and ideally PCIe Gen 5, otherwise I don't have much reason to upgrade.
reply