Archive for March 29, 2023

Wednesday, March 29, 2023

Apple Music Classical

Apple (Hacker News, MacRumors):

Apple today launched Apple Music Classical, a brand-new standalone music streaming app designed to deliver the listening experience classical music lovers deserve. With Apple Music Classical, Apple Music subscribers can easily find any recording in the world’s largest classical music catalog with fully optimized search; enjoy the highest audio quality available and experience many classical favorites in a whole new way with immersive Spatial Audio; browse expertly curated playlists, insightful composer biographies, and descriptions of thousands of works; and so much more.

Does this do anything to fix Siri for classical music? Currently, it’s almost unusable because the titles of the albums and tracks don’t really match the names that a human would think of. And those are not always the right granularity: even if I manage to request a symphony and it plays the first movement, it doesn’t know which of the subsequent tracks on that album are part of the same piece.

Joe Rossignol:

Apple today published a support document explaining why it decided to release a standalone Apple Music Classical app for classical music.

M.G. Siegler:

But it’s also odd in that it’s not like they stripped all of this music out of the main Apple Music app — as far as I can tell, it’s all still available there. So it’s not about de-cluttering that app. It’s more about letting classical music lovers dive deeper and find easier the music that they love. Fair enough, I guess. But I suspect the real reason Apple is doing this is simply because they acquired a beloved classical music service, Primephonic, a couple years ago. It was a way to help differentiate Apple Music versus Spotify. But they undoubtedly also knew that it would piss off users of that service if they killed off that standalone app. So here we are, two apps, one price.

[…]

It’s fine. It looks nice, as you might expect. But it’s also very similar to Apple Music.

[…]

Here’s something else that’s odd. After again, a year and a half of work, Apple Music Classical is iPhone-only. Every app developer knows that Apple likes to nudge, subtlely or overtly, cross-Apple-platform support for apps. “That’s a nice iPhone app. Would look great as an iPad app as well.” That kind of thing. How can Apple realisticly give such guidance if they themselves don’t have that ready to roll on day one of their services!?

[…]

My above stated routine actually doesn’t have me using an iPhone to listen to music. I’m using either my iPad or Mac. And yet I can’t use Apple Music Classical on either of those. […] And wait, it sounds like it may never come to the iPad or Mac! “Built exclusively for mobile”?! What on Earth?

Dan Moren:

Apple Classical is a bit of a strange beast.

[…]

I particularly appreciate the refinement Apple has done to the Library section. While it’s shared with Apple Music (you’ll see albums already in your Apple Music library that are also available in Apple Classical), there are also new sections for Recordings and Works, and refined sections for Artists and Composers. In all of those cases, Apple has implemented a Favorites system, where you can choose to save specific items, rather than simply providing an exhaustive catalogue of every single artist or recording you have in your collection.

[…]

There’s a lot of curation happening in Apple Classical, too. For example, searching for a specific piece will generally pop up a ton of recordings, but the app will identify an Editor’s Choice version as well as Popular Recordings to help separate the wheat from the chaff.

[…]

And lo and behold, Apple Classical includes soundtracks in its catalog, along with scores for Broadway musicals and operas.

Which is certainly cool, but the implementation as it stands right now is a little wonky.

Dave B.:

Something I immediately noticed is the new filter icon, with ‘Filter By’ functionality.

This is a good sign because it means there’s a solid chance we’ll get it in the main Apple Music app.

I’ve been asking for this for years!

Benjamin Mayo:

One odd UI nitpick with Classical I immediately noticed is if the table view is currently scrolling, the edge swipe back feature doesn’t work. This and a few other things gives me the sense the underlying codebase is more different to Apple Music than it first appears.

Now Playing UI report: more like Apple Music than Podcasts is but it’s still different. It has the iOS 16 Music swipey sliders but doesn’t have the same animations on the buttons and the album art transitions differently too. No long-press context menus anywhere.

Previously:

Update (2023-03-30): Kirk McElhearn (Mastodon):

To this day, there is no standard database of recordings that can be used to ensure metadata integrity.

I tell this story because it shows how difficult it is to set up a service enabling users to catalog and search for classical music. But without such a catalog, classical music fans have been forced to put square metadata into CD-shaped holes, starting with the earliest days of iTunes.

[…]

When Apple added the Work and Movement tags to iTunes 12.5 back in 2016, the company showed the first signs that it might finally be interested in classical music. Those tags enable you to manage multi-movement works, tagging them together, and they are useful for search.

Apple’s biggest task in preparing for the release of the Apple Music Classical app was to update the tags on potentially millions of tracks.

[…]

Nor can you download music to listen offline, though when you add albums to your library, they also get added to your Apple Music library, so you can download them in the Apple Music app and listen to them there.

[…]

Here are two screenshots: on the left is the Apple Music app, in which you can see that there’s no way to discern the name of each track; on the right is the Apple Music Classical app, which displays the work name on the top, in bold, and the movement names below, all visible.

He found some problems with search and metadata, but they’re much improved.

Chad Ossman:

I know this is going to sound hyperbolic, but as a lover of music in general, and a very amateur musician, I find the concept of separate apps for various genres almost offensive.

[…]

An important point she makes right away is that even if complex tracks are tagged with the correct metadata, the typical mobile interface doesn’t even have the physical screen real estate to display long text fields. But again, I contend that this is a solvable design deficiency, not cause for creating a walled garden, solely to accommodate one genre of music that typically has longer names than most others.

[…]

So, given all of these factors, and after going to all those lengths, why did Apple opt to create an entirely separate app to showcase its revised music catalogue and curated classical selection, as opposed to correcting the deficiencies of their existing product? I fail to understand what is unique about the broad category of classical music that would require an entirely different app/service than all the existing ones for, you know, music.

[…]

If Apple were to improve metadata support for the existing Apple Music streaming service, the benefits would extend well beyond classical music. For example, Beatles obsessives would surely appreciate quick access to the 2009 remaster of the 1970 mix of “Get Back” from Let it Be, not to be confused with the 2021 Giles Martin remix, the 1969 rooftop performance from Anthology 3 (released in 1996), the 2003 Naked version, or the 2009 remaster of the 1969 Single Version from Past Masters (released in 1988).

Update (2023-03-31): Benedict Evans:

Apple Music Classical is just totally unusable. Very disappointing. I can’t browse a library by composer or by piece, and that’s designed behaviour. 🤦🏻‍♂️

[…]

It gets better. The search function does not search your library. This really was not ready to ship. 🤦🏻‍♂️🤦🏻‍♂️

[…]

I have hundreds of albums, and my composer screen is blank. You cannot browse your library by any field have the other than the date you added it to the library

Dave B.:

Why are all these sections in the @AppleClassical app’s Library shown as blank?

I have tons of classical music in my library. But if it’s not auto-populating these sections, that kind of defeats the whole purpose.

Bug or bad UI?

Each of those is described as “Add Your Favorite”, but they are not designated as favorites sections. These are the views within the Library tab.

Update (2023-04-05): Nick Heer:

It seems Apple Music Classical is merely a version of the Primephonic app Apple acquired, skinned to resemble the standard Music app. After a week, I find this similar-but-different quality creates a muddled experience as a user. The Classical app does not support context menus, transitions between some screens see the title animate from one direction while the rest of the screen animates from the other, and the Now Playing view does not respond to dragging up or down as it does in Music. The Library view is uniquely confusing and buggy. It feels more like a third-party app in the style of something Apple could make more than it does than Apple’s own app.

[…]

By the way, the more I think about this app, the more I came to realize Apple’s interpretation of ID3 audio metadata is limited — and not just for classical music. The most recent spec was published in 2000, and contains fields for entries like remix artists, source medium, and live recordings, by way of descriptive text, but Music does not support any of these fields. It never has, even when it was called “iTunes”. Instead of using this comprehensive structure, songs in a local music library, in the iTunes Store, and in Apple Music resort to a series of parentheses for featured artists, live recording venues, and the album version or master — “How the West Was Won (Live) [Remastered]”, for example, or “Momma Sed (Alive at Club Nokia) [Live]”. There may be good reasons for these limitations. But it would be better if all the songs and albums in Apple’s catalogue contained more fulsome and better-structured metadata. The spec is already there and it needs mass adoption.

Update (2023-05-30): Hartley Charlton:

Much like Apple Music for Android, the Apple Music Classical app for Android largely mirrors the design of its iOS counterpart. Notably, the Apple Music Classical app for Android comes before Apple has made the app available for the iPad or Mac.

Update (2023-06-30): Benedict Evans:

I am still really disappointed that Apple put all that time and effort into building a classical music app, correctly identified all the reasons why it needs to be a standalone app, and then delivered something that is absolutely and completely unusable. It’s the new Apple Maps.

[…]

[There] is no way to build or browse a library. I have over 1000 albums saved and this is the screen[…]

[…]

There is literally no way you can browse your library except by the date you added the album. Things in your library don’t even show up in search.

Apple Pay Later

Apple (Hacker News):

Apple Pay Later allows users to split purchases into four payments, spread over six weeks with no interest and no fees. Users can easily track, manage, and repay their Apple Pay Later loans in one convenient location in Apple Wallet. Users can apply for Apple Pay Later loans of $50 to $1,000, which can be used for online and in-app purchases made on iPhone and iPad with merchants that accept Apple Pay. Starting today, Apple will begin inviting select users to access a prerelease version of Apple Pay Later, with plans to offer it to all eligible users in the coming months.

John Gruber:

Apple Pay Later is only available for online purchases — either on the web or in iOS apps. I guess that makes sense — competing services like Affirm, Klarma, and Afterpay are only available for e-commerce transactions too (I think?). I just never before thought of online Apple Pay being different from brick-and-mortar retail Apple Pay.

Previously:

Hachette v. Internet Archive

Jay Peters and Sean Hollister:

A federal judge has ruled against the Internet Archive in Hachette v. Internet Archive, a lawsuit brought against it by four book publishers, deciding that the website does not have the right to scan books and lend them out like a library.

Judge John G. Koeltl decided that the Internet Archive had done nothing more than create “derivative works,” and so would have needed authorization from the books’ copyright holders — the publishers — before lending them out through its National Emergency Library program.

[…]

He notes that the Google Books use was found “transformative” because it created a searchable database instead of simply publishing copies of books on the internet.

Mike Glyer (via Hacker News):

A copy of the decision can be downloaded from CourtListener.

[…]

John G. Koeltl, District Judge of the U.S. Southern District of New York, granted summary judgment based on the record, as was requested by both sides. His analysis of the case is illustrated in the following quotes from the decision.

dahart:

I have good feelings for the Internet Archive, but in this case it’s about a handful of books that are being copied and distributed a mere 5 years after initial publication, which I think a lot of people who want copyright shortened would still agree is quite a bit too short.

[…]

Keep in mind that this isn’t about the Internet Archive specifically. If the court ruled it’s okay for them to copy and rent books, then anyone can copy and rent books, it undermines the entire market for books (and also web sites and images and other media, because this is the Internet Archive).

Previously:

Update (2023-04-03): Chris Freeland:

For those asking how you can support the Internet Archive, there will be a rally on the steps of the Internet Archive on Saturday, April 8 @ 11am PT.

[…]

The nonprofit Internet Archive is appealing a judgment that threatens the future of all libraries.

Chris Freeland:

This decision impacts libraries across the US who rely on controlled digital lending to connect their patrons with books online.

[…]

But it’s not over—we will keep fighting for the traditional right of libraries to own, lend, and preserve books. We will be appealing the judgment and encourage everyone to come together as a community to support libraries against this attack by corporate publishers.

There are no details about what exactly they are appealing, so perhaps I’m missing something here, but this seems like an incredibly dishonest framing. I thought the whole point of the case is that Internet Archive was not doing what libraries have traditionally done. Their “library” was more like the book equivalent of Napster. And I don’t see how the ruling would affect “all libraries” or even those that “rely on controlled digital lending.”

Update (2023-04-04): The lawsuit happened as a result of the National Emergency Library and its unlimited “lending,” but it also covered the Internet Archive’s Controlled Digital Lending (CDL) program. So the part I wrote about CDL above is wrong. I had thought that this included services like OverDrive, but it has a more specific meaning. I think I was right about the other part:

Adi Robertson:

The Open Library is built around a concept called controlled digital lending, or CDL: a system where libraries digitize copies of books in their collections and then offer access to them as ebooks on a one-to-one basis (i.e., if a library has a single copy of the book, it can keep the book in storage and let one person at a time access the ebook, something known as the “own-to-loan ratio.”) CDL is different from services like OverDrive or Amazon’s Kindle library program, which offer ebooks that are officially licensed out by publishers. It’s a comparatively non-standard practice despite implementation in places like the Boston Public Library, partially because it’s based on an interpretation of US copyright doctrine that hasn’t been strictly tested in court — but this is about to change.

[…]

Publishers took aim not just at the National Emergency Library, however, but also at the Open Library and the theory of CDL in general. The service constitutes “willful digital piracy on an industrial scale,” the complaint alleged. “Without any license or any payment to authors or publishers, IA scans print books, uploads these illegally scanned books to its servers, and distributes verbatim digital copies of the books in whole via public-facing websites. With just a few clicks, any Internet-connected user can download complete digital copies of in-copyright books.” More generally, “CDL is an invented paradigm that is well outside copyright law ... based on the false premise that a print book and a digital book share the same qualities.”

Update (2023-04-06): marwis:

Amazing how reckless IA is with their library program while at the same time being extremely cautions with Wayback Archive to the point of retroactively applying robots.txt and removing website archives because “they can’t afford even the smallest risk of being sued”.

causi:

We all screamed “Hey, you idiots are going to ruin everything if you act like the pandemic has magically nullified the concept of copyright” and they fucking did it anyway, and now exactly what we said would happen is happening. […] Controlled digital lending had a chance of getting off the ground. The IA’s Emergency Library’s unlimited digital lending burned it to the ground and stomped on the ashes.

rsync:

However, as a long-time, regular, sustaining financial benefactor of IA, I was annoyed that they strayed into this set of activities in the first place and then dismayed when they pushed the envelope on it during C19 quarantines.

I, and others, predicted this trouble and even if there is not direct causation why do they have to tickle this dragon in the first place

Spooky23:

The “emergency library” was a gift on a silver platter for the publishers. It was a reckless and stupid action that made the worst case scenario that the publishers wouldn’t have been taken seriously about a reality. Obviously the 1-1 lending was an issue to the publishers, but higher risk from a litigation perspective.

jacquesm:

And it is strictly speaking not a thing that IA should be doing in the first place. They should have set up a completely different legal entity for this and kept them visibly separated on all fronts, especially online.

Their core service is too important to put at risk testing copyright law.