It was absurd trickery and PR spin trying to change "MP3 is now unencumbered by patents, enjoy!" into "MP3 is dead."

The significant difference from 2017 is that, now, Opus is well-supported everywhere.

Even in Apple's ecosystem?

It depends. Opus as a codec is, but Vorbis or WebM/Matroska as containers are not.

For VoIP applications, this is irrelevant, given that they often use RTP as a container for streaming and can supply their own decoding libraries for bulk files. (WebRTC on Safari also supports Opus natively).

It does matter for podcast feeds and other cases where the content provider does not control the playback app.

Wikipedia says "Due to the addition of WebRTC support in Apple's WebKit rendering engine, macOS High Sierra and iOS 11 come with native playback support for Opus audio encapsulated in Core Audio Format containers."

And of course VLC works on Macs and knows Opus.

The overwhelming majority of the audio files I have are .mp3.

I don't even know if my ".mp3 player" can handle any other format; it has trouble with a few file formats for sure, so I convert the files just to be able to listen to them. This is usually a once-and-done deal for me. I honestly can't tell the difference other than "does it play?" and I have decent hearing. Maybe if I were some purist with perfect pitch I'd be throwing fits, but I'll leave that for other people. The audio is fine. I don't hear the problem.

The concern is not a one-time conversion to MP3 but several generations of MP3s derived from each other, each one with more artifacts than the last. It's the same problem we experience with image compression. Lossy compression of a lossy source results in a lossier output. From a preservation point-of-view, it's best to have a lossless (perhaps compressed) copy and use that as a source for whatever arbitrary audio format we're using decades from now.

Previous discussion from May 16, 2017: https://news.ycombinator.com/item?id=14347648

I still use mp3 almost exclusively in my audio library, for the very reason the article states -- it's essentially universal. No other format achieves that. Also, mp3 can be more than acceptable if you dial the compression ratio way down.

Agreed. I'd like to also add that having the most pristine audio quality is probably a moot point if either (a) you have some sort of hearing loss or (b) your speakers are not of the highest fidelity.

My personal collection is in MP3 format at 320kbps. I'm very happy with it's ability to be played on pretty much anything and given the (a) minor hearing loss I've accrued over the years and (b) the fact that none of my speakers are ultra-high fidelity, I consider the portability of MP3 an acceptable trade-off over the slight improvement in quality that another format might provide. I also keep all of my CDs tucked away if I ever need to re-rip them for that higher quality.

> AAC and other newer audio codecs can produce better quality than MP3

What does this mean? There are many factors that contribute to a end-user’s assessment of quality, including the mastering, the DAC, the headphones or speakers. So what does “better quality” mean in this context?

Of course, but these factors would impact both MP3 and AAC in the same way, no?

So: Better quality on the same playback system. Is that a bad thing to want?

The article merely makes the (valid, in my opinion) point that often, the benefit of switching is not worth the effort.

Yep, they impact MP3 and AAC the same way. But what does “better quality” from a codec mean when these are controlled for?

It means AAC will reach transparency with lower bitrates than MP3. Transparency means a listener would not be able to tell the difference between the AAC/MP3 file and the original lossless file.