• pkjqpg1h@lemmy.zip
    link
    fedilink
    English
    arrow-up
    15
    ·
    2 hours ago

    "Anna’s Archive is a non-profit project with two goals:

    1. Preservation: Backing up all knowledge and culture of humanity.

    2. Access: Making this knowledge and culture available to anyone in the world."

    Thanks everyone who contributed this great project.

  • eggdaddy@lemmy.dbzer0.com
    link
    fedilink
    English
    arrow-up
    7
    arrow-down
    2
    ·
    10 hours ago

    I really don’t see this as being useful for anyone outside of hard archivest. The bitrates are pretty trash. I guess if you just a setup with an incredible amount of music no matter what, this is for you. IMHO the meta data is worth more than these lower quality sound files although we have meta data for what’s out there now.

    Outside of that here is what and how they are going to release. I’m guessing this drop was their “popular” track drop. From their site:

    For popularity>0, we got close to all tracks on the platform. The quality is the original OGG Vorbis at 160kbit/s. Metadata was added without reencoding the audio (and an archive of diff files is available to reconstruct the original files from Spotify, as well as a metadata file with original hashes and checksums).

    For popularity=0, we got files representing about half the number of listens (either original or a copy with the same ISRC). The audio is reencoded to OGG Opus at 75kbit/s — sounding the same to most people, but noticeable to an expert.

    • GraveyardOrbit@lemmy.zip
      link
      fedilink
      English
      arrow-up
      14
      arrow-down
      2
      ·
      7 hours ago

      Scientific studies have shown that audiophiles cannot tell the difference in lossless and lossy audio quality when given a blind experiment

      • mic_check_one_two@lemmy.dbzer0.com
        link
        fedilink
        English
        arrow-up
        9
        ·
        edit-2
        3 hours ago

        It really depends on how low the bitrate is. A change from 320kbps (the highest “near-CD” bitrate that .mp3 supports) to 128kbps (standard .mp3) won’t make a huge difference, but a change from 160 to 75 will likely make a big difference… Bitrate tends to be a game of diminishing returns, where a difference between 96kbps and 128kbps is typically noticeable, even by laypeople… But a difference between 320kbps and 640kbps is harder to hear, (or makes no difference at all), even though it’s a much bigger jump between numbers. As the bitrate continues to increase, you get fewer and fewer benefits while your file size begins to balloon.

        To be clear, there is a lot of snake oil in the audiophile world. I’m not denying that. I work in audio, (peep my username), and spend a lot of time dispelling snake oil myths as part of my job. My current audio rig is easily a quarter million dollars, and is located in an acoustically treated room, because it’s built for an entire audience. I’ve also worked in recording and system design. So I’m probably fairly qualified to speak about this specific topic…

        Like lots of snake oil, the bitrate conversation is built upon grains of truth; Just enough to be convincing to someone who only has a surface level understanding of the underlying principles. And audiophiles tend to focus a lot on hardware and manufacturer’s claims, instead of studying what makes that hardware work… Which makes them particularly susceptible to snake oil myths, oftentimes perpetuated by the manufacturers to sell more expensive products to unsuspecting customers. An extreme “low vs lower” bitrate difference is one of the few things that laypeople will be able to identify when presented with an A/B test. In fact, low bitrate comparisons are often used by scummy audiophile companies as a bad-faith “here’s what our competitors sound like, vs what we sound like” example. And to be clear, reducing from ~160kbps to ~75kbps is an extreme difference.

        I want you to think of the most crunchy and heavily compressed “downloaded from limewire on the family computer for your iPod” .mp3 file you’ve ever heard. Full of artifacts, absolutely no high end, sounds like it was recorded with a landline phone, and it crackles when the kick drum peaks. That was probably at least 96kbps, because that’s the lowest bitrate that .mp3 compression supports by default. And that’s after the mp3 compression algorithm has done its lossy “eh, people probably don’t care about this particular frequency” thing. 75kbps is crazy low, and you’ll undoubtedly hear the compression as a result. But again, increasing bitrates will have diminishing returns as the number continues to climb. Going from 75kbps to 160kbps will be a marked improvement, but going from 160kbps to 320kbps will be a much smaller change.

        The reason audiophiles tend to have difficulty with (or even completely fail at) identifying different bitrates is because audiophiles live in a magical land where going from 1200kbps (high-end FLAC quality) to 1411kbps (uncompressed CD quality) makes a noticeable difference. In 99.9% of cases it doesn’t make any difference at all, (because again, diminishing returns) but audiophiles will swear that the 1411kbps sounds better simply because the number is bigger. Again, the snake oil is built upon grains of truth, (differences in low bitrates are immediately noticeable) but only enough to be convincing to people who don’t understand the underlying principles, (at a certain point, bitrate stops impacting audio quality and only makes your file size bigger).

        All of this is to say that yes, the posted bitrate of 75kbps is laughably low. And even laypeople will absolutely be able to hear a difference between the two in an A/B comparison. Because as the bitrate approaches 0, the differences get more and more apparent. And (at least when compared to things like FLAC and CD quality) 75kbps is remarkably close to 0.

        • Lorindól@sopuli.xyz
          link
          fedilink
          English
          arrow-up
          1
          ·
          edit-2
          42 minutes ago

          You are correct.

          This anecdote is empirical, I know, but from my own experience I know how very hard if not impossible it can be to tell the difference between 320kbps and FLAC tracks, even with a high quality setup.

          I happened to find excellent vintage studio monitors some time ago and with my music afficionado friend we wanted to try if we could tell the difference. We are no audiophiles, but we both can tell the difference between good and bad sound.

          Both selected three favourite tracks from different genres and we converted the CD-ripped FLACs to 320kbps CBR and put them on a random playlist with the originals. Then we listened.

          Both got a few right, but I couldn’t really say what it was that made guess the FLAC. It was more like a feeling in the back of your head than anything substantial. “This sounds somehow more alive” is maybe the best description I can give. Or it was just dumb luck.

          Anyway we came to the conclusion that 320kbps can be enough to replicate an enjoyable sound, at least for us. Not one track sounded lacking and we had a good time with our little experiment.

          EDIT: Fixed typos.

        • Lka1988@sh.itjust.works
          link
          fedilink
          English
          arrow-up
          1
          ·
          2 hours ago

          Remember kids, when trying to nail a specific guitar tone, start with the thing that actually creates the audible sound you hear: THE SPEAKER!

    • mushroomman_toad@lemmy.dbzer0.com
      link
      fedilink
      English
      arrow-up
      10
      ·
      9 hours ago

      The point of AA is the archiving.

      Anyways, from a listening perspective, 160kbit vorbis is audibly lossless I think, and there are many songs on here that are not possible to find elsewhere. For popular songs you want, yeah, just download the Flac elsewhere.

    • skisnow@lemmy.ca
      link
      fedilink
      English
      arrow-up
      2
      ·
      8 hours ago

      Most of my mp3s from back in the day are 128kbit, so 160 is an upgrade for me.

    • prole@lemmy.blahaj.zone
      link
      fedilink
      English
      arrow-up
      1
      arrow-down
      3
      ·
      9 hours ago

      I can only speak to what a 75kbps mp3 sounds like, but unless Opus is like 3x better at compression, it’s going to sound like complete dogshit.

      • clubb@lemmy.dbzer0.com
        link
        fedilink
        English
        arrow-up
        1
        ·
        7 hours ago

        I don’t know what 75 kbps opus sounds like, but I can tell you how 32 kbps sounds. Versus mp3 at that bitrate, it sounds actually listenable, while mp3 sounds like you’re underwater.

        All things considered, the Spotify songs probably sound fine at 75.

  • Southern Wolf@pawb.social
    link
    fedilink
    English
    arrow-up
    18
    ·
    14 hours ago

    Hope Anna’s has a good Onion site setup… Cause they are gonna probably have to rely on that soon enough.

    • pkjqpg1h@lemmy.zip
      link
      fedilink
      English
      arrow-up
      1
      ·
      3 hours ago

      I don’t think so there is a mutual relationship with AI companies and the copyright’s future is not bright

    • daniskarma@lemmy.dbzer0.com
      link
      fedilink
      English
      arrow-up
      7
      ·
      6 hours ago

      I’m always astonished of how underused it’s the dark net for these kind of projects. Most torrent sites doesn’t have a dark net mirror despite how easily they get blocked in the clearnet.

    • mic_check_one_two@lemmy.dbzer0.com
      link
      fedilink
      English
      arrow-up
      2
      ·
      5 hours ago

      I don’t think they do, and this will probably be what changes that. I’m fully expecting the site to lose their remaining TLDs as a result of this.

  • Zedstrian@lemmy.dbzer0.com
    link
    fedilink
    English
    arrow-up
    56
    arrow-down
    2
    ·
    22 hours ago

    While their intentions are good, this will unfortunately probably lead to them losing their last two domain names.

    • Luke@lemmy.ml
      link
      fedilink
      English
      arrow-up
      50
      ·
      19 hours ago

      I don’t understand the concern, domain names are cheap and easy to get, they can just keep using new ones. Why does it matter if they lose the ones they have?

      Piratebay used to do the domain dance all the time back in the day (and maybe still do).

      • emeralddawn45@lemmy.dbzer0.com
        link
        fedilink
        English
        arrow-up
        3
        ·
        3 hours ago

        Is TOR not completely owned by the feds? I remember even back in the silk road days people were saying the FBI owns every endpoint. Is TOR still practical? I truly don’t know I’m asking for input.

        • chicken@lemmy.dbzer0.com
          link
          fedilink
          English
          arrow-up
          1
          ·
          edit-2
          1 hour ago

          In this scenario it wouldn’t matter because the idea is to use it as a way to access a website that would otherwise be accessed over clearnet but has become inaccessible. But if they made an onion site endpoints wouldn’t be used anyway afaik since the traffic doesn’t leave the network. Now that I’m thinking about it there might be some issues with practicality doing it this way if they have a big volume of traffic, but there are options for routing around censorship that don’t involve DNS.

          • emeralddawn45@lemmy.dbzer0.com
            link
            fedilink
            English
            arrow-up
            1
            ·
            55 minutes ago

            I don’t understand this comment, can you elaborate? Why wouldn’t the endpoints be used? This is probably my ignorance but I thought all traffic was routed through the onion network and then eventually to the end device, but all that extra routing can’t help you if the Feds control the last stop before whatever server you’re trying to contact… are you saying that if a site is entirely hosted on TOR then no information makes it to an endpoint?

            • chicken@lemmy.dbzer0.com
              link
              fedilink
              English
              arrow-up
              1
              ·
              edit-2
              3 minutes ago

              are you saying that if a site is entirely hosted on TOR then no information makes it to an endpoint?

              Basically yeah. My understanding is that exit nodes are special and using them is a vulnerability, but you only use exit nodes to access clearnet sites from Tor, and you are less vulnerable if you aren’t doing that and rather going to sites with .onion urls. Which, unfortunately I can’t find one for this website, but I’m thinking they’d probably consider making one if they can’t maintain any clearnet domains anymore.

        • Venia Silente@lemmy.dbzer0.com
          link
          fedilink
          English
          arrow-up
          1
          ·
          edit-2
          2 hours ago

          If enough people set endpoints, then the feds will own a fewer proportion of the total. AKA: we have to be the change we want to see in the world.

          • emeralddawn45@lemmy.dbzer0.com
            link
            fedilink
            English
            arrow-up
            3
            ·
            2 hours ago

            Yeah but even if you could get it down to like 50% why would anyone want to take that risk? Idk I might be misunderstanding something about how TOR works but it seems no more anonymous than the clearweb from what I’ve heard.

  • skisnow@lemmy.ca
    link
    fedilink
    English
    arrow-up
    2
    arrow-down
    1
    ·
    13 hours ago

    Where? I checked the torrents JSON mentioned there and there’s no text match on ‘spotify’… did it get removed or am I looking at the wrong JSON?