MusicBrainz - Robert Kaye - Community, Open Source, and Metadata - The Lorenzo's Music Podcast (Transcript)
Listen on:
Tom Ray: Hi and welcome to another Lorenzo’s Music Podcast. I am Tom Ray, and on this show I like to talk to musicians, and I like to talk to people that create things for musicians. And I also like to kind of delve into the open source, open web sort of realm. And today I am talking to someone who has built something for musicians, does things for musicians, does more than that, and that’s what we’re here to find out more about. so why don’t you introduce yourself and tell the people what it is you do.
Robert Kaye: Hi, I’m Robert Kaye. I am the founder of MusicBrainz and as I like to call the lead geek of the MetaBrainz foundation, which is the nonprofit that organizes MusicBrainz as well as several other projects. And what do I do? I created that project, 25 years ago because, well, there was a precursor project to that as well. We can go into that tangent if we need to, okay, about open source communities and so forth. and I started that project because the CDDB database was taken private. quick reminder of that. CDDB was this database that was one of the first open databases. This is before the terms open source was even coined for, having information as to what’s on a cd because the track titles were not actually put on a CD very intelligently enough. And, a couple of college kids made a database and allowed other people to contribute to it. And I typed in a couple hundred CDs because I thought this thing was great, right? And then actually one day, the next find out that, they’re taking this private and, sold it off to big company. They all, you know, pocketed a few million, which is big money back then. And they went on about their happy business. And, I was pissed off, and I continue to be pissed off so pissed off all the time that this happened. That one time, standing there continually crutching to a friend at a party back in California and he said, like, man, man, just shut up. It’s time to start your own open source project and just, just leave me alone. Just start your own project. And you turned heel and walked away. And I was like. And that was literally the beginning of, of the, very impetus of music writings. wow. I’m sticking with a couple of steps, right? But, I mean, he really, you know, got me to act, he got me to do things right. There was a quick precursor project called the CD Index. And I had no idea about open source, no idea about communities, nothing. I started hacking on some perl scripts Because I got fired from a really shitty job and I was wondering what to do next. then there was a slashdog post and the slashdog post said, hey, what are we doing as a replacement for cddb? And I sat down and buckled down, finished a couple of scripts and then I said, I’ve got the replacement for cddb. Then the slashdot effect came down on me and my site. Really everyone and their brother came and started shouting over how everything like SQL will never scale. You cannot build a webpage like this. All these other things.
Tom Ray: They instantly went into. You used the wrong thing instead of like what, a great thing you did.
Robert Kaye: Alan Cox showed up and was insistent that we use basically CD lookup over DNS and all kinds of other really, really bizarre things. The community, since I didn’t know how to manage it, went toxic immediately. It was just a shit show and I want nothing to do with it. And I got disillusioned with the whole thing and I just, just kind of walked away from it. then came the dot com crash and I had my beloved S2000 Honda Roadster, which is still the best car I’ve ever driven. So fun. and I had to do some soul searching. I had to decide what am I going to do? Am I going to go move to the Bay Area? And I was living in San Luis Obispo at the time. move to the Bay Area and work on somebody else’s dream and keep my dream car or work on my own dream and sell my dream car. By the way, I’d regretted selling every single Honda I’ve ever sold. To be honest, that said I sold the car and buckled down and said like okay, this needs a better name. This needs to be forward looking. This needs to not just be about CDs, it needs to be about music because CDs are on the way out. My timing was a little off on that one. and we need to look towards future digital, yada, yada, yada yada. And I decided to call it Music Brainz and started working on Music Brainz and got some, a little bit of traction from people because I was also working the Freeamp project which was sponsored by E. Music, my, my next employer. Then, which is all fun and good and I just piggybacked off of that and I don’t know, for years I worked on the site and like, okay, a database is now cool enough. I think we should reset the database and now we can start fresh. Everyone absolutely flipped their lid and said no no, no, you can’t restart. We’ve been working on this. This is fine. I was like, okay, literally that database is what exists today. I mean, we’ve dragged it along through many versions of database changes and all these other things, but, it’s been in continuous operation ever since. Right. And it’s grown into a much more significant sort of project than I ever dreamt it would and so many more people’s lives than ever thought it would. you know, put it into context in a lot of ways. we are now an elemental cog in the infrastructure of the Internet when it comes to music anyways. Right. So and the thing when, you know, we are not very well known unless you like tagging your music collection and so forth, but in general we’re not really one of the better known open source project. And we’re okay with that. We’re okay with that. This is all about communities and communities don’t like runaway growth. So we’re actually kind of cool with that. But we find ourselves in this position of actually being these, these kind of strangers. But we can say, can walk into any room and say, like, our data is on your phone. And not be lying about that because everybody’s got either an iPhone or an iOS phone in their pocket and it’s got like YouTube music or it’s got Spotify, or it’s got something in it, it’s got our data in it. Right. And people like, whoa. Really was like, yeah, but we’re the Internet backbone for music metadata. I mean, okay, you’ve got grace note, you’ve got other companies and so forth, but the open stuff is obviously going to win the day. Right. Because everyone can use it and you know, that’s where everyone can use it. As one of the early, axioms that I put in place, no barriers to entry to get data in and you know, remove the barriers to entry, make it as easy as possible for people to get hooked on our data. And I literally called it a drug dealer business model. Right. First one’s always free, Right. But then you come and you want to use it in a commercial context. Fuck you. We got you. Right.
Tom Ray: Look here, Walter White.
Robert Kaye: In a lot of ways, sure. So you know, we made it easier for people to dive in and to use the data and including for commercial, commercial purposes. Right. To other companies, on an hourly basis.
Tom Ray: And I mean that’s, that is the thing on open source is that oddly enough it is also used to make, you know, copyrighted and corporation like they’re built on open source things, so. And that’s the point of it. Now there are a few things I want to ask you about too. Now, first of all, where are you located right now? Because you were talking about being in San Luis Obispo and the Bay Area. But, but right now I know I’m talking to you and you have a time difference from me, so you are in a different location. Where are you currently located?
Robert Kaye: I’m in Barcelona.
Tom Ray: Barcelona. Wow, nice.
Robert Kaye: Okay, so I am originally German. I still am German. Grew up there for 13 years, moved to the States, ended up in San Diego, graduated high school there, and then went to Cal Poly and San Luis. Ah. And I said, like, when this amazing community of friends dissipates and breaks apart, I’m leaving the United States. That only took another 26 years. So. And then after 26 years, a group of friends was finally done and I packed up my stuff about, 12 and a half years ago, so, and moved to Barcelona because I realized it didn’t need to be in California anymore, to live in Europe. And I’ve always wanted to live in Barcelona. So, I mean, why not?
Tom Ray: You know, I would like to too. Maybe I, maybe that’ll happen for me soon.
Robert Kaye: The key here is that, that European passport, right? Because at that point it’s just pick up, pick up sticks and go. Right? Like not having to deal with a visa or any of that. Key.
Tom Ray: My, my. Actually on a, on a side note, my sax player, just like there’s some sort of thing where you can apply for citizenship if you have a, a relative or some sort of lineage to, I want to say Yugoslavia or something like that, but you could actually get one where you can freely move between the US and he’s actually applying for that. Not that he’s going to do it. It’s just like, oh, that’s kind of a neat idea. And he’s just doing it for the sake of doing it. And also just like you never know. So I’d never heard of before. Yeah. anyway, so, and the other thing I wanted to mention too, and this is first of all, going through that story, mentioning some of the things totally forgot about. Slash dot cd, cdd, Jesus, cddb, which I forgot existed. So you clearly did. Or at least in my mind, your part, because it pissed you off. I remember when it closed down or it was difficult to use or you had to. I don’t even remember what, what the deal was, but it was like, it was hard to attach it to things or you couldn’t connect it to certain players or some dumb thing like that.
Robert Kaye: Yeah.
Tom Ray: And I want to say that’s when I discovered Music Brainz as well. Because I also around that same time was toying around with I was dual booting Linux computers. I was, I was using a Windows computer, but I was learning Linux. And I think breezy Badger on Kubuntu was the first distro that I ever installed and I was trying to find ways to as you said, get my CDs. That. Because then what you did is you put them on, you know, your hard drive and listen to them on your favorite player because that’s what we did. Or you know, let’s be honest, you downloaded it from wherever you could find it. But that was the thing is all the information was garbage or it was labeled wrong or like, you know, the CD when you ripped it, you didn’t necessarily have the right thing. So I want to say that’s where I first heard about Music Brainz. Now I looked at my account and I had been on there since I signed up in 2005.
Robert Kaye: So. Yeah. Yeah, that’s impressive.
Tom Ray: Yeah.
Robert Kaye: Thank you, thank you. yeah, and what you were just saying is interesting because the whole my music collection is shit because I downloaded off of Napster and has shitty tags and I can’t shove it into my clean, happy new iPhone. A boon for us. Right. Because people, people came and needed to use Picard to clean up their music collection. Yes. And you know the, the metric I always put out was, you know, if somebody takes a thousand hours of effort from us and gives us an hour back, we’re happy.
Tom Ray: Right.
Robert Kaye: the same sort of thing with Picard. If you download all of this data and you add two or three new releases to the database, we’re cool. Right? Very happy with that. And that’s basically how it worked. And there were tons and tons of people came wanting to get their database, of files cleaned and tagged and that’s what helped us build a community, a company and everything else around it, nonprofit, more specifically around it, to let us actually get up on our own two feet to become a self sustaining organization.
Tom Ray: And did you? I wanted to know too. You said when you were talking to this person at the party, the person who basically got so pissed off it made you go, okay, yes, I will do something just so I can prove that you shouldn’t be pissed off at me sort of situation. But did you already have a programming background? Where you. Or did you kind of learn. Okay, all right, you did. Because you never know. It could be one of those things like. So I sat down and I read a bunch of books on how to run databases.
Robert Kaye: And no, it wasn’t that far fetched. I was more a graphics sort of geek and very budding, very early on web sort of geek. That was very early days when I jumped into that. And I’d always avoided database courses in uni because like, oh my God, who wants to work with databases for the rest of the.
Tom Ray: Oh my God. Yeah, that’s me. That’s why I’m working on a Jekyll site right now. Because I’m like, well, I can just kind of tell it what to do and give it the information I want and just write it into a text document and. And then it spits it out and it’s just a regular HTML site and I don’t have to maintain a thing. Yeah, yeah, well, and that, that leads into the, the database itself. So now the music database is a source of of the information people, but it’s also, it can be maintained by, if you join as a member, but you also have people. Well, here’s what I want to ask you. So I still get sometimes alerts that information that I put and truthfully so I mean back then when I first started using it, I was using it to go, oh, this information is wrong on an album. But then as time went on I just update my own band on it. So when we, when we do a release or you know, but I will still get information of like somebody will correct it or go, oh, this one thing that, this one thing that we had shared it to no longer exists. And I’ll get updates like somebody’s going through there and updating it. So is that the community? Is that people that works there?
Robert Kaye: No. So it’s the community. And when I explain about where the real value of meta Brainz and music Brainz lies, there’s no value in any of the source code. There’s no value in any of the servers. There’s no value. There’s hardly any value in the team itself and there’s no value in the database. The value itself is in the community because the database is. When we start dumping the database, by the time the database dump is done, it’s out of date because we update things come to the database once a minute or so, every 30 seconds new edits are applied. But once those edits stop, the value stops and the team does not edit. Every once In a while we get some people that are confused and we just look at it and say, it will take me 2 seconds to edit this or 20 minutes to try and explain this person how to edit this. And they don’t need to go through this page, then we’ll fix it for them. Right. As a sort of a customer support. Yeah, Right. But on the whole, our team does edit quite a lot. And my, you know, I don’t have a right hand man per se, but one of my closest team members, he’s got over 4 million edits or so, right? Just 4 million edits. Okay, let’s see. I’ve had 4 million heartbeats and 4 million breaths and batted my eyelashes 4 million times. Probably 4 million steps. Right. 4 million edits to a database as a volunteer is insane. Right. And unfortunately, and my hat off to Dr. Saunder, a long time contributor, had been part of music Brainz for 18 years, recently died and he made 300 edits to the database and every day for 18 years.
Tom Ray: Wow. Okay.
Robert Kaye: Right. I mean, when you start an open source project, you know, I think everyone should have a moment of, you know, I’m just clicking Create Repository or something. Right. But we should have a moment to think of what might actually happen in the future. Right. And this is a particularly good scenario. Right. We can, I can see that people have gotten married and bought houses and all of this other stuff because I started my open source project and community congealed around it. Right. But this sort of message is actually a message that I think more people should really be thinking about when they start their projects, because of the negative consequences. Right. There’s too few people that actually think about how could this thing that I’ve just created or planning on creating be used for evil rather than for good. And interestingly enough, Mary Shelley’s Frankenstein was that some of the key themes of the book, which of course doesn’t come through in the horror movies that we get today. Right, right. To me that what haunted me about the book was the whole time he ran away from his creation the entire time. And while I can empathize with that at times, I still am, you know, running the show. Well, running the show. And as much as an open source project like that, you ever run the show. Right, right. Yeah, the.
Tom Ray: Well, and it really is the community too. So you run it in the sense that you’ve created it, but it really lives through the way people use it and what gets done with it. Now that being said too, we had talked about. So Music Brainz, the database. And you had also mentioned Picard. Now Picard is actually an app or software. It’s something you download to use and interact with. Music Brainz. Now tell me how that came about, how, you went from. I mean, it’s essentially you built two separate things, but they had to connect and one of them runs to update the ID tags on the files themselves.
Robert Kaye: That’s. Yeah. So the card was always. So for me, when I first started looking at this project and how to make it work, I needed to identify what was called, what I called beneficial feedback loops. So the idea of a beneficial feedback loop is if someone can come in and say, okay, you have an application that makes it a lot easier for me to tag my music collection and I do this thing and then I give you a little bit of effort back. This creates this feedback loop where the people that are taking it are giving and the ratios don’t have to be perfect. Like I said, you can be really lopsided and still work perfectly fine. This feedback loop, Picard, was our original feedback loop that then allowed the hordes that had their shib mapser collections to tag their collections for us to build our database and, for everyone to walk away a bit happier. Right. So Picard is, is that tool and Picard in use? I mean, it’s 20 years old. Right. So Picard in this use for, I don’t know, the middle 10 years was. There were a few releases. It wasn’t particularly, There wasn’t a lot of happenings. But now, and I’m sure you’re aware of this as well, is that a lot of people are really pissed off at Spotify and Spotify is no longer really serving a lot of communities. And so people are now going back to their music collections. Me, myself, yeah, I’m, you know, I’ve dusted off my collection and I’ve, copy of Navidrome that I’m looking at right here on the desk that I’m. That I’m streaming from and so forth. And I’m much happier doing my own thing because when I upload a, release to my server, it’s not going to disappear tomorrow can’t be set for Spotify anymore. So there is. What we’re seeing is a resurgence of people that are really pissed off by Spotify to go back to take care of their own music collections. This sort of personal digital sovereignty over your own music experience, which I’m really behind because Spotify is really Spotify and Everyone else have pretty much let us down in the grand scheme of things. So people are, rediscovering their music collections and falling in love with them. And this is actually, believe it or not, where I’m doing a lot of work. Because, the work that we’re doing at Better Brainz extends far beyond just Music Brainz. we also have the Listen Brainz project, which is like, last step down, right? Which is its own. It’s a completely different thing and quite, complementary to Music Brainz. and I’m personally working on another project called Listen Brainz Local. And the idea with this. And this is. This was actually. I don’t know about.
Tom Ray: Yeah, tell me about that.
Robert Kaye: It’s not out yet, right? It’s still early days. but, when I was thinking about Music Brainz and creating Music Brainz, I was originally kind of pissed off by the fact that, come on, it’s the year 2000, right? We’re still picking this track and picking this track and then picking this album, this track. Like, isn’t there a better way to do this? Right. I, was wishing for a recommendation engine, basically, right? The terms weren’t coined yet and all of that. I said, like, I want to write that. I want to write that. What do I. Oh, I need data. Oh, I need a mountain of. Okay, let me do music banks for a few years and then come back to it, literally. It’s taken me almost 25 years to get back to that, but now the whole getting back to it. We’ve been working on building recommendations with Listen Brainz and, all of these tools for music discovery. If you go to Listen Brainz and go to the Discover tab, there’s some really cool stuff, including the fresh releases, which it shows you everything that’s coming up. And it’s custom tailored to your music taste because we know what you’re listening to. So when an artist you’ve listened to more than three times has a release coming up, it shows in your feed. It’s really cool. You need to check it out. It’s really nice. So we build all these tools, but there’s a lot of people, like, that’s nice. I don’t use Spotify. What do you got for me? Right, so to basically build these tools, but to actually do them for local collection is a completely different headache, right? And you can’t just say, hey, upload the contents or the metadata of your collection to our server. We’ll make recommendations for you. That’s one, too much data. And two, as you said, we don’t always know what people downloaded their stuff from, they may not be entirely proud of where that the provenance of their music collection. Fair enough, right. Privacy issues. There’s just, that’s a non starter. So then the idea is like, okay, you have to actually go into the user’s collection and then scan the collection, match it against Music Brainz That’s a big one. Right? That’s the job of Picard. But I’m working on a new engine to make all of that easier. So there’ll be an adjunct if you download a tiny little file for Picard in the future, which is looking to be about 63 gigabytes right now.
Tom Ray: Oh, Jesus.
Robert Kaye: Drastically turbocharged your tagging ability in Picard. But you know, that’s sometimes sometime for next year. The idea then behind Listen Brainz Local is something that like a navi drone, like a locally hosted music collection, you would run alongside it. and the, the real challenge is that these other sorts of services have never really embraced everything. Music Brainz Because if not all of the music, not all of the collection is tagged with Music Brainz IDs, it’s the recommendation systems cannot work. Right. M. If the metadata is loosey goosey, it just isn’t going to work. So we have to require that the collection is 100% tagged with music Brainz Right? Yeah. Drops a lot of people off and then these people won’t add these sorts of features to their own services if they don’t work for all of their users. Right. So we’re sort of taking the tough step and saying we’re working with MusicBrainz ID tagged collections and if you don’t have that, we’re sorry, you can’t play with that. Here are the tools to fix it. Right. but the idea then is shove the recommendations, shove the discovery, shove all these things into a local tool that runs on your own service. You can inspect it, you can have all the privacy you want and so forth and you can take some of the things from the cloud, like our recommendations, global recommendations, and then resolve them to your own content, your own collection. Right. Which is a lossy process. Right. Because your collection is a subset of what exists out there. Right. But at least it’s a start to start bringing recommendations and discovery tools to a local collection. And I think, and I’m hoping that’s kind of where we can pick up a little bit of more users. Because it’s incredibly difficult to actually attract people to your project these days because between shitty Google AI and so much spam, it’s really impossible to get your project noticed. Right. So one of the key things that we want to do is basically build tools that feed into other tools like these Navidromes and the Goniks and so forth of the funk whales of the world so that we can piggyback off of their communities.
Tom Ray: Right.
Robert Kaye: Because they’ll go talk to their communities and them’s like, oh wait, I can have recommendations for my funkoil. How do I do that? Right.
Tom Ray: Well, and that’s one of the reasons why I’m really into and still figuring out, but love the idea of and still learn every day, the Fediverse in itself too, where those things do talk to each other and this project fuels the other just by the fact that they’re able. You can look at one or two, follow one from each particular instance or type or you know, like I can follow the pixelfed account on my Mastodon account without having to go to pixelfed, that sort of thing, just to give an example for anyone listening who doesn’t know what it is. And that’s what you’re saying. I also wanted to ask too, with reading some of that information, are you advancing or using a new way to do stuff that you had to accomplish this in the past? Like I know you’ve had acoustic id, which I always thought was great, which was an audio fingerprint, but you also have the, the MDI MBID as well with the identifier. So are you still using those or do you, do you have a new method for doing this?
Robert Kaye: When I say, you know, musicbrainz tag to collection. It’s all about the MBID’s.
Tom Ray: Okay. Okay.
Robert Kaye: It’s all about the IDs. And we’re, I mean we’ve, we’ve never wavered from that. That is our, that is our magic sauce. By the way, acoustic ID was never. Our project was created by a friend of our ours. But, oh, they’re just, you know, that’s what he wanted to do and I didn’t want to do it. So he created his own project and it’s tightly integrated into music Brainz and everybody’s happy.
Tom Ray: Okay.
Robert Kaye: Yeah. All right.
Tom Ray: So I just, I just remember when I heard about those, I was like, that’s an interesting idea that it takes an audio fingerprint. Okay, continue.
Robert Kaye: Now here’s something I want to, Want to, want to throw out to you and to, to your listeners, which is an emergent pattern that I cannot wait. And I’m actively working towards this emergent pattern because it’s Going to be good.
Tom Ray: Okay.
Robert Kaye: Really good. So, so, so you’re familiar when I just said funk whale, you, you nodded. You know has the feature of pods and then you can join pods together. Yes. Yeah. Okay. I’m not a fan of funkwhale to be honest. Nice try. But the thing is that Fediverse model they’ve got is that ah, I can choose to basically join another pod and if I’m joining the RAA infiltrator pod, that’s going to go bad for everyone. Right. So security model. I don’t really like this one. The better security model that I start seeing emerging is one where everybody’s got their music collection on Subsonic Interface. Right. That’s the interface that that’s under Navidrome and Funkoil and all these other ones. Right. So if you have your collection online, there’s an API for it. Now you can start plugging it into different tools. Right. So if you’ve got a collection and I’ve got a collection and my team’s got a collection and then we plug all of these sources into one tool, into one multi source aware tool for music.
Robert Kaye: I created my own Spotify. It’s not nearly as expansive. Right. But it’s a, it’s a streaming service because I’m pretty happy with the streaming, with the stuff I stream off of my own music collection. Now let’s, let’s say I add another five or six, music mavens to that. Right. And you know, now we have access to their collections. Right.
Tom Ray: And then you’re saying you’d be able to join different databases, different music databases of different friends and you could all stream from the same thing and then.
Robert Kaye: Recommend into the whole collection. Okay, the whole collection.
Tom Ray: That would be the key. That’s what I was going to say. being able to add to it even past those databases would be the key being able to update it. Now what about. Okay, think of it this way. Let’s get into this. I have some other questions now. Let’s go back to that recommendation sort of thing. I mean that’s really what it is. It’s like I have a music collection too, but sometimes I’m like, I don’t know what I want to listen to today. I can put it on shuffle and it’s like, Jesus, it keeps giving me the same songs that I shuffle over and over again. so it’s like I want to pick by category, I want to. Or I just give me something that sounds like it. Even if I don’t have it. Go. You may like this. So go get it. You know, that sort of thing. Or go find it. Or here’s where it is. would there be something like that? Are you talking about that? Where it’s just going to be like, oh, listen to rock today, or create a playlist based on what the genre is? Would there be additives like that that you’re talk about?
Robert Kaye: They will be listening by color.
Tom Ray: Yeah, there you go. No, I like that.
Robert Kaye: Listen Brainz has, this feature right now. It needs some refreshing and so forth, right? Choose a color and it’ll just pull up all the COVID art that matches that color as a predominant color. You can start clicking on stuff and listening to stuff, right? So. So all turquoise releases from 1984, Gimme. Yeah, totally reasonable question. I don’t know what I’ll get, but who knows, right? Some people might really enjoy listening to that sort of stuff, right? I want to be able to, get some other ideas of what I’m talking about. Listen, Brainz also has a concept called LB Radio. It’s a very geeky sort of engine that lets you make playlists, right? So you can give a seed artist and say, also include similar artists and another seed artist, but don’t include similar artists from that. And give me some Tag Radio and some of the stats, my favorite things that I played last year, and mix it all into one thing and play. Yeah, you can write a query, a prompt for that now and it’ll do all of that for you. Right? So that stuff exists already and I’m building this stuff to now work on local. Right? So the music industry has this idiom and basically says discovery doesn’t pay. Which is why these discovery tools have, you know, either never been created or have fallen by the wayside, right? Imagine if Spotify had some really incredible tools, that would appeal, to nerds like you and I that will go deep into the music, right? That would be fantastic, right? But it would be huge amount of work. And the consensus is that you can’t make money with that. Nobody cares about that except for you and I. But you and I are statistically insignificant. So it doesn’t matter. You can’t make money off of that, right? So, that pisses me off to, to no end. Because, you know, here we are, we’re sitting here and Spotify keeps shoving the same on our faces, right? I mean, I still use Spotify, but now I can use Spotify with Listen Brainz So Listen Brainz will actually do, do Try and do best job as possible to give you new stuff without bias. Right. And because we’re not trying to sell you anything, all we want you to do is to actually get to the music that you want to listen to. Right. M. And you know, the, the, the we have daily jams and weekly jams and weekly explorations that are, you know, discover weekly sort of things. Right. And if you listen to a track yesterday and that playlist suggests you should listen to that track. You’re not going to get that track because that would be overplaying that track. Right. So every chief complaint by everyone here is just like, hey, you’re overplaying this. Right? Yeah, yeah. And so I’m. One of the things I’m working on is to make sure that we don’t overplay things. Right. Okay. And because the overplaying is a financial motivation on Spotify’s part because like hey, this track we get, you know, we have to pay low, low royalties on it or something like that. So we’re going to favor this track. Right. Which is just the, the evilness of Spotify through and through. And that’s how the money influences the recommendations they’ve got going on. Yeah. and we don’t do any of that. Right. all of our stuff, go read the source code, go look at it. we’re trying to be as transparent as possible. We’re participating in a streaming transparency project here in the EU and so forth. So we’re really trying to be, you know, the, the good people that are not that are actually trying to help you understand music, collection your own music taste and help you find more of what you want and then connect to people, people and so forth. That is honestly our goal. And luckily since we’re a non profit, we function on a completely different economic model than everyone else does.
Tom Ray: Right.
Robert Kaye: So I mean our team is nine people big with I don’t know, six and a half full time equivalent, which is nothing for what we’re producing on the grand scheme of things. Absolutely nothing. It’s a round off error on Google’s daily operations. Right. so what we’re able to do with all, with all of this stuff is, you know, we have a very small thing, we’d love to have a larger team but the licenses coming in for commercial use, covering for MusicBrainz data is only so large and it’s not something that I can choose to expand or contract the market. Right. Like people either want to use the data or not and there’s not Much else I can do about that. but still taking donations from end users to. Because running, listen, Brainz Is mostly hosting cost.
Tom Ray: Right, right.
Robert Kaye: Hosting costs. Like, okay, servers are not that expensive anymore. Now if I can put, I don’t know, 4,000 users on a $50 server a month, you know, now you can figure out what the per user per month cost is and then make sure that the donations roll into that amount. Okay, we’re cool, but we’ll keep building stuff. Yeah. So it’s a really neat position to be in, to be running a, I call it a business running inside of a non profit. We don’t really go out asking for money. Right. So all we ever do is we basically run this database for the public, for the public to download and then we take commercial support payments from people, donations from people, and that’s it. Right. The irony about all of this and we haven’t talked business. I’ll get into business for just a second.
Tom Ray: I was, I was kind of going to transition into that because you’re raising some questions where it’s like, I do want to know more about that. Yeah. Okay, continue.
Robert Kaye: and this is Tongue in Cheek. If you look at it too closely, it falls apart a little bit. But the Tongue in cheek version it describes our business model is we’re 100% profit, non profit, licensing free shit. We don’t own right. So the 100% profit part is the weirdest part because we run a non profit and we do everything for the people. Right. And then we say, oh, hey companies, if you only use stuff, just download it there. And we don’t have any incremental costs for an extra download, so the costs are zero. Right. So when a company gives us money, we do nothing. Right. We have the infinitesimally small accounting overhead where I have to send them an invoice and account for payments and like that. Right. That’s the, that’s the, that’s not 100% profit, but it’s like 99.99 or something. It’s, it’s might as well be 100. Right?
Tom Ray: Yeah.
Robert Kaye: So but we’re running that inside of the nonprofit and you know, biting by all the terms of the IRS so far, not really given us any grief for what we’re doing. And and then, you know, licensing free shit, it’s all, it’s all public domain Data. It’s all CC0 on the whole. Right. licensing free shit we don’t own. Right. We don’t own any of this. Stuff, it’s just preposterous. So all we really do is we’re the curators. I already described where the real value comes from. Right? The value comes from the enormous community of people that are making changes to the database. Right? But if you are an MBA and you come at this with your, I don’t know, whatever evil business models that try and instill in your head in an mba, this makes no sense, right? This makes no sense. It will never work. Right. And it’s like, you know, here’s our, here’s our balance sheet and here’s our profit loss. It’ll never work. No, that’s our actual profit and loss. Like we’re making money with us, we’re not getting rich, but we’re making money. It’s never gonna work. right, right. So it’s, it’s very weird, but it’s kind of fundamental. And then when I look at how to run the foundation, I don’t want to run a foundation.
Tom Ray: The foundation, is that Meta Brainz? Is that the foundation? Okay, Yeah.
Robert Kaye: I want to manage a team. I want to do technical things. Right. I don’t want to run the fund, the foundation, but somebody’s got to do it. So I’ve written a lot of software, not surprising to basically as much as possible. Right. So a lot of things are automated and so you know, at the end of the day there isn’t a whole lot of stuff that needs to be done, but somebody still needs to, needs to see, after it all and you know, deal with the accountants and the lawyers and everything else and in between, and that’s me, but you know, so the other unofficial title is that I’m in umbrella. So whatever comes from the outside world, it hits me and I protect my team from it. Right. So lawsuit that’s happened. you know, really how to keep the lights on all of that stuff. Yeah, yeah. The lawsuit wasn’t particularly fun. if you want to go into details on that, I recommend, Corey has a boing boing article on that. Corey Doctorow Prof. Yeah. Search, search matter, Boing boing Meta Brainz lawsuit. You’ll find it.
Tom Ray: I may have even read it. But it, I’m assuming it has something to do with, as it always does. Even though you are not hosting any music, you’re not doing anything that involves pirating music or anything like that, but you’re hosting information about music and they’re like, well, you have the name of our artist on your database that’s stealing or Something stupid like that is what.
Robert Kaye: I’m assuming is what it’s probably about. What you’re assuming is the, you know, the weekly fraudulent DMCA takedown requests that we get. Yeah. Everybody’s like, oh, no, you’re downloading your. No, there’s, there’s, there’s no download links. Right.
Tom Ray: Yeah.
Robert Kaye: Just telling people that their DMCA request is fraudulent is wrong. Is a huge amount of effort. But no, actually what happened here was much more evil. And I, completely did not see this coming. So Wikimedia Commons hosts a lot of images for artists. So these are photographers of different types that will upload pictures of artists and then put them under Creative Commons licenses. What they don’t really tell you is that they have their own criteria that they tack onto that about how Creative Commons attribution should be given.
Tom Ray: Yes. Most people do it as a drive by, as what they think create Creative Commons means you can use it for free is what people think. And that’s. It’s true to a point.
Robert Kaye: But this photographer in particular in question, was basically, is like, no, you need to have attribution of my name in 150 point font, bold print, blink, otherwise proper accreditation.
Tom Ray: Okay.
Robert Kaye: All right. so I mean, what this guy does basically is he rasses anybody that uses his images and just basically send me $10,000 and this problem goes away. Right. And so he sued us and we’re like, no, we just ignored him as we were, as our board advised us, ignore him. Like, these people go away. This one didn’t go away. He sued us. Right. We’re lucky. We have a guardian angel, at Cavazos, at, I forget the firm that he’s with. but he’s, he’s been our, guardian angel for many years. And he basically went to his firm, said, hey, can we defend these guys pro bono? And they said, sure. So then they went pretty much on the offensive and subpoenaed both Wikimedia.
Robert Kaye: Wikimedia and Digital and Creative Commons and a, discovery of information quest to figure out, like, what the. Is going on here. That Wikimedia foundation is entirely happy to leave, copyright trolls operating on their platform, and they have no enforcement against copyright trolls. M. I’m so pissed. I’m still pissed off by that. I will never contribute another penny to Wikimedia Foundation, I’ll tell you that.
Tom Ray: Okay.
Robert Kaye: And we dug deep and yeah, and the basic policy is like, we don’t want to do any enforcement in our community. And if somebody gets sued. Yeah, whatever.
Tom Ray: Right. So our platform is to for free use Creative Commons licensing attribution. But you know, we’ll be over here, let us know if you have any trouble and we’ll go. Sorry, I’m m busy sort of thing. That’s interesting. Okay, right.
Robert Kaye: And when you get sued, through the Creative Commons and you, and you try and actually talk to the creative comment through Wikimedia comments and so forth, guess how much. Like can you pick up the phone and talk to them? Of course not. Right, right. Do you get any help from them? No. Anybody that works for Wikimedia, when the word lawsuit comes, they’re instructed to just stay mum and the conversation stop. So lawyers figured this out real quick and basically just, you know, subpoenaed them. Right. Which really woke up Wikimedia and creative comments. They were just like, holy, what are you doing? Like we have a good legal team. In the end, they basically hit the photographer so hard that he basically just like, oh, I’m tapped out. No, no, no, no, no, I want to drop this. And then just dropped it. Right. But then good.
Tom Ray: I mean it’s essentially like it’s, it’s an individual but yeah, that’s like shutting down an entire spammer farm, you know, or telemarketer farm. It’s just like dig that. Yeah.
Robert Kaye: So, so the guy basically went away and everything was, was resolved. And I don’t know if this is still the case or not. I haven’t looked in a while. But at that point the Wikipedia page for a copyright troll listed Philpott was the, was the plaintiff’s name, Barry or Larry Philpott as a copyright troll example, who was operating Wikimedia comments. I thought that was really brilliant. So that really blew up and blew up face. So, because Music Brainz has always played a very clean game on all fronts, we continue to have people who are willing to support us. Corey Doctorow has been on the board of directors for 21 years now. My directors have been there for, you know, the other one celebrated 15 years the other day as well. I think it was 15 years. my team has got a great tenure, the community members have great tenures and so forth. It’s all about building a non toxic place, which is incredibly difficult. Right. Especially in the Internet where everyone just wants to fling poo at each other. but it’s really critical that you start off with, hey, this is a place we’re respectful for each to, towards each other with a Continual reinforcement of that message every single time that somebody goes off the rails and then empowering your community to do the same. Right. And then when somebody else comes in and it goes off the rails, they feel empowered to basically continue the line about how we’re respectful for it towards each other. Yeah, it doesn’t always work, but we have a community manager who people can report people to and then they get thrown out of the community domain. Basically keep the peace. Right. And if something starts doing something illegal, oh man, they get tossed real fast. We don’t tolerate that sort of stuff. Right. So there’s a difference between having a completely uncontrollable community and that’s Wikimedia and having a community that is not controlled. But you, you basically want to see, you want to carefully set up your, your guidelines, not rules, guidelines for behavior. And anybody that’s too far into outlier just needs to get chopped off.
Tom Ray: Yeah, right. Well, so on, on top of that, I want to add too. So you were talking before about how you know, the different levels that there are to maintain, even though there’s not a lot of people there, or it’s not even a huge organization by, you know, comparative standards to a corporation, but it still is doing its work, making a profit. As you were saying that. Here’s the really weird thing, and it’s funny that you are involved in the music side and then me being a musician, like I’m kind of going, as you’re mentioning, you have to do this to do this. And it’s not just, it’s more than one thing. You do multiple things. And that’s one of the. Being a Creative Commons open source band ourselves, like I’ve set it up so that we are able to run. I mean we essentially just pay for our domain each year, you know, and maybe like, I don’t know, like a dollar or something of server costs for And we even, you know, use open source operating systems, open source recording systems. Like I’ve maintained it so we have the smallest overhead possible, which I’m sure you’ve done too. But also on top of that, you can’t just go. And then that’s how we’ll make the profit. It’s like, no, it doesn’t work like that. So you have to think of other things that are related to like this stuff behind me. I, I like collecting vintage stuff. So I sell it so that way I can fulfill my need to buy, collect it and justify buying this stuff. But then I sell it to make a profit as well now getting into that just, it was, I was just seeing how like man, it’s almost like musicians are, are, non profit organizations, ironically. But so explain to me then how with Listen Brainz and being able to use it, and you mentioned that Music Brainz is inside of so many things and that’s kind of where I want to get at with this whole thing today is like, so what, what is the organization as itself that keeps it alive, keeps it running, keeps it you know, free from just being drowned in cost? Like, what would you say is the key to all of it?
Robert Kaye: Me.
Tom Ray: Okay, all right, I like that answer.
Robert Kaye: I mean, I, I, it’s a terrible bus factor answer. Right, And I’m very well aware of that. But fundamentally it’s, we just have this, this philosophy of like if we don’t have to spend money, don’t spend money.
Tom Ray: Right, exactly.
Robert Kaye: Yeah. Nobody in the company has authority to spend money except myself and the system administrator who doesn’t have access to the purse strings. But if he needs to order a new server because reasons he can order a new server. Right. That has impact. But yeah, and the whole idea is just how can we make it as lean as possible? So how many assets do you think we have on the books?
Tom Ray: I mean, probably I’m going to think a lot.
Robert Kaye: I don’t know.
Tom Ray: Zero.
Robert Kaye: Okay, Right. We have an office in Barcelona. But like none of the assets. And it’s like there’s furniture, there’s. Right. Okay. Technically the organization pays for my computer. Right. So I guess that one computer, that one PC that’s going to depreciate immediately is our app. Right. It’s not worth accounting. Right. It’s just, it’s. We have no assets. Right. We don’t want any service. We have nothing. We have cash in the bank. Right, that’s fine. but so if you don’t have assets. And then for instance, so we needed to move from the hosting facility we had in California to we wanted to move to somewhere in Europe and we needed to go in the cloud, as it were. Right. This was now 10 years ago. And we did an exhaustive study of three providers, one plus one in Germany, Hetzner in Germany and Amazon. Of course. Everyone was basically like, I’ll just go to AWS M and be done with it. Right. Like, wait a minute, we just published a long blog article that said we’re going to Hetzner because AWS is five times as expensive as HeadSmart. Hm. And people didn’t get that. Right. But we just figured out, like, how can we do this as cheaply as possible? And this ethos is, you know, I mean, sometimes you need to spend money. And we spend $15,000 in the annual summit where we fly the entire team either to India or to Barcelona. Right. You do need to spend money on these things, but you can be really careful about where you spend your money. Right. So rather than spending useless money on aws, we went with the, German company Hetzner. And our experience with Hetzner has been the strangest thing ever. So strange. We started off with, I don’t know, eight or nine servers, and our bill was just under €2,000. And then we expanded a little bit, and then we got the €2,000. And for the last eight or so years, our, bill has been around €2,000, and we’re now with 50 servers. Wow. I. That wouldn’t have happened AWS, I tell you that. No, it’s been. I mean, the cost of their services has been so drastically shrinking that, you know, we can just look at it like, oh, we’ve had the server for two years. Let’s just swap out the server. Not because we have any problems with it, but we can swap out the server and get a cheaper one that’s more powerful.
Tom Ray: Yeah.
Robert Kaye: So we. And since we’re a nonprofit, and this is. So we are a business running inside of a nonprofit. But then you can still. We have the 501C3 classification. So we can go and say to Sentry, hey, can you give us a free account? GitHub, can you give us a free account? we get all these free accounts from people, right? So our services. One of the m most important, expensive services that we pay for on an ongoing basis that isn’t hosting is our stupid QuickBooks account. That is. Right. That’s just dumb. let’s not make $800 for that piece of shit software. Right? But our accountants say we must do it, and our accountant’s keeping us out of jail, so. Right. QuickBooks a day. But, you know, this whole idea of not spending money if you don’t have to, it just pervades, and it becomes a point of pride for us. Right? And, yeah, no, we. We do not have. you know, I managed to buy a flat here in Barcelona by hook or by crooked. It was tough. Right. this is not luxurious, but I want people to be paid fairly. Right. you know, so I have to be aware of, like, hey, what’s inflation in India doing this year? I have to Know all these things where my team members are so I can actually have a fair conversation about making sure the team is fairly compensated, right? And the second most important thing I do is to actually get the fuck out of my way of my teammates, right? Let them do their job. Mostly because I’m lazy, right? But let them go do their jobs, and then they’ll go do their jobs. Pay them well. Let them do their jobs is a magic combo, right? Then you throw on top of it. It’s like, hey, look, we’re trying to spend as little money as possible because we want to do all these cool things, but what I can give you is a high quality of life. And people like, yeah, okay, okay, I can sign up with that, right? So we’re all more or less of this, you know, have you heard of the. The expression, enough, I have enough, right? You know, I have a flat, I. I have what I need in Barcelona. I don’t need to become rich. I don’t need this. I have enough, right? Yeah. And life is more important, you know, to go out, spend time with friends. And, you know, I live in Spain, where that is really important, right? And I’m here for a reason, for that reason, right? Because I really, you know, I consider my move to Spain to be semi retirement. I’m still working, working a lot of hours, but, it’s. It’s. Life has just become a lot easier, right? And, you know, the, the whole idea is if something’s not working for us, we’ll stop doing it, right? And then, you know, let’s just not spend any money and let’s be nice to each other. And somebody says, like, oh, we have a team meeting tomorrow. And somebody says, oh, I can’t because I have a concert. We’re all like, yeah, you’re going to concert, right? You just got to have the right attitude, right?
Tom Ray: Yeah.
Robert Kaye: And give your team the space to do the things. And you can do amazing things. Absolutely amazing things. All right? I feel like I’m really babbling, but, you know, maybe. No, no, no.
Tom Ray: I mean, that’s all very positive, and that’s really what we all hope for when we want to imagine going into a business of our own or even be part of a business that is doing something we love. And that’s really just, you know, hearing that or the way that you’re explaining, it’s like, yeah, that’s the way we all picture it. Like, even myself, I’m like, I just want to, I don’t want to go, like, I Want to get a big mansion and fancy cars. It’s like, no, I just want to do what I want to do and not have to worry about paying for the way I am now. You know, just sustainable. I mean, that’s all it really is. And sustainable really is the way to be. So and so, one more thing I want to ask you before we go today. You mentioned the, the projects that you’ve been working on, the ideas that you have now. What, is there a timeline for when these are coming out? You were talking about sort of the music server, sort of the, the idea, that just what you’ve been working on, the, the, the hinting at the projects that aren’t available yet. When it, when do you think we might be able to expect something like this?
Robert Kaye: Early next year.
Tom Ray: Really? That soon?
Robert Kaye: Well, it’s been in the works for a couple years now.
Tom Ray: Okay.
Robert Kaye: Sort of like, you know, everybody’s got their own pet projects, right? And this is the pet project I’ve been working on. But the project I really need to work on is Listen Brainz And then really complicated. but, that’s a scalability issue, right? We can see, it’s like, okay, we can take on another 10,000 users and then we need to get another server or replace this piece of software with something better. And we wanted, we want something better. So I’ve been working on that for months now, but that’s top priority. When I get that one out, then I’m going to go to the Walk in the park, the Listen Brainz local, and just pound out some really cool things. I mean, it’s a point where people can start playing with it, but it’s very rough, right? Like if another engineer wants to pick it up, I’d be very happy for that. But, really, And I don’t celebrate Christmas per se very much, so I celebrate the fact that it’s going to be quiet. So I’m just going to go, you know, have a week to myself and then recycling a lot, cook some good food and then just hack the rest of the time and build stuff. But I mean, that should sound so.
Tom Ray: Sad, but it doesn’t. It seems so. It’s so pleasant.
Robert Kaye: It’s marvelous. Right? As long as you can get over the fact that like. Well, I mean, I am not enamored with Christmas, right? I was when I was a child. But you know, nowadays just see the stressors and all the capitalistic motivations behind it. So easy for me.
Tom Ray: Well, I want to thank you so much for talking with me today. This has been great. And I really have been looking. I mean, essentially, I’ve been using your stuff for years. So I love the fact that I finally get to meet you. So this has been fantastic.
Robert Kaye: Well, thank you very much for the invitation. It’s been a very fun conversation.