Kevin Nelson is a web and mobile software engineer based in Berlin. His current clients include the music streaming service Rdio and the music licensing and distribution company Loudr.

Kevin Nelson Interview

Kevin shares his thoughts on music tech scenes, choice paralysis, and the future of digital music. Note: Because he works as a contractor for music companies including Rdio, his opinions below are his own & don’t reflect Rdio’s or any other company’s opinions.

After The Show: How does Berlin compare to San Francisco in terms of each city’s music tech scenes?

Kevin Nelson: Interesting question. Music Tech is an interesting space concept to me, because lately it seems to have two main branches, music consumption tech and music creation tech, and I’m not so convinced that they intermingle very well. I also think it’s important to recognize the importance of the music scene in general when thinking about the music tech scene. If you’re trying to bootstrap a music tech scene in a city that doesn’t have a vibrant music scene, you’ll either flounder, or wind up creating a vibrant music scene in the process.

Historically, music consumption tech has been borne of commercialism and economics. Investors, producers, label executives, and the rest of the money people all want to innovate on technology so that they can sell more sound waves. First it was the phonograph record, then the variety of tape recordings, the compact disc, and the implosion of file sharing.

From what I’ve seen, Berlin doesn’t have a huge community of these sorts of people. Obviously there are some labels and business folks here, but the word “producer” in Berlin means more the conceptual artist or director rather than the executive responsible for footing the bill.

Meanwhile, in San Francisco, amidst the gold rush of TCP/IP venture capital, it’s sales, eyeballs, and advertisements that drive the train, so you have business- and growth-minded people building tools and media that they think will propel Los Angeles, Nashville, and New York into the next generation of music monetization.

The other branch of the tree, music creation tech, is the extended evolution of recording studio innovations. Transducers, signal mixers, magnetic tape. This is where all of the excitement of new music lives, the branch of the tree where you see a couple of out Philadelphia invent a new kind of music controller that sparks a slew of imitations from bigger companies. Or where you see recording artists release the iPhone app they’ve developed to record and use live. To me, this is where the excitement and joy is…

Berlin is definitely much more focused on the creative branch, which you can see in the companies that thrive here. Ableton and Native Instruments are both music creation companies, and even though it’s changed a lot over the years, Soundcloud was also primarily focused on creation when it launched. Of course, the companies don’t define the scene, but they’re a manifestation of its vibrance.

Berlin’s also a less expensive place to live than San Francisco, so I finally have time to actually make music again, which might also color my perspective a bit.

Sometimes when I open Spotify, the nearly unlimited listening options are too overwhelming, so I close it and open my more manageable iTunes library. Besides offering playlists and recommended/similar songs (“You listened to Alvvays so you might like TOPS”), how can music tech companies combat choice paralysis?

I think this is actually the hardest problem that music streaming services are facing right now, and I don’t think anyone has come anywhere close to being successful. Choice paralysis is a huge issue when you can listen to (almost) anything in the world. This is why you see services like Pandora and Slacker take off — most people don’t want to think that hard about putting on music. They just want to push play.

Ultimately, I think this is actually the service that companies like Rdio and Spotify are selling, and I think that if it’s done correctly, it could be really remarkably great. I used to have a collection of hundreds of CDs in high school, and I found myself falling into a similar problem. The choice paralysis isn’t as strong, but it’s still there. Picking from the 300 CDs on the shelf is a lot harder than picking from the 3 in your disc changer.

That said, I don’t think anyone’s doing a good job of it right now — actually, I think everyone is doing a pretty terrible job of it (Sorry, friends at the Echo Nest and at Rdio, but it’s true). This is, of course, in the context of a range. What the Echonest does is much better than pure randomness, and it’s certainly passable, but it’s not something I would call “good” by radio DJ standards. It’s also an incredibly difficult problem to solve, with way more complexities and nuances than you might expect, so I don’t mean to say that the folks working on these problems are slacking off or not talented — just that the systems aren’t there yet.

If I may step off of that soapbox for a moment, and onto another one, the iTunes-style Library feature is something that Rdio has had forever, and when I started working there it was one of the biggest features that drew me into the product. You can create a virtual music collection (now called “Favorites”) without having to stuff albums into playlists in the way that Spotify forced you to do. Of course, Spotify eventually caught up and introduced their Library feature within the last year, but it still does completely idiotic things like preventing you from organizing the old iTunes style list of songs by “Artist-then-Album-then-Track Number”. Instead, it sorts alphabetically as the second tier by default, which is not how anyone ever thinks about looking at their music collection. It’s super dumb.

Rdio has also been working on their Home feature for a while, which presents a Facebook News Feed style list of activity from your Rdio social network: comments on playlists, trending albums, songs, and stations, more music by artists you’ve been listening to, etc. It’s a huge first step, but there’s also a lot of noise if you don’t follow the right people.

At the end of the day though, I don’t think an algorithm will ever make as pointed of a recommendation as a good friend who knows your tastes. Someone saying “Hey, I think you’ll like this because it meets my standards, and it sounds like these other things that I know that you’ll like” is a much stronger signal than an algorithm saying “Hey, you should listen to this 2 AMG star album by the same artist that released that 4.5 star album that you were listening to yesterday”, and you’re much more likely to enjoy the recommendation.

How do you like being an independent contractor vs a full-time/on-location employee?

This is my second big round of independent contracting, and each option has its ups and downs. The first time around, I struggled a lot to find clients and work, and struggled a lot with trying to charge what I thought was a reasonable rate. I feel really lucky that I found the clients that I did, and got to work on the projects that I did.

In general, the work of finding clients, and making sure they actually pay you is really challenging, and not usually something you have to think about when you’re working as a full time employee. I’m lucky right now, though, since I’m still doing regular contract work for a couple of clients that are great to work with.

Depending on who your clients are, or where you work, there are other pros and cons. It’s nice to work from home, for example, but living with a partner who also works from home can be stressful. The flip side of that coin is that it can also be lonely and isolating. If your client doesn’t force you to be onsite, you’re not going to have any water cooler (or coffee robot) conversations unless you get creative with where you work. It can also feel aimless and unfulfilling to constantly be doing hired-gun work, but you might also have a similar existential issue if you find yourself in a full time job that you don’t particularly enjoy.

The positive sides are really nice though: Since I work remotely and in a far-away timezone anyways, I also have the luxury of taking long lunches when I please. Usually that means I have to work later into the evening, but that also works well with the timezone thing, since my clients are awake and at work during my evenings.

The biggest two perks, though, are 1) that I was able to move to Berlin, and 2) that I’m able to schedule personal work time to focus on things like actually making music again, or working on my own app projects. I’m sure it’s possible to find a full time job that accommodates those things somewhere in the world, but those negotiations and discussions are easier when you get to be in charge of yourself.

I’ve also discovered that billing 8 hours in a day is a lot more work than showing up at a full time job and working in the office for 8 or 9 hours. I’m now pretty convinced that the average 8 hour work day only really consists of 4 or 5 hours of actual work.

How do software engineers communicate effectively with non-tech people at a music tech company? Is there a department that serves as the bridge/translator between the technical and non-technical sides?

I don’t think there’s anything particularly special about music tech here. Any company with an engineering department is going to have these kinds of stereotypical “engineering vs other department” issues. With music streaming, you have a content licensing department, but with music creation, you just have the normal company set of things. Product design, marketing, sales, analytics.

Maybe music tech companies attract engineers who are more interdisciplinary, but I think other companies can attract these kinds of people as well. It also depends on company size. If you’re big enough to have middle management, then you need to make sure that the people you put into those management roles can bridge any communication gaps between their team and other teams.

How do you envision the future of digital music, particularly music discovery and streaming? What frontiers do you think we’ll push past?

This is a hard question for me to answer for a couple of reasons, but let’s just say I’m not incredibly optimistic. To start with, streaming is not currently an exhaustive listening experience, and I don’t think it will ever be one. You can’t stream Queen Crescent on Spotify. You also can’t stream the Beatles, but the point of using a band that you (hopefully?) haven’t heard of is that unless a service aggregates literally everything, the scope of discoveries that it can provide will be limited.

Today, for instance, I discovered Breaking Circus for the first time, but it was because I was reading about related bands on Wikipedia and clicked on some links, not because of some algorithmically curated radio station. Their music isn’t on any streaming service because the label it was released on is now defunct, and no one’s tried to recover the rights to redistribute the content. It’s also not for sale in iTunes. Will Spotify or Rdio (or Beats/Apple, or Tidal) ever be able to solve that problem? Probably not, unless the major labels and copyright restrictions magically disappear…

At the end of the day, though, streaming definitely has changed the way we think about obtaining music in the same way that downloading did at the turn of the century. We didn’t have pocket computers back then, so it makes sense that with this new technology we have a new distribution mentality. The commercial services, locked into unfavorable deals for eternity, will do their best to stay afloat, so they’ll try to encourage lock-in: only listen to music in Spotify. This works for everyone who listened to terrestrial radio. For the music nerds, the best place for music discovery is the same place that it’s been all along: small (online) communities of friends and people with similar tastes.

Thanks Kevin! Keep up with Kevin’s projects at his website + Linkedin.