mp3.com

“Those who cannot remember the past are condemned to repeat it.” – George Santayana

As someone who does a lot of work in the area of research, I say that one of the most common mistake startup people make in this regard is harboring the assumption that the internet “has everything”, or “everything that matters”. Having worked on digitization projects within library systems before, I can attest to the fact that what we see on the internet now is only a tiny fraction of what has been published, and not everything that’s being published right now gets digitized right away. With enough time (and incentives) we may reach a point where the world becomes “fully-digital”, but we’re not at a point where that claim can be made yet — not even close.

Librarians are working night and day trying to make their archived content digitized and available to the public, but due to the bad economy and budget cuts these efforts have become severely hampered in recent years. The unfortunate thing about this scenario is that the items that are available for public viewing often go unnoticed because they don’t tend to rank well on search engine systems. A lot of the times it’s just a matter of going to Worldcat or the library’s website in order to get what you need, but because libraries don’t usually have an allocated marketing budget these items never see the light of day. (Even if the quality of its content is very high.) If you can buck the trend and diversify your sources of information, though, you’re much more likely to get a better balanced perspective than what can be achieved on Google search alone.

In the same vein, it’s dangerous to assume that all of the musical talent in the world has already been “digitized” and uploaded onto the net because it might cause you to completely ignore the community of overlooked and under-appreciated artists who simply don’t have the means or methods of making themselves a known entity on the web. Music ventures that are interested in cultivating independent musics should especially be aware of this, because they’re inherently reliant on emerging and new talents in order to draw more people toward their site/app. And the talent pool in this regard is huge, if you consider some of the simple facts above. How would you entice these artists — many of whom already have busy schedules teaching, performing, composing and rehearsing their current projects — to come and use your product on a regular basis?

The Story of Mp3.com

Here’s a little bit of perspective that might be of some help to some — a story about the music tech industry that has largely been pushed to the waysides of history, dug up by the Internet Archive’s Wayback Machine. The rise and fall of Mp3.com.

During the turn of the millenium a daring new music website had emerged as the next contender to the record industry: it acquired the mp3.com domain and adopted it as part of its company name. Mp3.com’s CEO, Michael Robertson, envisioned an environment where independent artists could gather, share, and express themselves through their music, using the world wide web as its primary platform.

mp3.com archive image

Archive image of mp3.com via Internet Archive’s Wayback Machine.

The look of the site seems somewhat dated now, but if you read some of the links that are on there, it had most, if not all of the features that many startup companies might tout today as being “revolutionary” or “game changing”. There were streaming services, genre rankings, advertising/promotional tools, hosting tools, licensing tools, developer APIs, radio stations, CD publishing tools, merchandise, and so on and so forth. There’s even a subsection for the women of MP3.com listed there — these gestures are pretty common now, but for a tech company of its time that was fairly unheard of.

When the site was made available to the public, thousands of indie artists flocked, enamored by the potential and possibilities that it had to offer. At its peak, Mp3.com was hosting millions of songs, getting thousands of streams and hits every day, being featured in major news and media outlets as the “next big thing”. At one point, even Alanis Morissette decided to be a major investor in the company, convinced that the idea would work.

But more importantly, it had a devout, highly impassioned community of artists who really believed in what the site had to offer, participating and giving feedback to the developers in any way they could. In 1999 they introduced a “pay-per-play” system where artists were paid every time the song was streamed, no questions asked. Compared to Spotify and YouTube’s partnership programs that exist now, the pay rate was quite generous and this kept many artists interested in keeping up with the company’s progress and developments. This system gave artists a glimmer of hope that making a living through alternative channels was in fact a real possibility.

Unfortunately, the experiment proved to be a fatal mistake, eventually leading to the demise of the company as a whole. Due to intense competition and pressures to get to the top of the charts, artists began gaming the system, using playbots, “play posses”, and hacking tools in order to artificially inflate their royalty and ranking numbers. Honest acts were robbed of their efforts, while the integrity of chart rankings gradually eroded into nothingness — the developers did their best to curb people’s cheating, but in the end their efforts were proven to be futile.

Two years later, the financially weakened company was acquired by Vivendi/Universal for about $5 a share, down from its $60 peak. The company removed the pay-per-play system entirely, then immediately started to evict independent artists off the site, eventually replacing all of them with major label acts. Having lost its support from both its founder and its parent company, the site was eventually shut down and sat idle for years on end, largely forgotten by the public at large. Michael Robertson now seems to be tweeting about conservative politics and supporting Republican candidates whenever he has the time — maybe this could have something to do with his experiences having witnessed some of the ugliest parts of human nature during his time and tenure as Mp3.com’s CEO.

- – -

In recent years many musicians have voiced concern over the low royalty rates that come from services like Spotify or Last.fm, but this story might give an explanation why companies now are very wary of aggressively adopting the pay-per-play model. Google spends an insane amount of effort “cheat-proofing” their search algorithms, because they know that if the trust in the company disappears, it won’t be long before everything else starts crumbling down. When money becomes involved, efforts on all sides become much more intensified (including legal issues), making the system much more unstable and complicated. So the first lesson to be learned here is: fairness matters, as does the security needed in order to maintain its balance.

Mp3.com was an example of a tech company that was closer to being a “mining excavation” — i.e. a company that was interested in the development and cultivation of artistic talent. By paying its artists directly, it burdened the risk of content creation, in hopes that new and innovative content will attract and retain more visitors to the site. The project was ambitious, fair-minded, progressive, and way ahead of its time — nonetheless, it failed because the model was proven to be unsustainable in the long run. And the most unfortunate part about its downfall was that it was the artists themselves, the people Mp3.com was trying to help the most, who sabotaged their own opportunity of being emancipated from the monopoly of the record industry.

Figuring out the reasons why this happened might give a glimpse into how music-based ventures might be better-handled in future and upcoming projects. Engineers often pride themselves on their “objectivity”, thinking that if they create a system with all of the right rules and formulas, the cream of the crop will rise to the top “naturally”. But in recent years tech companies (including Google, who has been consulting Hollywood executives as of the late) have become interested in the idea of curation, using editorial, rather than autonomous, processes in order to drive content towards its intended viewers.

The idea here is that you can’t be everything to everyone — in order to get people what they need, there’s no avoiding having a selection process for the types of artists you’re willing to work with or reward. Unfortunately for engineering purists, this means having to abandon the quest for the holy grail, the “perfect” algorithm for creating the ideal democratic world. In a lot of ways Facebook’s success was based on it being exclusive (back when it required a verifiable college-based email) rather than inclusive, which gave the company an identity and sharper sense of focus that the other social media sites weren’t able to provide.

In the same way that people became tired of the rampant anarchism of MySpace, the music world may be ready for a similar convergence of ideas and practices in its own ranks. The novelty of discovering your own band — often described as a product “feature” by music sites — has largely lost its appeal because people have now begun to see it as work rather than play. What they’re looking now is for someone to cut through all the noise and show them musicians of high caliber and quality, and in order for that to exist, there needs to be a curatorial process.

Part 3 — What do Artists Really Want?

  • McIntosh Bonthera

    Great post. I think what it comes down to is poor execution. They had a great idea but the implementation was lacking. Of course, in hindsight, we can point to cheaters being the problem but the founders really should have thought deep about the repercussions of the pay-per-play feature. Some companies are caught off guard by people who perverse the system (i.e. Chatroulette – what did you expect with strangers on camera?) but luckily, these beginnings paved the way for web 2.0 startups. 

    If artists feel that there still isn’t a viable startup that caters to their needs out today. That may be because entrepreneurs have a tendency to solve their own problems or problems they are familiar with. Perhaps we need more musicians/programmers out there in the world. 

    • http://ryangtanaka.com/ Ryan Tanaka

      I think a successful music startup will have to be lead by someone who has a deep knowledge of music, particularly in the area of underground/indie scenes.  Isn’t that whom the tech industry is going after anyway?  But yeah, there is definitely a disconnect between engineers and artists right now, but bridges are being formed as we speak.

      Looking at it from this point of view it’s pretty interesting — there’s lots of potential, but the road will be pretty rough.

  • Pingback: (¶/♪) Ryan Tanaka | Disrupting the Music Industry (Pt.3, What do Artists Really Want?)

  • Pingback: Yo Yo Ma: Music, Meaning, and Technology