Isaac Jaffe is a composer, performer, and Head of Research at TuneSat, a company that uses audio fingerprint technology to monitor TV channels and websites for unlicensed music.
After The Show: You juggle so many projects – composing music for film and TV, performing with SugarBad, playing upright bass in jazz bands around NY, teaching piano/guitar/bass, and running TuneSat’s Research Department! What’s your schedule/time breakdown like?
Isaac: My schedule tends to move around based on what’s going on, and fortunately all of the people I work with give me flexibility which is key. Most weeks I do 4 or 5 days in the office at TuneSat, leaving around 5 pm each day to teach a few music lessons. After teaching, I usually do one rehearsal and 1-3 NYC performances a week. Those are long days…
SugarBad also plays out of town every few months. This is either a long weekend or a 10 day stretch, which means I have to put my students on hold and work remotely. That can get a bit crazy, but being on tour is so much fun so I don’t mind. Working while traveling definitely wouldn’t have been an option in the days before free wifi.
Fortunately, composing music is relaxing for me so it isn’t too hard to fit this into the free time I find here and there. I live with an upright piano and two other musicians, which helps a lot! While juggling everything is a challenge, the nice thing about shifting my focus is that when I feel burnt out at the office or on the road I’m never too far from shifting gears.
That’s so cool that you were a music librarian for the 2010 Winter Olympics…What do you remember about that experience?
This was a really interesting job for me right after college. I was in charge of checking and formatting all of the cue sheets for much of the network and cable programming (cue sheets are submitted to Performing Rights Organizations like ASCAP and BMI in order to distribute royalties to composers and publishers). This entailed spending 2 or 3 months working 10 hours a day, 7 days a week in a windowless room at 30 Rock watching every hour of curling and figure skating and ensuring that the fingerprint technology was picking up all of the music…
Usually it was a matter of checking each of the instrumental underscores coming in and out of commercial or during replays, making sure that the durations and use counts were accurate. Sometimes, we would have to figure out what was being used by ear. I remember winning an argument with another librarian over whether one track was sung by Ella Fitzgerald or Billie Holiday. Occasionally, your jazz degree pays off!
You’re in charge of TuneSat’s team of music investigators – what kind of research goes into determining if a client’s music is fingerprinted and properly catalogued?
The acquisition and cataloguing process for our clients’ music is pretty straightforward – we have a fingerprinting and metadata system that keeps things pretty well organized. What is really challenging is sorting through different uses of music on the internet and find commercial uses that are likely unlicensed.
There’s just so much media online being used in advertising and being monetized on its own, through Youtube and elsewhere. It used to be that music was only transmitted on terrestrial radio and television, which only leaves a handful of frequencies that are being broadcast at once. Now, there is new media being uploaded and consumed constantly from a virtually infinite number of sources.
This is the main challenge for TuneSat – finding ways to determine the signal within the noise, the media that is really generating revenue living alongside user-generated content. Now that consumers aren’t paying for CDs and are moving away from buying downloads, sync licenses and performance revenue are becoming a more and more important part of the music industry. It’s shocking how much music is incorporated into commercial uses on the internet without a proper license. Our technology has found music that I’ve written being on jewelry advertisements and pornographic videos without my permission, and without any compensation for my work. It’s really exciting being challenged every day to build solutions that become an essential part of how musicians are paid.
Thinking about music data management, what technical skills and programs besides Excel and Salesforce are really important to know? What do you use on a daily basis?
Excel and SalesForce are where I do most of the heavy lifting at work. I will say that while I have no formal training in computer science or programming, the skills I learned designing websites as a teenager have come in very very handy. A lot of what we do involves web-crawling and data mining on the internet, and being able to modify API calls and sort through the source code of websites looking for media files is a huge help. Long story short, if you understand how HTML works and know how to ask Google for help, there’s an awful lot you can get done!
How much technical/coding knowledge do you need to work with developers and administrate the backend and client side of a website?
Just enough to communicate! Fortunately, I work with a really smart and talented team, and the programmers are really patient when they have to bring me up to speed on why something may or may not work.
Do you have a favorite age student to teach music to? Have you noticed any similarities between students who stick with learning an instrument vs. students who give up?
It really has to do with attitude, some of my favorite students are in their 50s, some are younger than 10. Regardless of talent or passion, if a student is willing to show up to a lesson ready to learn and have a good time, we tend to make a lot of progress and stick with it long term. I’ve had very talented students who want to be professional musicians but get frustrated and lose focus when they hit a plateau.
What I’ve found is that progress happens in two stages: first the ears get better, and then the hands get better. As soon as the ears get better, the student hears what [he] has been doing wrong for the first time. Often, this can be frustrating and even make it seem as though things are getting worse. Instead, hearing what is going wrong is the first step to addressing it; once the ears find the problem, we can work on fixing it by building technique. This is how the hands get better. Of course, this is a cycle, and one reason why musicians always seem to fluctuate between feeling great about our playing and feeling terrible. I find that if you can recognize this process and stay on top of where you are it makes the journey less frustrating and intimidating, so I always try to help my students see their progress in context.
What advice do you have for people who work with music data?
When I was a teenager, everyone told me that becoming a musician was a bad idea. The record industry had just started its rapid and historic decline, and the prevailing opinion was that there wouldn’t be any feasible way for musicians to make a living anymore. Fortunately, I’m very very stubborn, and just couldn’t see myself being anything other than a musician. Instead of the dust-bowl of a music industry that we were promised, what we’re faced with now is a blank slate for the new digital economy. Those of us working making music and technology have an opportunity to build this new system, which is fun, and if we do it right, a chance to ensure that musicians are paid more fairly than they were in the 20th century.