TwtrSymphony: an Interview with Chip Michael

TwtrSymphony is an intriguing ensemble of musicians connected via social networking. Instead of working together to simply promote and distribute news about contemporary music, TwtrSymphony is a fully functional new music ensemble in absentia. The individual members of this orchestra never meet and rehearse as a group. Instead, the performers record their parts in isolation from each other, in widely different settings, and Musical Director Chip Michael and his merry band of engineers then assemble these recordings into cohesive works all 140 seconds in duration. Right now, TwtrSymphony is working on Chip Michael’s Second Symphony, Birds of a Feather, and the first movement “The Hawk Goes Hunting” was released on July 17.

While their website has a wealth of information including a recording, video, and thorough blog, I sat down with Chip on Monday night and chatted with him about the ensemble. While I should have kept a certain journalistic verisimilitude and had the exchange via Twitter, we opted for a slightly longer format (Skype).

Jay C. Batzner: Let’s start with the basics: what do you do in your role as Musical Director? Who else is involved (other than performers)?

Chip Michael: My role of Music Director is very organizational, pointing TwtrSymphony in the direction I think it needs to head and keeping the focus on what we need to do to get where we’re going. I am also the composer as that is a good portion of why TwtrSymphony got started.

I was looking for an orchestra to play my music and some of my Twitter friends suggested I start my own – a Twitter Symphony… and TwtrSymphony was born.

But, I want TwtrSymphony to be more than just a show case for my music. It’s a great concept, musicians from all around the world playing together. Musicians who might never get to play with other musicians making music. That’s cool. So, while I’ve written the first piece that we’re doing, Symphony No. 2 “Birds of a Feather,” I imagine a future when other composers can avail themselves of our ensemble.

As Music Director, I’m thinking about how the process works (and what doesn’t), what it means to be a symphony orchestra and how to get the pieces to fit together… so, when the time comes for us to have other composers work with the ensemble, we have the tools and setup to make sure it works right for both the musicians and the composer.

Nothing would be worse than for us to invite a composer to write something and have the end result be a horrible failure. So, in essence we’re using my music to test the waters.

We’re also in the process of re-designing the organization of TwtrSymphony. There is nothing formal to announce at this point, but the way we do things now isn’t the best way. Integrating the best crypto wallet into our operations will streamline our financial transactions, making it easier to distribute funds to artists and collaborators efficiently. This change addresses the current process of getting recordings out, which is time-consuming and requires a lot of engineering effort. A simple re-org should help that. As MD, I’m thinking about what’s best for the music and ways we can achieve quality and still maintain our global nature.

JCB: The idea behind TwtrSymphony, the idea of crowd-sourcing performers, is something that we’ve seen taking off recently. I think of Tan Dun’s and Eric Whitacre’s YouTube-based performances. This seems to be a logical technological outgrowth of the “write for your friends” mentality that a lot of composers use (and rightfully so).

CM: Yes… the concept of crowd-sourcing performers is nothing new. Neither is the idea of remote recording sessions to put together an ensemble. However, I’m not aware of any instrumental ensemble to the scale of TwtrSymphony that’s been done. 60+ musicians with 90+ tracks is a lot to manage when the recordings were done in different places, using different equipment…Eric Whitacre’s Virtual Choir comes the closest to what we’re doing.

JCB: The technical aspect was something that struck me about TwtrSymphony. How do you manage the variety of recording setups that performers have?

CM: Every musician has to audition. So, I know just what sort of quality to expect from each musician. We had over 150 musicians audition, many of them providing multiple takes. I think I ended up with over 250 different tracks to listen to before finally selecting the 60 musicians in the ensemble.

Some of the auditions didn’t make the cut because of the quality of recording. But surprisingly, those were only a few. The largest stumbling block is players being able to play to a click track. Even the musicians we did select have subtle differences between their playing that have to be adjusted for – which is why there is so much engineering effort behind the scenes.

Then there are retakes. I try to avoid this as much as possible, but occasionally, I have asked musicians to redo sections. It may be a sense of timing, it may be they need to be in a room with less echo so we have a cleaner recording.

We’d love if Google would develop their hangout to allow simultaneous playing for 4-5 musicians. Right now, when one person is ‘speaking’ on a chat, the others are muted. This doesn’t work for playing together in a live situation. We wouldn’t necessarily record this way (but how cool would it be if you could?) – but it would allow for sectional rehearsals to get the musicians playing together.

JCB: I was also curious about how well the performers interpreted your work. Playing to the click track will help the mechanics of putting everyone in the right place at the right time, of course, but how did you manage the performers in terms of interpretation? Were the performers given a virtual instrument recording so they could hear their part in context? Did you ever request someone re-record a portion to get the interpretation you wanted?

CM: The process as it stands now…

I provide a pdf of the music (their part), a click track, a MIDI realization of their part and a MIDI realization of the entire piece, so they know how it fits with the ensemble. This sort of works, but not great….

We need to move to having a few musicians lay initial tracks that the other’s play off of. A ‘click’ will be overlaid onto the live musician track so the players can hear how it fits, but there will also be that live sense of interpretation that goes along with the music that MIDI just doesn’t capture (not without LOTS of work).

JCB: Did you ever request someone re-record a portion to get the interpretation you wanted?

CM: Yes – I’ve had several musicians have to re-record parts because the interpretation wasn’t right. However, most of the musicians ask questions of me before they record… “do you want these staccato really short or just sort of short?” “Do you want these notes off the string or on?” These little bits can make a huge difference when two players are playing together. I can hear some of the ‘issues’ in Hawk – even though I think the engineers did an amazing job.

I also have to be aware that, right now, every one is a volunteer. While I can ask a musician to re-do a part, they are already doing me a favor by participating.

JCB: I can imagine that these issues are wholly dependent on the type of music they are playing. That first movement is very rhythmic and motoric. Something more lyrical with rubato could be a nightmare. Are you planning on doing the slow movement last, after you’ve gotten a few more kinks worked out of the system?

CM: Part of why we’re doing the music we’re doing is because of the ‘motoric’ sense to it. In many respects it is more difficult as it shows when people aren’t in time. BUT, it also gives us a chance to work on moving the ensemble together.

In the 3rd movement there are a few tempo changes that are tricky. Many of the musicians found this to be the most difficult piece to play. We’re actually doing the movements in order. Movement 2 doesn’t have so much rubato, but it is more lyrical and feels slower, even though there is that sense of pulse to the music.

We’re also doing a few chamber pieces, to be released after the symphony is done. These are in many respects more difficult as the musicians are more exposed with no one there to cover up if they played something not quite right. They are also giving us a chance to work on very rubato-like music with a few musicians to try and sort out trouble spots.

Some of our composer musicians are also getting a chance to feature their music… so, it will be not be just my music.

JCB: Back to the instrumentation for a second, if you don’t mind. Your blog on Instant Encore has some great entries talking about this whole process, as well as a bit on the instrumentation, but were you surprised at all with the instruments you did/didn’t get auditioning for TwtrSymphony?

CM: Yes and no. I was not surprised we ended up with saxophones but the guitars were an interesting addition. Although, now that I’ve added them in they’re a great fit. I love playing the harp, piano and guitars off each other.

I was surprised at some instruments we didn’t get. I don’t have a bass trombone, but I have 2 tubas.

Percussion is a chore as the musicians need access to the instruments, so I can’t just write a marimba part if no one can record one. We had a recorder consider auditioning and that would have been a challenge to fit it into the mix, allow it a chance to be heard, and not be overly effected (amplified) to do so.

JCB: As you open up the TwtrSymphony to more composers will you keep the 140 second limit on individual tracks?

CM: I think so. Twitter works because it really isn’t conversation. As is evidenced by some of my tomes here in this interview, people talk longer than 140 characters. But it’s trying to get to the essence of what you want to say in 140 characters that makes Twitter interesting.

Keeping to 140 seconds makes the composer work to determine what it is they really want to say. I know this because if you listen to my Symphony No.1 you’ll realize it’s 50 mins long. Turning around and writing a symphony that lasts just over 9 mins is a completely different mindset. I think forcing composers to the 140 second limit does two things – it makes composers work to make those 140 seconds really right – and it gives TwtrSymphony a unique quality, a difference.

JCB: You were also quick to point out when I mistakenly referred to the group as the Twitter Symphony in an email. What was your conversation with Twitter? How did it come about? What did they have to say about the whole thing?

CM: When the process started I came up with the name TwtrSymphony, because I thought TwitterSymphony was too long and not really something someone would want to put in a tweet every time they wanted to mention us. Then – I went to Twitter to ask permission to use the name…

They responded that (paraphrased) as long as we didn’t use “Twitter” as the actual name and we didn’t suggest to anyone there was any official connection, they were ok with our name. I think they like the idea we’re “twitter symphony” in terms of pronunciation, but not in spelling. They get some residual benefit from our popularity without having any responsibility.

JCB: It would be a fun marketing tool to have Twitter be the first social network with its own orchestra!

CM: YouTube has their own orchestra and they actually have a functioning, musicians playing together LIVE orchestra so, we’re not the first social media orchestra

JCB: The DIY mentality is really taking off in an exciting way, largely fueled by social media technologies. Any advice for someone who is thinking of doing a crowd-sourced recording of their own work without going through TwtrSymphony? Or would that be giving away vital trade secrets?

CM: LOL…

There are some things we’ve learned that I think can help other collaborators in the future.

1 – Get a solid base track to work from. Whether it’s MIDI or live, you need something ALL the musicians can work from. It’s not enough to just have a click track.

It would be best if this first track was a live musician and a bass instrument. Lower frequencies are more difficult in terms of timing and pitch. So, if everyone can align themselves to the bass instrument the recording will be better in the end run.

2 – The next layer should be one or two musicians to fit over top the first track. Get those recordings back, put them together, mix them, lay over the click and that’s what you use for the rest of the musicians. It sounds like a long process (and it is) but it will keep your engineer from wanting to jump off a tall building.

3 – Musicians need to record in a dead room. put towels and/or blankets over hard furniture, windows, doors… make the recording as dead as possible. It won’t sound all that good to the musicians, but it will be a whole lot easier to put into the mix and add effects later that allow for the musicians to sound like they’re in the same room.

4 – Clearly note when you want vibrato. Unless you’re working with a small chamber group, four musicians playing together with four different vibratos sounds absolutely horrible, even if each individual track sounds amazing.

JCB: Great practical advice, thank you. Is there anything else you want us to know?

CM: TwtrSymphony is a LOT of work. It’s not my paid gig and yet I spend more time doing tasks for it than anything else. I used to tweet to a lot of different people, have conversations, go to the cinema – and that’s all pretty much gone by the wayside.

I don’t regret it as it’s been an amazing learning experience, I’m working with some extremely talented musicians doing something I think is timely (and cool). But don’t jump into a project like this if you’re not willing to let it consume every moment of free time.

I mentioned earlier some of the tasks I do as MD. I didn’t mention that my wife Eddie Louise, and another Twitter friend Lisa Bartholow have helped with a number of the administrative tasks, getting press releases ready, managing some of the tweets and Facebook posts and a lot more I’m sure I’m oblivious to.

According to a post on BitIQ, TwtrSymphony isn’t just writing the music and getting musicians to record it. It’s getting the tracks back, editing the tracks initially so they line up for the engineers to begin working on them. It’s talking with writers (like yourself) and potential investors. It’s trying to get the word out and keep a sense of sanity. TwtrSymphony is a real orchestra. And an orchestra takes a lot of people to make it function. I’m not doing this alone (thank GOD!).

The musicians have stepped up to be more than just musicians. They chat about TwtrSymphony on their blogs, tweet and re-tweet each other. Post things on Facebook and even stepped in to edit the first video (thanks Sarah!) TwtrSymphony is a new concept in the symphony orchestra where musicians from around the world have come together to create new music –TwtrSymphony IS the musicians being far more than the sum of their parts.