Writers of writers

In my most recent creative technology explorations, I’ve been really interested in autonomous software that can generate music or art. In reading about this stuff on the Web, I often find comments along the lines of “the computers will put the composers out of work.” In a fascinating post about neural-network-based machine intelligence writing prose and poetry, Ross Goodwin has the perfect response.

When we teach computers to write, the computers don’t replace us any more than pianos replace pianists—in a certain way, they become our pens, and we become more than writers. We become writers of writers.

Yes, please. I want to do that.

Since the above-linked post was written, Goodwin has written a new post linking to this delightfully insane film, shot from a screenplay written by his neural network algorithms.

(The writers I’m writing are much dumber than the ones Goodwin is writing, but it’s still fascinating and thrilling to me. Here’s an example of one of the better things my little guy made.)

Kadenze is my new home for learning about creative technology

Massive Open Online Courses (MOOCs) made a big splash a few years ago with the launch of Coursera. If you’ve not heard this wacky term, a MOOC is an online course that is usually free or cheap, and involves a nearly unlimited number of students to take a class simultaneously. I’ve read of Coursera courses that have had tens of thousands of students. This creates an active community, and a low-pressure learning environment. Coursera is a platform hosting college courses from major universities on a wide variety of topics.

Kadenze is a relatively new MOOC platform that focuses on creative technology. It was founded in part by professors at California Institute of the Arts and Princeton University (among others) who specialize in computer music. You can audit all of their classes for free, which allows you to watch lecture videos, read course materials, and participate in class discussions. To submit assignments and receive feedback, you can sign up for a premium account at $10/mo.

I’ve audited parts of four or five courses on Kadenze in the last year. The quality of the content and instruction is extremely high, and the topics are right up my alley. One small frustration is that there is no way to watch the videos offline. I often want to watch on my iPad while I’m in low-connectivity areas, and there’s currently no good way to do that.[1] Apart from that minor quibble, my experience has been great, and I would strongly urge anyone who has summer tech learning goals to check out their upcoming offerings, many of which focus on free, open-source software.

  • Introduction to Programming for Musicians and Digital Artists – This is a class by one of Kadenze’s founders, Ajay Kapur, on music programming language ChucK. The videos are a little hokey at times, but it’s a terrific starting class. One of these days, I’ll post a little about a project I’ve been working on in ChucK. It’s not quite ready for primetime just yet.
  • The Nature of Code – Dan Shiffman on creating visual programs in P5.js that mimic natural processes like flocking, schooling, and branching.
  • Programming Max: Structuring Interactive Software for Digital Arts – Not open yet, but I’m all about this. For those on team Pure Data, there’s a PD course scheduled for the end of this year with Sergi Jorda and the Notorious MSP himself.
  • Touring Modernism: From the French avant-garde to American Pop and Beyond – This one is a little bit not like the others, in that it’s a little more theoretical than applied, but the first unit of lecture videos that are available as I write this are stellar! Professor Lisa Wainwright of the School of the Art Institute of Chicago[2] is a really engaging lecturer, and she gets to use examples from right there at the Art Institute!

So if you’re like me and set unreasonable learning goals for the summer time, these courses are worth every penny of either the free or paid tiers. Kadenze is one of those rich resources that makes me want to quit all my jobs and responsibilities and just dive in to learning stuff really deeply and making wild and reckless creative decisions. If you’re making something with what you learn there, please share. I’d love to see.


  1. In fairness, this is not entirely Kadenze’s fault. While they provide the MOOC platform, they don’t own the content. Rather, they license it from the people and institutions who create it. I suppose this arrangement helps keep costs down, but I would definitely pay the $10/mo. if I could watch offline.  ↩
  2. …of Latter Day Saints, I suppose.  ↩

Support

This past Sunday’s #musochat, hosted by New Music Gathering, was focused on support for new projects. I’m trying to get better at deciding to take on new projects, not take on other projects, and stopping projects that have run their course. It made me think of this talk by composer Martin Bresnick. Some might think of this direct discussion of money as a little crass.[1] But lately I’ve been thinking of “support” in terms of how much time and energy I can devote while maintaining my ideal level of mental and physical health.

This is one of Chamber Music America’s First Tuesday sessions from March 2016. In it, Bresnick talks about how much a composer should charge for a commissioning fee on a new piece of music.

There are three parts to this: you the composer, the next part is the commissioner, and the third part is the work itself.

I read a nice suggestion a while back. If the response to a new project isn’t “Hell yes!”, it has to be “no.”


  1. I might argue that those people are being a little precious.  ↩

Better Feedback on Creative Projects

If you have ever made something and presented it to an audience, you’ve probably had the horrible experience afterward of being told how great it was. To some, I’m sure that sounds dumb. Being told that you’re great shouldn’t be horrible. Unless, of course, you were hoping to learn something about how the audience perceives the things you make. As an audience member, I have found the exchange to be equally unsatisfying.[1] As a teacher, I often place myself and my students in these positions hoping that we will all learn something from it. A social media “friend” a while back directed me to a feedback system called Critical Response that I began using in my own sessions; and, it has dramatically improved their value. Remember, we don’t have feedback sessions for fun, even though they can be lots of fun. We have them for their utility: to learn to make stuff better and, in doing so, make better stuff.

Choreographer Liz Lerman developed the Critical Response Process (CRP) to solve the issue I described above. It does an excellent job of separating the personal tastes of the “responders” (audience, in Lerman’s system) from that of the artist. The artist gets to determine how well they[2] achieved their goal without getting into aesthetic disagreements. I’ll describe it here in broad strokes, but if you’re interested in implementing it, I strongly recommend her short book on the topic.

Lerman defines three roles in a CRP session: artist (composer), facilitator, and responder. After presenting the work, the process runs through four phases. What’s particularly interesting is the direction to hold value judgements (positive or negative) to the very end. Again, this allows everyone to focus on the artist’s goals and the audience perception without passing any judgement on those goals directly. As I describe the process, I’m going to assume we’re talking about music. But as Lerman’s subtitle explains, the system is valid for any creation, “from dance to dessert.”

  • Step 1: Statement of Meaning – The facilitator asks responders to explain what they heard as specifically as possible. Lerman suggests “What was stimulating, surprising, evocative, memorable, touching, or meaningful for you?” You may have noticed that these are not all value-neutral adjectives! In my limited experience with CRP, it’s hard to avoid this at this stage; but, I try not to let anybody get too effusive or negative.
  • Step 2: Artist as Questioner – The composer asks questions about specific elements of piece. This is not the time for “Did you like it when…?” or the well-worn “What did you think?” Those are just soliciting general opinions. There’s time for that later. Artist questions can provide insight into what the artist thinks is important about the work. They might ask, “Could you hear the gradual harmonic shift from measures 28 to 40?” Specific questions yield more useful responses.[3] When I first started, I should have done a better job at preparing composers (especially students) for this. Good questions here can really raise the value of subsequent steps. I would encourage composers to take a few moments to write some questions down.
  • Step 3: Neutral Questions from Responders – These questions are tricky for responders who are new to CRP, as it can take some effort to phrase certain questions in neutral ways. So instead of asking why the third movement was so long, the responder might ask how the composer is thinking about the structural proportions of the work. There might be a good reason for the third movement to be long. Maybe the issue isn’t the length, but rather how the third movement is prepared by the previous two, or how it develops, or something else. A neutral question allows everyone to frame issues in the context of the composer’s own goals for the piece. Neutral questions are a great way for the composer to learn how the audience perceives the music.
  • Step 4: Permissioned Opinions – Responders get to offer direct and clear opinions here for the first time in the process. However, they should ask permission first. The script goes something like “I have an opinion about [specific thing]. Do you want to hear it?” This gives the composer the chance to avoid getting bogged down in parts of the work that are still under heavy revision. So if there are certain elements of orchestration that are still being worked out, a composer might not want to waste time in a back-and-forth about what they consider to be a placeholder decision. A nice side-benefit: the time of asking the question allows the artist a quick moment to recall the reasoning that went into whatever the opinion is going to be about. This might allow them to respond more thoughtfully. In my experience as a composer, when I’m not “prepared” for an opinion, it’s much easier for me to get defensive or ignore the feedback entirely.

This process usually takes around 30–35 minutes for my composition students. When you add time for listening to the work, this is approaching our weekly meeting time of 50 minutes. During certain times of the year, I may try to cram two of these into one meeting time, which often means that I have to cut a section short, or cut it entirely. I nearly always eliminate Step 4. Not only can these opinions be expressed and discussed after class, I think that both composers and responders actually get a lot more out of the questioning steps than the opinions. These are people that see each other and discuss music regularly anyway, so they tend to be quite familiar with one another’s musical taste. In the best of all possible worlds, we would always have time for all four steps. In this world. We sometimes live with three-and-a-half steps.

If you’re looking to implement a form of CRP in your feedback sessions, I have a few bits of advice. First, to paraphrase a great American philosopher, you don’t have to take my word for it. Buy the Lerman book. My outline above is a very high-level overview. The book offers a lot of detail on variations, examples, and specific advice for facilitators of the process that can make or break a session. And of course, I would hate to think that somebody read this 300-word summary instead of supporting Lerman’s work. Second, be sure you do a good job of explaining the process before you do the first run. Explain each step; give examples; and most of all, explain why the steps are structured the way they are. For those accustomed to either a critical firing squad or anti-critical Care Bear Stare, this will be a little uncomfortable and may come across as arbitrary at first. In my experience, it is worth getting over that hump, and the efficient way to do that is adequate preparation for the first sessions.

I would be curious to hear from any readers about feedback sessions you use for your creative work. What is your process like? Do you use a form of CRP? Any advice you’d like to share? Drop a note in the comments, or find me on Twitter.


  1. In fact, I know there are times when I’m so concerned about what I’m going to say to a composer or performer at the end of a performance, that I’m distracted from the music they’re making right in front of me. I know. I’m a terrible person.  ↩
  2. Yes. I used a singular “they”. You should too.  ↩
  3. Lerman does warn against questions that are too specific, but that has not be a problem in my sessions so far.  ↩

The Great Internet Deflation continues

This week, the The New York Times announced that they would join the Associated Press in decapitalizing (decapitating?) the word “Internet.”

“In our view, it’s become wholly generic, like ‘electricity or the ‘telephone,’ ” he said. “It was never trademarked. It’s not based on any proper noun. The best reason for capitalizing it in the past may have been that the word was new. But at one point, I’ve heard, ‘phonograph’ was capitalized.”

Dumb analogy. The Internet, as a physical object, is arguably humanity’s greatest creation, and certainly among its most influential. You know how many phonographs there were? Me neither, but I’m pretty sure it’s more than one.

Furthermore, as we grapple with issues of network neutrality that may actually create many Internet-like networks, it’s important to use language to remind ourselves that the Internet is useful precisely because of its singular nature.

Long live Internet.

Career Candor

Remarkably candid post from pianist Andy Lee on NewMusicBox recently.

First, a deer chaser makes noise, and that’s exciting. There are a lot of things I want to do with my career, but making some noise and getting noticed seems like a good place to be. Second, all the water that pours out of the deer chaser has to go somewhere. No, we don’t always get to control where it goes, but that water still nourishes the soil and helps create new growth.

Likewise, career progress is often difficult to see. I haven’t sold any CDs as a result of that post, nor have any gig offers come my way, but I’ve expanded my new music network (to use a crass term), and I’ve gotten my foot in the door with an entirely new community. It was also a useful reminder that I enjoy writing. That’s not nothing.

Not only does he talk about his career struggles directly, but also the psychological impact of those struggles on his approach to future opportunities. Read the whole thing.

And when you’re done reading, buy Andy’s disc. It’s great.

SN234: Q is for Qhord

I was probably in high school trying to use Sibelius 1.2 to write an arrangement of something for me and my brass-player buddies the first time I got help from “Daniel at Sibelius.” Well, it’s been a long fifteen years for both me and Daniel Spreadbury; but, he’s as delightful and nerdy[1] as ever. This week’s SoundNotion was now the third time we’ve had him on the show to talk about Steinberg’s new scoring application, Dorico. If you’ve not been following the drama of the last few years, Daniel and many of the original Sibelius developers have been working in secret on Dorico for the last three-and-a-half years. Take a listen to our excellent conversation with Daniel. (show notes)

I gave Daniel (now “Daniel at Steinberg”) a bit of a hard time at the end of the episode on the licensing tech that Steinberg plans to impose, and I may have implied that I wouldn’t buy this thing if I couldn’t use it on two computers. I’ve been thinking about that a bit more since we recorded this show last Friday. And I’ve reached the following conclusions.

Second, it’s ok that Dorico 1.0 is not a one-for-one replacement for Sibelius 7.5 for me. I can use it for projects that fit it, and installing Dorico on my Mac isn’t going to break Sibelius. I can use them both. No, it doesn’t have chord symbols. But you know what? I don’t usually use chord symbols, and if I need them, Sibelius is a click away.

Third, this is a huge project serving a niche market. I want to show Steinberg that they aren’t wasting their time and money with it. If I don’t support the competition, I don’t think I can whine about Sibelius.

But first and foremost, the most significant feature of Dorico is not its proportional spacing algorithm, slur arc controls, or even the miraculous open meter implementation. It’s actually the people that make it. My experience corresponding with Daniel and other members of his team lead me to trust them. Trust is not a word I’m usually comfortable associating with giant multinational organizations; but, the reason I use it here is that I’m not describing a big organization. I’m describing people.

So Steinbergers, you may have my dollars later this year. I’ll even pay whatever a VAT is if I have to.


  1. On this blog, as in life, “nerd” and “geek” are terms of honor and endearment.  ↩

The eye of the $100k-beholder

Mark Stryker of the Detroit Free Press has a great write-up of the innaugural M-Prize, a chamber music competition hosted at the University of Michigan. I love that chamber music is getting such a high-dollar award – $100,000 grand prize and $200,000 total.[1] However, Stryker had some strong words on the judges selection. The winning group was a string quartet (sigh), playing Debussy (sigh), Haydn (sigh), Mendelssohn (sigh), and a token two minutes of Webern. Another finalist, the adventurous piano-percussion quartet Yarn/Wire, played two new commissions.[2] Living in Florida, I didn’t get to hear the finals. But, I’ve heard Yarn/Wire before, and so I’m certain Stryker does not exaggerate when he writes:

[Yarn/Wire] played with a cohesiveness that at least equaled the Calidore quartet and a depth of expression and distinctive interpretive identity that surpassed the winner.

But more disappointing than that was this exchange between Stryker and a member of the winning Calidore quartet:

It’s troubling that the Calidore’s programming was so relentlessly conservative. The most recently composed music the group played was the two-minute first movement from Anton Webern’s Five Movements, Op. 5, written in 1909. After the semifinals Thursday morning, I asked Calidore violinist Ryan Meehan if his group plays any contemporary music. “Yes,” he said. “We played Webern this morning.”

The whole article is definitely worth your time. Stryker is one of the tiny handful of music critics that is willing to express an opinion.


  1. Arguably, chamber music might be better supported by giving $20,000 to each of five chamber groups, or some other division, but that’s another issue.  ↩
  2. Perhaps unsurprisingly, the other of the three finalists was a saxophone quartet.  ↩

Klinghoffer Cancellation Thoughts

Note: This was originally published 24 June 2014 on Medium in response to some of the controversy surrounding the Metropolitan Opera’s production of John Adams’s The Death of Klinghoffer.

This week, the New York Metropolitan Opera and general manager Peter Gelb cancelled simulcasts of John Adams’s The Death of Klinghoffer. (NB: They’re still performing it live.) Gelb and the Met were concerned about accusations of anti-Semitism that have been cast on the opera, which depicts the true story of a terrorist hijacking of a passenger liner in 1985. The passengers on the ship were both Israeli and Palestinian, and the hijackers were affiliated with the Palestine Liberation Front. Leon Klinghoffer, a Jewish man, was shot by the hijackers, and his body was thrown overboard.

The events on which the opera is based occurred nearly thirty years ago. The opera premiered nearly twenty-five years ago. Is it really “too soon” for the arts to address these issues? I would argue that it is never too soon to start talking about something. If you don’t want to listen, don’t buy a ticket.

Those arguing against the opera (and in favor of the Met’s decision) say that Klinghoffer glorifies terrorism or elevates extremists in some way or another. Either these people have not seen or read about the opera, or they are willfully misunderstanding it. I saw a beautiful new production of it at Opera Theater St. Louis in 2011. If you aren’t familiar, take a moment to listen to some (thanks, Spotify!). You won’t regret it.

The work opens with two choruses, sung back-to-back: “Chorus of the Exiled Palestinians” and “Chorus of the Exiled Jews.” It acknowledges that the struggle between this two groups is deeply rooted, it has touched the lives of millions of people. Furthermore, it presents the first of several uncomfortable truths: that nobody is without fault and nobody is without suffering. Only a very naïve person could believe otherwise.

One of the things I love about opera, and other kinds of storytelling, is the way it plays with what I think of as the “cultural scale” of a character. We often get to see larger-than-life characters treated as regular people, as in Wagner’s pantheon of mythical characters in The Ring. Other times, we get to see regular people treated as tragic heroes, as in the starving artists of Puccini’s La Bohème.

Throughout The Death of Klinghoffer, the audience gets to know some of the characters involved in the eponymous act. Yes, you read that correctly. It’s an opera about a terrorist act, and some of the characters are terrorists. Klinghoffer is in an interesting position regarding the scale of its characters. They actually were regular people. Then, they got on a boat, and their lives changed. They went from being regular people to being characters in history books and Wikipedia pages. Now, John Adams and librettist Alice Goodman have the interesting task of turning them back into people. The audience is forced to confront a terrorist on a personal level. The opera tells us “These people do evil things. But they’re people just like you. They have needs and wants and desires just like you. But some of their needs, wants, and desires involve acts of terrorism.” This turns these comic-book-scale supervillains into people. This is another very uncomfortable truth because it forces each of us to consider: Am I capable of that kind of evil? In Tweet-length: The Death of Klinghoffer isn’t an opera about religion. It’s an opera about people. See there, I even left you 55 characters for hashtags or whatever the kids are doing these days.

Here’s the thing, though. Art often makes us uncomfortable. In some sectors of the arts, making the audience uncomfortable is seen as a feature, not a bug. Making us uncomfortable helps us to think about things in new ways, see sides of the world and ourselves differently. If we are to argue that art has a purpose and relevance in contemporary society, we should focus on this artistic function.

Also, let’s not confuse the intentions with a character with the personal beliefs of the creator. I don’t mean to burst your bubble, but if you run into Bryan Cranston on the street, he’s probably not going to sell you any crystal meth. The presence of antisemites in Klinghoffer does not make it or its creators into antisemites. People don’t protest performances of Don Giovanni because it advocates misogyny. Grown-ups understand that Giovanni is an asshole, but that he’s also a character distinct from the actor, composer, director, and all the real people that go into creating an opera character.

Hello world!

There’s something you should know.

I’m seriously considering starting a blog.