How it works: Part one and part two are each subdivided into numerous fragments (each being one or more phrases). Each pianist is engaged to learn, and then record with MIDI uploaded to the Currawong platform, two fragments from part one, four from part two. The total amount of music each pianist learns and uploads will be about three minutes.

Each pianist will upload numerous takes of their respective fragments. After each recording of a fragment they are free to reject (delete) or accept (upload) to the platform. We expect something on the order of 8 - 10 uploaded versions of each fragment to populate the source queues. In this first phase, artistic management of the project will determine which takes should be considered most representative of the score fragments.

All pianists are encouraged to refresh their queue with more uploaded recordings from time to time. On an ongoing basis we will listen to new versions and accordingly update the queues as warranted.

Over time pianists may choose to learn, record and upload additional fragments so that what was initially one pianist / one fragment queues become populated over time by more pianists. The significance is that the Installation/Performances will be relatively stable initially, but over time will show increasing variance.

Let's try to explain that one by getting into more detail about the digital sound transformations. As noted earlier, transforms are, at the macro level, pre-set in the sense that the type of transform is selected in advance for every phrase that is subject to transformation. But beyond that the sound transformations are highly subject to how a preceding passage was played. As the piece progresses, the computer is not only causing piano sound to be modulated, but also continually evaluating characteristics of the performance (based on criteria we "instruct"). As more pianists "weigh in" by uploading their performances of more of the many available fragments, the overall variability of these performed fragments will increase. Accordingly the overall spectrum of sound transformation over time will enlarge as well.

Implementation: An Installation/Performance will, at the first level, consist of each of the captured fragments in the order they occur from start to finish. They will be drawn from the regularly updated queues of approved takes -- somewhat like assembling a studio recording to produce a complete performance of a work, except that the performances in the source queues will from time to time refresh. Thus there is no "definitive" version of any fragment, let alone the piece itself. (The order of fragments within the piece itself never changes. This is a piece composed from beginning to end and the score is the score.)

The evaluation criteria are determined partly by ear, but also over time by data analytics, which are designed to analyze and detect patterns in how phrases are played. More about the data analytics.

The thing I want to do in any piece for a soloist is to gain as much purchase as I can on the full range of expression the instrument offers. And there is nothing up for that challenge like the piano.

Now does that make for a piece hard to play? Absolutely. The piece is really difficult, start to finish. But - important, in my view - it's not ridiculously difficult. There are pieces that are arguably unplayable. This is not one of them. And when I think of the technical skill required to play Ravel or Scriabin or Liszt or Prokofiev... this is definitely in the zone.

But let's get to the more interesting question, the one about musical language itself. I will back into an answer this way: No matter how complex the moment, no matter how much is going on in a measure, I want a listener never to lose a sense of "place" at any moment. I don't mean grounding at all in the sense of traditional harmony, yet something not totally at odds either - rather the goal is a certain balance between dreamlike imagination and stability.

All very nice, but how to achieve? When I think about the time we roughly identify as Modern to Contemporary, the past 75 years or so, the works that strike me as the most glorious achievements were not products of various "movement solutions" (e.g. serialism, new romanticism, minimalism) but rather from the cracks, as it were. With feet planted in more than one space - hybrids, arguably.

I'm sure there are many great pieces out there I have not been fortunate enough to hear. A few I have been privileged to write about, such as Nono's Promoteo and Grisey's Les Espaces Acoustiques and Schnittke's string quartets. The fact that these pieces are as breathtakingly great as they are proves that music that speaks largely in the realm of the abstract can be just as emotionally and intellectually powerful as the most revered works in the classical music canon.

A composer can't credibly proclaim "I achieved this" but can fairly say "this is what I tried to do". So let me point to two short piano pieces by Stefan Wolpe, Form (1959) and Form IV (1969) that were of singular influence on my development of a musical vocabulary. (With the caveat that when you are any kind of artist in your teens and early twenties - as I was when I first heard these pieces - anything you like becomes influential.) In a way - I realize this only in retrospect - the two parts of get used to it man ... are, in relationship to each other, something like between Form and Form IV.

If I may add this about Wolpe: The few pieces he managed to write in his last years, perhaps at moments when the cloud of advanced Parkinson's disease that he suffered from had briefly lifted, are to me among the most profound accomplishments of music from that era. The time period is around 1969 - 1971. They are not lengthy or complex - rather, they are gossamer and concise, almost fragile. But they point to a path forward.

What about the electronics? Elsewhere on the site there's plenty of detail on that, so here I'll try to say what they mean to me compositionally. First, the piece is perfectly OK on its own; it doesn't "need" the electronics. I try to push the envelope as to what the piano - and the pianist - can achieve. The electronics are meant as (1) a kind of inevitable extension of imagination (as in, how weird can things get and yet retain a constant sense of place); (2) a repository of the history of interpretation and performance that in turn shapes future performances - something that we could only contemplate very recently. An essential aspect of composing with this medium is that the electronic transformations will never be the same twice. They can't be, because they are informed by how the piece is played during that particular, never-to-be-repeated performance, and further derive from contextualizing a passage in performance by comparison with how others played the same passage at other times.

Back to Wolpe for a moment more: My first years as a student at Columbia coincided with Wolpe's last years with us. His music was prominent at new music concerts, and so I would see him in the audience. There was one moment - just one - when we actually crossed paths. We were both living at Westbeth in the West Village. I was entering the building as he was being carefully wheeled in. He looked warily over his left shoulder at me, and I nodded, no doubt imagining to myself that he recognized me from the audience. I said nothing, of course. Looking back how I wish - hindsight is easy, right? - that I had said something like... "Oh Mr. Wolpe, your music means so much to me!"