Space between notes

Indian music is about the space between notes. It’s also about the space within.

For over twenty years, every one of my attempts at resolving the problem of expressing my musical ideas digitally hit the wall of above realization.

Indian music is fundamentally monophonic. If mono-melodic is a word, that might be more apt. It is about creating a single melodic line that remain complex enough to be interesting. There’s a gap in time that a strand of moving frequencies want to traverse. Of all the possible paths, I need to find the one I didn’t traverse before, but the one that leads to the next.

My favorite part of Indian music, especially Hindustani classical, is the primacy of improvisation. It calls for continuously creating new ways of moving in time and frequency. It’s like building an intricate structures on the go, while maintaining uniqueness. Something like the growth of crystalline structures in a super saturated liquid. And like that process, here too, I need to super saturate my brain with musical ideas, dissolved melodic strands.

I like doing that. Very much. For the music that get realized is something new for me as well. Performance here is not about retelling, nor recreating.

When a group perform music together, It’s about being part of the whole. Our individual expression needs to be congruent to the rest. It is relative to the rest. Producing music as group is exhilarating, and very fulfilling. But to make this togetherness happen, there need to be a musical framework that supports this. European classical music traditions (let’s call it western classical), probably influenced by group worship traditions of Judaic religions, came up with a strategy based on creating spectral structures. It try to fill the audible spectrum with static structures. The movement then is morphing of these spectral artifacts to the next in quantized time. Being written for and performed by many people (or polyphonic instruments like piano, organ), it also specifies the limits to individual freedom.

In contrast, Indian music revolves around an individual and it mostly involves monophonic instruments. It’s not restricted by conditions related to other simultaneous sounds. It’s about creating a structure in time. Expression doesn’t happen across the spectrum in static time, but as a defined, narrow frequency band continuously moving along the spectrum. It is by exploiting this temporal movement that Indian music finds its expressive power.

Digital music, and electronic music in general is a product of the western classical and art music traditions. Naturally most of the electronic music products in the market align more with western music. Until two or three years ago, the availability of even sampled Indian instruments were a rarity among virtual instrument and synth vendors. While Harmonium is currently widely used in Hindustani music, Piano key arrangement and playing techniques are not very conducive to produce Indian music.

Another aspect of electronic art music is that, a large portion of it is still produced in non-live settings. Even in the live EDM, majority chose live remix/DJ strategies than creating new and improvised elements. There are, of course, exceptions, but since tool makers usually fulfill the requirements of majority users before attending to the niche.

So, for a non-commercial musician like me, there were not a lot of options to work purely in digital domain, especially in a live setting.

In the last couple of years, there have been a few instruments that try to get away from the traditional piano keyboard approach to MIDI controllers for digital music. In general, these instruments provide a vastly enhanced level of interactivity. While it is already possible to manipulate and control large number of parameters of virtual instruments and sound modules, the way to interact with them were at best cumbersome. One could setup control surfaces with many buttons, pods and sliders, but remembering and fluidly using them live is a tremendously difficult task. If one add to the demands of improvising something like an Indian raaga, this very soon become impractical.

The key to playability is easy and contextually clear access to various parameters of the sound. This means that, I should be able to access and modify the parameter in an intuitive way that doesn’t require me to come up with difficult to execute workflows.

As an example, lets assume we want to play a sampled violin. In a traditional keyboard setup, I will use the piano keys to communicate the note and its initial velocity using my fingers, use the expression pedal for controlling dynamic expression, additional key switches on the keyboard for selecting the right stroke or sound, pitch bend lever or wheel, mod wheel to control timbre, pedal, a few buttons on the controller to change additional characteristics of the sound like the bow pressure or speed etc. Not only that we run out of limbs pretty fast, it also causes a huge information and perception load to effectively utilize.

If one looks at how a violinist plays the physical violin, she is manipulating several things at the same time as well. However, everything a violinist has to do is within the immediate context. There is no need to take your hands off and tweak a knob to change the pressure of the bow, or use a different mechanism to play a legato or portamento. The resulting workflow is much less cumbersome and easier to remember and develop muscle memory. This is crucial when producing original music, especially in an improv setting, as most of the brain is working on the evolving structure of the music rather than the details of workflow.

In a follow up article, I will describe how some of these new protocols, virtual instruments and hardware controllers come together to create a setup that is very close in its interactivity and expressiveness to some of the traditional physical/analog instruments.

0, 1, infinity: a digital art music tour

It is only recently digital music production capabilities start to provide means for expressing nuances of Indian music. This is true with software and tools, virtual instruments and the means of human-computer interaction.

The most commonly used music creation tools, viz., sequencers/daws provide an interface and workflow that suites creation of orchestral compositions using western musical idioms quite efficiently. Many tools provide advanced capabilities for transforming and creating novel forms that adhere to the principles of western music. Indian music, which is pretty archaic in its representation and even codification*

Virtual instruments (I mostly use VSTs these days) is another area of digital music that has expansive support for traditional, conventional and novel instruments and sounds and many of them are eminently playable. However, the same design and implementation decisions that is heavily biased towards western music production makes them rather clumsy to use in an Indian music context. It is true there are Indian instruments available as VSTs (typically packaged under “ethnic”, “world” or “exotic” category!), but their playability, especially live playability is quite suspect. Some instrument developers try to compensate for the limitations of the daw and keyboards by providing prebuilt transitions and phrases. However, I find them of very limited use since my compositions need more than just occasional riff of an “exotic” instrument.**. Recently, however, more VSTs that consider such dynamic and nuanced articulation and provide a way to execute them even in live setup more or less naturally. Many of the new modelled instruments (as against the sampled ones) seem to have a better handle in this at this time.

Influence and adaptation of instruments from other musical cultures including the European to Indian music has been going on for some time. However, major impact in musical expression itself started happening the wide spread adoption of orchestral instruments as the background for popular music. Harmonium is another one of those instruments that changed the way Indian music is expressed. MIDI keyboards and other controllers for interacting with digital music tools were a direct copy of piano keyboards with some additional capabilities. However, except for some fringe, experimental ones, none of them provided a way to provide nuanced tonal control that Indian music demands. This was the case until a few years ago when a new class of MIDI instruments stated to appear in the market. These are collectively referred to as MPC (Multi-dimensional polyphonic controllers). A new extension for the age old MIDI standard to accommodate larger amount of per note data to support these controllers were also developed alongside. The result is several new MIDI devices that even look drastically different from standard piano keyboards that came to the market in last two/three years.

One such instrument is Linnstrument, which has a matrix layout of keys instead of the linear piano one. This is similar to the fret layout of many string instruments. It also provides four different parameters to be controlled separately for each note, viz, velocity, pressure, timbre and pitch. This is much closer to what a physical instrument like sitar or violin provides. Breath and bite controller from TEControls is another device, which while not provide note data, is capable of capturing x-y movements and bite pressure along with breath.

0, 1, infinity is in some sense, celebration of these tools and methodologies available to produce digital music that includes the extensive melodic nuances of Indian Music. It is also my journey from being an analog bamboo flautist to a purely digital musician.

The video above is from premier performance of the tour at David Hall, Fort Kochi. The tour will continue till March exploring more and more aspects of these new possibilities. This is movement 3 from the tone poem named “Night in the Meadow”.

*While Indian classical music, especially Karnatic has very strong body of formalizing, this is more about the static structure of music rather than a dynamic performance. Thus the gamakas (meend), exact durations, microtonal assignments etc. are left out of the representation system. So, a written version of Indian classical music only gives an outline.

**The situation is better for percussive instruments though. There are excellent Indian percussion libraries available. My complaint is mostly about melodic instruments, as this is where Indian music drastically differs from other musical expressions.

Pixie Flux theory of Quantum Consciousness

In a video by Sixty Symbols (see below), Prof. Moriarty while giving a dress down of Dr. Lanza and his theory of quantum woo mentions that one could postulate pixies coming in and out of existence to create spooky effects. I think this is a serious proposition. One could come up with a hypothesis without violating any of the laws that we currently hold that, this is what actually happens.

Prof Moriarty on Quantum Woo

This is how it works. Since anything can happen without violating any physical laws within plank time (like the creation of annihilation of particle-antiparticle pairs), one could hypothesize that, there are magical Pixies that come into and poof out of existence under 10-43 seconds. These Pixies are the ones that maintains the reality and causes all kinds of quantum spookiness. The Pixies are also carriers of consciousness. Well, actually consciousness is produced by the fluttering of the Pixie wings.

This hypothesis is named the Pixie Flux Hypothesis© of Quantum Consciousness. I am claiming the copyright for this hypothesis. Unlike actual theories of physics, for which I can write a paper, get it peer reviewed and published, this can only be ascertained by brute force of an enlightened mind (mine).

Creative Commons License
Pixie Flux Hypothesis by Salim Nair is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Based on a work at


My visit to India after a year was, as always filled with situations very close to what Tim Minhin, inimitably showing in this video. Enjoy

Tim Minchin: STORM

ഇവിടുത്തെ ആദ്യത്തെ മലയാളം പോസ്റ്റിങ്ങ്‌!

കുറെ നാളായി ഞാന്‍ മലയാളത്തില്‍ ഒരു പോസ്റ്റ്‌ ചെയ്യണം എന്ന് വിചാരിച്ചുകൊണ്ടിരിക്കുന്നു. പക്ഷെ കൊള്ളാവുന്ന ഒരു IME എഡിറ്റര്‍ കിട്ടിയിട്ടാവട്ടെ എന്ന് കരുതി. അവസാനം ഇന്നലെ എന്‍റെ ചേച്ചി എനോട് ഒരു സഹായം ചോദിച്ചു, ഒന്ന് രണ്ടു ലേഖനങ്ങള്‍ PDF-ല്‍ നിന്ന് കണ്‍വേര്‍ട്ട്  ചെയ്യാന്‍ പറ്റുമോ എന്ന്. അപ്പോഴാണ്‌ മലയാള ലിപികളുടെ confusion ഒന്നുകൂടെ എനിക്ക് ഓര്‍മ്മ വന്നത്.

വര്‍ഷങ്ങള്‍ക്ക് മുന്‍പ് ഞങ്ങളുടെ ആദ്യകാല കമ്പനി ഒരു മംഗ്ലീഷ് എഡിറ്റര്‍ ഉണ്ടാക്കാന്‍ തീരുമാനിച്ചു. പക്ഷെ അതിനു വേണ്ട ഫോണ്ടുകളുടെ കാര്യം തപ്പാന്‍‍ തുടങ്ങിയപ്പോള്‍ ആണ് കാര്യങ്ങളുടെ കിടപ്പ്‌ ബോധ്യമായത്. നൂറു കണക്കിന് (ok, പത്തു കണക്കിന്) DTP സോഫ്റ്റ്‌വെയറുകള്‍ മാര്‍ക്കറ്റില്‍ അന്നേ തന്നെ ഉണ്ടായിരുന്നു. പക്ഷെ ഓരോന്നിനും അതിന്‍റെ മാത്രം ഫോണ്ടെ ഉപയോഗിക്കാന്‍ പറ്റുള്ളൂ. പലപ്പോഴും ഇത് സ്വാധീനം ഉറപ്പാക്കാനുള്ള ഒരു വഴിയായാണ് ഉപയോഗിച്ചിരുന്നത്. എന്‍റെ സോഫ്റ്റ്‌വെയര്‍ ഉപയോഗിച്ച് തുടങ്ങിക്കഴിഞ്ഞാല്‍ ‍ you are stuck with it.

ഏതായാലും ഒരു സംഗതി മനസ്സിലായി, ഇത്രയും കാലമായിട്ടും കാര്യങ്ങള്‍ക്ക് വലിയ മാറ്റമൊന്നും ഉണ്ടായിട്ടില്ല.

ഞാനും ശോഭയും കൂടി ഇന്നലെ രാത്രിയും ഇന്ന് ഇതുവരെയും കണ്‍വേര്‍ട്ട് ചെയ്യാനായുള്ള വിവിധ മാര്‍ഗങ്ങള്‍ തപ്പിക്കൊണ്ടിരിക്കുകയായിരുന്നു. പത്തു പതിനഞ്ചു ഫോണ്ട് ഡൌണ്‍ലോഡുകളും അസംഖ്യം ബ്ലോഗ്‌ പോസ്റ്റിങ്ങുകളും കഴിഞ്ഞ് ഒടുവില്‍ ശോഭ Google transliterate project കണ്ടെത്തി.

അവിടെ നിന്ന് IME എഡിറ്ററിലേക്ക് അധികം തപ്പല്‍‍ വേണ്ടി വന്നില്ല. എനിക്കറിയാം ഈ പ്രൊജക്റ്റ്‌ അത്ര പുതിയതൊന്നും അല്ല എന്ന്. എന്നാലും ചിലപ്പോള്‍ നമുക്ക് നല്ല ഒരു കാരണം വേണം കാര്യങ്ങള്‍ കാണാന്‍.

ഞങ്ങളുടെ തപ്പളിനിടയില്‍ പല മലയാളം മോഴിമാറ്റ സംവിധാനങ്ങളും കണ്ടിരുന്നു. പക്ഷെ അവയ്ക്കൊന്നും ഒരു IME എഡിറ്ററിന്‍റെ ഉപയോഗത കണ്ടില്ല. എന്തെങ്കിലും എഴുതുന്നതിനു മുമ്പ് ഒരു web browser തുറക്കണം എന്ന് വച്ചാല്‍ പ്രശ്നമാണ്. എനിക്ക് വേണ്ടിയിരുന്നത് OS-ഉം ആയിട്ടുള്ള സമ്പൂര്‍നമായ സഹപ്രവര്‍ത്തനം ആണ്.

കഴിഞ്ഞ അര മണിക്കൂര്‍ ടൈപ്പ് ചെയ്തതു നോക്കിയിട്ട് ഇത് പ്രായോഗികമാണ്, തികച്ചും. IME വാക്കുകള്‍ തിരഞ്ഞെടുക്കുന്നതിന് ചില കുണ്ട്രാണ്ടികള്‍ സംഭവിക്കുന്നുണ്ട്. കൂടുതലും ചില്ലക്ഷരങ്ങളും ചില കൂടക്ഷരങ്ങളും ഉപയോഗിക്കുന്നതില്‍. പക്ഷെ മിക്കവറും എല്ലാ സന്ദര്‍ഭങ്ങളിലും അധിക സമയം ചിലവഴിക്കാതെ തന്നെ എനിക്ക് ശരിയായ മാര്‍ഗ്ഗം കണ്ടുപിടിക്കാന്‍ കഴിഞ്ഞു.

മലയാള വ്യാകരണത്തിന്‍റെ കാര്യം വേറൊന്നാണ് – ഞാന്‍ മറന്നു പോയി. പല വാചകങ്ങളും എനിക്ക് മാറ്റിയെഴുതെണ്ടി വരുന്നുണ്ട്. അതുകൊണ്ട് തന്നെ ഇനിയുള്ള ദിവസങ്ങളില്‍ ഞാന്‍ കുറെ മലയാളം എഴുതാന്‍ പോവുകയാണ്.

For my non Malayalam readers, here is my language. There will be more like this coming in because I found this nice IME editor for my language (Google IME)

Chevy Volt has WHA!?

It has 10 million lines of code, exalts Wired magazine, making it the King of software cars.

Now, if we look at kings in real life, who are overly pompous, over spending, ego maniacal AHs, this title may suite.

Oh, no, I have nothing against Volt. I would buy one (or another Plug-in) when it is time for my next car. But there had been stories similar to this taking rounds in the web. It looks like the manufactures want to make this a story. Look at us, we have 10 million lines of code!

First of my problems with this statement is, lines of what code? Since lot of the code are controller and other microprocessor code, it could be just instructions. In that case, 10 M instructions is not that big a code base. It might be a good thing to have only that much. But if this is 10 M lines of code in some high level language like C, things look much different. Then the question is why the heck so much lines of code!

There was a time when people used to boast about the numbers of lines of code in their code base. There were even places which used to pay per lines of code. But, if someone these days try to bring up the number of lines of code with a sense of achievement, unless it is to show how few lines there is, it is unimpressive.

A lot of people assume that all the work in creating a software product is in finishing it up for the first release. But, the fact is, it is only the beginning of work. Not just fixing bugs but also keeping up with user demands for new features, accommodating new scenarios etc. So, the total cost of a software development process is overwhelmingly decided by its maintenance costs.

That is why the best software product is the one that does not have any lines of code. So, Chevy Volt is 10 million times worse than the theoretical best Open-mouthed smile

Fallacy of Exaggeration

In an article about the brain (a wonderful article) Carl Zimmer has this quote.

The brain is, in the words of neuroscientist Floyd Bloom, “the most complex structure that exists in the universe.”

Now, I understand the awe we feel looking at the complexity and ingenuity of our brain, but the most complex structure in the universe?! Biological systems tend to accumulate entropy to create complex systems, but there are many other phenomenon in the universe that has much more entropy. Think about the nuclear fusion in the center of starts, and how about the super massive black hole! Now think about possibilities of life (as self replicating, self regulating systems capable of building complexity) in the rest of the universe. Some of it might have had billions of years to evolve. May be there are sentient networks that cover whole solar systems.

How do you even conceive even a thought experiment to verify that statement?

I have to disclose that I did not read Bloom directly and doesn’t know the larger context from where it is taken.

Such statements, like any other unverifiable statements, should be avoided when someone writes about science.

Who is she/he?

Today evening, when I got back from work, we found her/him in our backyard. First I thought this was a ground hog, but in closer inspection, it looks more like an overgrown rat. Can anyone identify?




Update: Just found out. She is an Opossum. No wonder, she is quite lethargic! Unfortunately, most of the links I could find are for removal of them. Since she is in my backyard, I don’t care. She is not that ugly!

Create a website or blog at

Up ↑