MIDI: The Handshake Between Rivals

The Night in the Hotel Room

In January of nineteen eighty two, Dave Smith walked back to his hotel room at the Anaheim Hilton convinced that the whole thing was dead. He had just come from a meeting at the winter NAMM show, the National Association of Music Merchants convention, the music industry's biggest trade gathering, where representatives from a dozen synthesizer companies had gathered in a conference room to discuss his proposal for a universal interface that would let any electronic keyboard talk to any other. The meeting had been a disaster. Some manufacturers wanted an expensive high speed connection. Others did not see the point of an interface at all. A few openly worried that compatibility with competitors would destroy their market advantage. The room could not agree on anything, and Smith, a thirty one year old engineer from San Jose who had already revolutionized the synthesizer industry once, left thinking he had wasted his time.

Later that evening, there was a knock on his hotel room door. A Roland engineer named Sakai stood in the hallway. He had been sent by his boss, Ikutaro Kakehashi, the founder and president of Roland Corporation. Kakehashi was Smith's direct competitor. Roland made synthesizers that competed head to head with Smith's Sequential Circuits products in every music store on earth. And yet Kakehashi's message, delivered through his engineer across the hallway of a convention hotel, was simple. Do not give up so easily.

That knock on a hotel door is the reason every electronic instrument on earth speaks the same language today. Within fourteen months, Smith and Kakehashi would shepherd a specification through an industry full of rivals and skeptics, connect two competing keyboards on a convention floor to gasps of amazement, and give the world a protocol so robust that it would survive for more than four decades on version one point zero. They never patented it. They never charged a licensing fee. They gave it away, free, to an industry that included their own competitors. The protocol was called MIDI, the Musical Instrument Digital Interface, and the story of how two rivals chose cooperation over lock-in is one of the most consequential design decisions in the history of creative technology.

The Prophet from San Jose

To understand why Smith was the person standing in that hotel room, you need to go back to a garage in San Jose in nineteen seventy four. Dave Smith was born in San Francisco on April second, nineteen fifty, the kind of kid who took things apart to see how they worked. He studied computer science and electronic engineering at UC Berkeley, played piano in college bands, and graduated into the first generation of engineers who understood both music and microprocessors. In nineteen seventy two, he bought one of the first three hundred Minimoog Model D synthesizers ever made, and he fell into the gap between what the instrument could do and what he wanted it to do.

The Minimoog was monophonic. One note at a time. You could play a screaming lead line, but you could not play a chord. And every sound you created was ephemeral. The knobs and switches on the front panel controlled the oscillators, the filters, the envelopes, but there was no way to save a setting. You spent twenty minutes dialing in the perfect sound, and then you moved a single knob by accident and it was gone forever. Every synthesizer in the world had this problem. No memory. No presets. No way to recall a sound you had made yesterday.

Smith was working a day job programming microprocessors, the tiny general purpose computers that were just beginning to appear inside consumer electronics. He had an idea that now seems obvious but at the time was genuinely radical. Put a microprocessor inside a synthesizer. Use it to remember the position of every knob and switch. Let the musician store sounds and recall them at the push of a button. He assumed that Moog or ARP, the big established synthesizer companies, would build this instrument any day. He waited. Nobody did.

In early nineteen seventy seven, Smith quit his job and began working full time on a design he called the Prophet-5. Five voices of polyphony, meaning five notes at once, enough to play chords. Forty programmable memory slots, each storing the complete state of every parameter on the instrument. He demonstrated a prototype at the NAMM show in January nineteen seventy eight and shipped the first units later that year. The Prophet-5 was the first fully programmable polyphonic synthesizer in the world, and it changed everything.

The list of musicians who used the Prophet-5 reads like a history of popular music in the early nineteen eighties. Michael Jackson. Peter Gabriel, who called it his old warhorse. Depeche Mode. Kraftwerk. Phil Collins. Madonna. Vangelis. The warm, fat, programmable sound of the Prophet-5 became the sound of a decade. Smith's company, Sequential Circuits, grew from a garage operation to a major manufacturer.

But Smith was already thinking beyond the instrument itself. In his studio, he had a Prophet-5 and he had synthesizers from other companies, an Oberheim, a Roland, a Moog, and none of them could talk to each other. Each manufacturer had its own proprietary interface, or no interface at all. If you wanted two synthesizers to play the same note at the same time, your options were limited to control voltage and gate signals, an analog method that worked unreliably across brands because every manufacturer used different voltage standards. Roland used one volt per octave with different gate polarity than Sequential. Moog used yet another standard. Oberheim had its own parallel bus. Connecting them required custom cables, adapter boxes, and a degree in electrical engineering.

I do not look for what would have the best sales. I build what feels right to me.

What felt right to him in nineteen eighty one was the idea that any synthesizer should be able to talk to any other synthesizer, regardless of who manufactured it. An open, universal, royalty free standard. He started writing a specification.

The Watchmaker from Osaka

Six thousand miles away, in Hamamatsu, Japan, a man twenty years older than Smith was thinking exactly the same thing, and he had been thinking it for much longer. Ikutaro Kakehashi was born on February seventh, nineteen thirty, in Osaka. His parents died of tuberculosis when he was two years old. He was raised by his grandparents in a country that was sliding toward war.

Much of his childhood was spent in the shadow of conflict. At fourteen, he was sent to work at the Hitachi shipyards in Osaka. During the war, American bombing destroyed his home. When the war ended and the occupying forces arrived, Kakehashi was fifteen years old, orphaned, homeless, with no money and no prospects in a devastated country. He tried to get into university but was rejected on health grounds. At sixteen, he moved to the southern island of Kyushu, alone, and did what survivors do. He looked around, saw what was needed, and taught himself a trade.

In nineteen forty seven, at seventeen, Kakehashi opened the Kakehashi Clock Store, a tiny watch repair shop. Post-war Japan was full of broken timepieces and no one to fix them. He taught himself the craft. He soon expanded into repairing radios, and through the radios he discovered his passion. Not for electronics, exactly. For music. Kakehashi had no musical training. He could not play an instrument. But he loved listening, and he understood with an engineer's clarity that the gap between wanting to make music and being able to make music was a technical problem. A problem he could solve.

Then tuberculosis found him too, the same disease that had killed his parents. At twenty, Kakehashi was hospitalized. He spent three years in a sanitarium, weak, uncertain if he would survive. He became a test patient for an experimental antibiotic called streptomycin, one of the first antibiotics effective against tuberculosis. The drug worked. Kakehashi recovered, carrying with him the knowledge of how close he had come to following his parents into an early grave and a determination to spend whatever time he had on something that mattered.

He founded a company called Ace Tone in nineteen sixty to build electronic organs and rhythm machines. Ace Tone's FR-1 Rhythm Ace was one of the earliest commercial drum machines, and it found its way into pop recordings through the late nineteen sixties. But Kakehashi's ambitions outgrew Ace Tone. In nineteen seventy two, he founded Roland Corporation, and over the next decade, Roland produced a string of instruments that reshaped popular music. The TR-808 drum machine, whose booming synthetic bass drum and crisp hi-hat would become the rhythmic backbone of hip-hop, electro, and house music. The TR-909, which did the same for techno. The TB-303 bass synthesizer, designed as a practice tool for guitarists that nobody wanted, until it was rediscovered years later and became the defining sound of acid house. The Jupiter-8 and Juno-60 synthesizers, workhorses of the new wave era.

Graham Massey of 808 State, a band that literally named itself after a Roland product, put it this way.

The Roland gear began to be a kind of Esperanto in music. The whole world began to be less separated through this technology, and there was a classiness to it. You could transcend your provincial music with this equipment.

But Kakehashi faced the same frustration Smith did. His instruments lived in isolation. Roland had developed its own proprietary digital interface called DCB, the Digital Control Bus, which debuted on the Juno-60 in nineteen eighty one. It was technically clever, offering control of filters and amplifiers in addition to simple note values. But it only worked with other Roland instruments. And Kakehashi, the orphan watchmaker who had spent his career trying to make music accessible to everyone, understood something that most of his competitors did not. A walled garden is the enemy of music. If instruments cannot talk to each other, musicians cannot create freely.

The Paper, the Meeting, and the Arguments

At the NAMM show in Chicago in June of nineteen eighty one, Kakehashi approached Tom Oberheim, the founder of Oberheim Electronics, about the idea of an international industry standard for synthesizer communication. Oberheim, who had already built his own proprietary parallel bus for connecting his instruments, was interested but knew he was not the right person to write the specification. He suggested Kakehashi talk to Dave Smith.

Smith visited Oberheim's office in late summer of nineteen eighty one to preview a paper he was writing. With his colleague Chet Wood, Smith had drafted a specification for something he called the Universal Synthesizer Interface, or USI. It proposed a serial digital connection running at nineteen thousand two hundred baud, using standard quarter inch phone jacks, the same kind of cable used for guitar leads. It was a straightforward proposal. Simple. Cheap. Easy to implement.

Smith and Wood presented the USI paper at the Audio Engineering Society convention on October first, nineteen eighty one, in New York. The paper was titled simply The USI, or Universal Synthesizer Interface. The reception was encouraging. Here, finally, was a concrete technical proposal that could solve the isolation problem. But a paper is not a standard. For a standard to work, competitors have to agree on it. And competitors, by definition, do not like agreeing with each other.

Which brings us back to that conference room at the January nineteen eighty two NAMM show in Anaheim. Smith and Kakehashi organized a meeting with representatives from the major synthesizer manufacturers. Sequential, Roland, Oberheim, CBS Rhodes, Yamaha, E-mu Systems, Korg, Kawai, Octave Plateau, Passport Designs, and several more. American companies. Japanese companies. Big players, small players. The agenda was straightforward. Review Smith's USI proposal. Get everyone on board. Build the future.

The meeting collapsed, and it collapsed along specific fault lines. Dave Rossum, the co-founder of E-mu Systems and one of the sharpest engineering minds in the room, argued that Smith's serial approach was fundamentally too slow. E-mu wanted something closer to an RS-422 serial protocol running at much higher speeds, or better yet, an existing networking standard.

Why not just use Ethernet? It is fast, it exists, and it is only about ten dollars to implement.

Rossum's argument was technically sound. Ethernet was faster. But it was also more expensive, more complex to implement, and required every manufacturer to adopt networking hardware that most synthesizer engineers had never touched. Tom Oberheim pushed for a fast parallel bus system, similar to the one he had already built for his own instruments, which could transfer data at much higher speeds than any serial connection. Other manufacturers thought even Smith's proposed serial speed was too fast for reliable operation with home computers. And some companies did not see the point of an interface at all. Why should we make it easy for our customers to leave?

The fundamental tension in the room was the tension at the heart of every standards negotiation. The companies who had invested heavily in proprietary interfaces, companies like Oberheim with his thirty seven pin D-Sub connector and Roland with its Digital Control Bus, had real reasons to resist. A universal standard would make their proprietary investments worthless. It would let a musician connect a competitor's keyboard to their equipment. And a lowest common denominator standard, which is what a serial protocol looked like to engineers who wanted parallel speed, would limit what the interface could do. These were not irrational positions. They were the reasonable arguments of companies protecting genuine competitive advantages. Smith walked out discouraged, convinced that the project was dead, and then Kakehashi sent Sakai to his hotel room with a message that changed everything.

Building the Standard Across an Ocean

The message from Kakehashi was not just do not give up. It was a signal that Roland, one of the two most powerful synthesizer companies in Japan, was committed to the project regardless of what the others did. And it was a signal from a man who understood, better than anyone in that conference room, that cooperation creates larger markets than competition protects.

Think about what Kakehashi was risking. Roland had its own proprietary interface, DCB, already shipping on products. A universal standard would make DCB obsolete. It would let musicians connect a Roland keyboard to a Sequential Circuits synthesizer, or a Yamaha drum machine, or an Oberheim sequencer. It would destroy whatever lock-in Roland had built. And Kakehashi chose to do it anyway, because he believed that a world where all instruments could talk to each other was a world where more people would make music, and a world where more people made music was good for everyone, including Roland.

The next morning, the dynamic shifted. The Japanese manufacturers, Roland, Korg, Yamaha, and Kawai, met with Smith separately. All four were willing to continue. With the American and European companies still arguing among themselves, the work moved forward as a collaboration between Sequential Circuits in California and the four Japanese companies, with Roland serving as the liaison and coordinator on the Japanese side.

What followed was a year of faxes, letters, and occasional face to face meetings across six thousand miles of ocean. Smith described the process years later with characteristic understatement.

We worked via fax. Sequential in the USA and Roland coordinating in Japan. And a couple of meetings along the way, in Japan and California.

The Japanese companies agreed to meet about once a month to respond to and improve on Smith's initial design, then relay their proposals back to California through Roland's fax machine. The exchanges were substantive. On July twenty third, nineteen eighty two, a fax arrived from Roland with a proposal that would reshape the entire specification. The Japanese companies had agreed that every channel should be either monophonic or polyphonic, and that the number of channels should be increased from eight to sixteen. Smith's original USI had proposed eight channels, enough for eight notes of polyphony. But Sequential was already planning the Prophet-10, a polyphonic synthesizer with ten voices, and eight channels would not be enough. The Japanese engineers saw the same problem and pushed the number higher.

Karl Hirano from Yamaha designed a grounding scheme using optoisolation to prevent electrical noise and ground loops between connected instruments. This was an elegant engineering choice born from practical experience. Anyone who had ever connected two synthesizers from different manufacturers knew the problem. Different power supplies, different grounding points, and suddenly an audible hum bleeding through every output. Hirano's solution was to make each MIDI connection electrically isolated. A tiny LED inside the receiving port would flash in response to the incoming signal, and a photosensor on the other side of a gap would read the light. No electrons crossed the gap. No ground loops. No hum. You could connect instruments from different manufacturers, plugged into different power outlets on different continents, and the signal would pass clean.

The Japanese companies also proposed replacing the quarter inch phone jack with a five-pin DIN connector, a round plug that was not commonly used for audio equipment. The choice was deliberate. A unique connector meant you could never accidentally plug a MIDI cable into the wrong socket. It was a connector specifically for this purpose and nothing else. The baud rate changed too. Smith's original USI proposed nineteen thousand two hundred baud. The final specification settled on thirty one thousand two hundred and fifty baud, a number that looks odd until you realize it is an exact division of one megahertz, the common clock frequency of the microprocessors used in synthesizers at the time. This meant any manufacturer could generate the MIDI clock signal precisely without expensive custom hardware. Just divide your processor clock and you get the exact baud rate. It was a choice that prioritized ease of implementation over raw speed, and it was the right choice. A protocol that is easy to build into products gets built into products.

The data format was equally pragmatic. Each byte had eight bits, but the first bit was reserved as a flag to distinguish status bytes from data bytes. That left seven bits for values, which gives you a range of zero to one hundred and twenty seven. One hundred and twenty eight possible notes. One hundred and twenty eight levels of velocity, how hard you hit the key. One hundred and twenty eight values for any continuous controller, a mod wheel, a pitch bend, a sustain pedal. In nineteen eighty two, when synthesizers were just learning to play more than one note at a time and the Yamaha DX-7 had only thirty two parameter positions, one hundred and twenty eight felt like more than enough. It was not. But it was good enough. And good enough, shipped on time, beats perfect, shipped never.

The name changed too. Universal Synthesizer Interface became Musical Instrument Digital Interface. MIDI. The Japanese companies had initially suggested UMI, the Universal Music Interface, pronounced you me, which had a charming collaborative ring to it. Smith countered with MIDI, and the name stuck. Someone involved, the records are not entirely clear who, had realized that the word synthesizer was too narrow. This was not just for synthesizers. This was for any musical instrument. Drum machines, sequencers, samplers, eventually computers. The name change was more prophetic than anyone involved could have known.

Five Minutes at the Roland Booth

On January eighteenth, nineteen eighty three, at the winter NAMM show in Anaheim, California, Dave Smith carried a Sequential Circuits Prophet-600 to the Roland booth. The Prophet-600 was the younger sibling of the Prophet-5, a more affordable synthesizer that happened to be the first instrument in the world with a MIDI port built in. Waiting at the Roland booth was a Jupiter-6, Roland's flagship polyphonic synthesizer, which had also been fitted with MIDI. A five-pin DIN cable connected them.

John Bowen, Sequential's head of sound design, was there. He had been up late finishing the presets for the Prophet-600 and was nervous. They had tested the Prophet-600's MIDI connecting to another Prophet-600 and it worked. But connecting it to a Roland Jupiter-6, an instrument made by a different company with different engineers and different firmware, was something they had never tried. Nobody had. That was the entire point.

Smith nearly did not make it to the demonstration. He arrived late, and according to accounts of the day, had trouble getting past security at the Roland booth because he did not have a NAMM property pass. Here was the man who had invented the protocol, carrying one of the two instruments about to make history, and he was being waved away at the entrance. Someone sorted it out, and Smith set up the Prophet-600 beside the Jupiter-6.

He pressed a key on the Sequential keyboard. The Roland played the same note. He pressed a chord, and both instruments sounded together, the warm analog voices of two rival companies singing in unison through a cable that cost a few dollars.

He moved the pitch wheel. Both responded. For the first time in history, two instruments from two competing companies were speaking to each other in real time, through a standard that either company could have vetoed, through a protocol that neither company would ever charge a cent to use. In the iconic photograph from that moment, John Bowen stands with his back to the camera in a red Sequential Circuits jacket. Beside him is Roland's Jim Mothersbaugh. Dave Smith leans over the keyboard. The expressions in the room are what you would expect when people see something that they understand will change their industry.

Thirty years of version one point zero shows we got it right.

MIDI was published as a complete specification in August nineteen eighty three. It was free. No patents. No licensing fees. No royalties. Any manufacturer could implement it in any product without asking permission or paying anyone. The MIDI Manufacturers Association, formed in nineteen eighty five, would maintain the specification as a nonprofit, funded by membership dues, not by gatekeeping the technology. This was, as one retrospective put it, an incredible achievement substantially unmatched to this day.

The Imperfect Standard That Won

MIDI's critics arrived almost immediately, and they were not wrong. Thirty one thousand two hundred and fifty baud is slow. A single MIDI note-on message, the instruction that says play this note at this velocity on this channel, takes about one millisecond to transmit. That is fast enough for a solo keyboard. But if you are playing a ten-note chord across sixteen channels while sending control changes for pitch bend, modulation, aftertouch, and sustain, the messages queue up and you get latency. Notes that should sound simultaneously arrive a few milliseconds apart. Performers call this MIDI choke, and for decades it was the most common complaint about the protocol. Dave Rossum, the E-mu engineer who had wanted Ethernet in that first meeting, could reasonably claim that he had warned them.

The seven-bit resolution was another limitation. One hundred and twenty eight velocity levels sounds like a lot until a concert pianist tells you that the difference between a whisper and a shout on a Steinway involves gradations of force that one hundred and twenty eight steps cannot capture. One hundred and twenty eight positions on a continuous controller means that turning a knob smoothly produces a staircase of discrete values instead of a continuous curve. For synthesizer parameters like filter cutoff frequency, the steps were sometimes audible, little zipper noise artifacts as the value jumped from one step to the next.

The protocol was unidirectional. Data flowed one way, from MIDI out to MIDI in. If you wanted bidirectional communication, you needed two cables. And there was no way for a device to announce its capabilities. A MIDI keyboard had no way of telling a connected synthesizer how many keys it had, what controllers were available, or what it expected in return. You just plugged things in and hoped for the best.

These were real limitations. But here is the design insight that Smith and Kakehashi, whether by brilliance or by luck, got exactly right. A standard does not need to be perfect. It needs to be implementable. It needs to be cheap. It needs to be unencumbered. And it needs to ship. MIDI was all of those things. The five-pin DIN connector and the optoisolator circuit cost pennies. The serial protocol could be handled by any microprocessor with a built-in universal asynchronous receiver transmitter, the standard serial communication hardware that was essentially ubiquitous by the early nineteen eighties. The specification was short enough to read in an afternoon and simple enough that a single engineer could implement it in a week. Within two years, every major synthesizer manufacturer on earth was shipping instruments with MIDI ports.

MIDI works because it is simple and free, which led to it being ubiquitous.

And ubiquity, once achieved, is nearly impossible to undo. By the mid nineteen eighties, a musician could walk into a studio with a Roland drum machine, a Yamaha synthesizer, a Sequential Circuits keyboard, and an Oberheim expander, connect them all with inexpensive MIDI cables, and control the entire rig from a single keyboard or a computer running sequencer software. This had never been possible before. The creative explosion that followed, the sophisticated layered productions of the late nineteen eighties and nineteen nineties, the rise of the home studio, the democratization of music production, all of it was built on that simple, imperfect, free protocol.

What Happened to the Handshakers

The protocol survived. The companies and the people behind it had harder journeys.

Sequential Circuits began to struggle in the mid nineteen eighties. Smith, the engineer who built what felt right to him, had pushed his small company into computer audio products, chasing the promise of a huge mass marketing opportunity for synthesizers in multimedia. It was the right idea at the wrong time. The market was not ready, and Sequential did not have the capital to wait. By the fall of nineteen eighty seven, things were desperate. The company occasionally could not make payroll on time, and employees were asked to hold their checks for a few days while money came in. Sequential released one final product, the Prophet 3000, a sophisticated sixteen-bit sampler that was arguably ahead of its time. Only several dozen units were produced before the doors closed.

We were too small and undercapitalized, and we were a few years too early in the market. It drained our resources, so by the time we pulled back to professional instruments, it was too late.

Yamaha bought the remains of Sequential in December of nineteen eighty seven and formed a small team called the Dave Smith Division to continue development. Within a year, Yamaha shut that division down too. The man who had invented the protocol that every instrument on earth now spoke, the man who had chosen to give it away for free instead of charging a licensing fee that would have made him wealthy, watched his company die and his brand disappear into the archives of a Japanese conglomerate. Smith and several members of his original team, including John Bowen, were picked up by Korg. In May of nineteen eighty nine, they founded the Korg R&D group in California, where Smith helped design the Wavestation synthesizer. It was good work. But it was not his name on the front panel.

Thirteen years passed. In two thousand two, at the age of fifty two, Smith founded Dave Smith Instruments and began building new synthesizers under his own name. The Evolver, a hybrid analog and digital instrument, was his comeback. He could not use the Sequential name because Yamaha still owned the trademark. But Smith was building again, which is what mattered.

Kakehashi stepped down from Roland gradually. In two thousand one, he resigned as chairman and became a special executive adviser. In two thousand two, he published his autobiography, titled with the phrase that defined his life. I Believe in Music. The book told the story of the orphaned watchmaker who built an instrument empire, and it sold widely in the music technology world. In two thousand thirteen, after a reported clash with Roland's management, Kakehashi left the company entirely. He was eighty three years old, and most people would have retired. Instead, he founded a new company called ATV Corporation, focused on audiovisual electronics. He launched his first ATV products, electronic drums and video converters, in November of two thousand fifteen. He was still building instruments.

Then, in January of two thousand fifteen, Kakehashi did one more thing in the spirit that had defined his entire career. He personally contacted the president of Yamaha and recommended that Yamaha return the Sequential Circuits trademark to Dave Smith. Yamaha agreed. After twenty seven years, the name came home.

I feel that it is important to get rid of unnecessary conflict among electronic musical instrument companies. That is exactly the spirit of MIDI.

That quote tells you everything about the man. Thirty three years after sending Sakai to a hotel room in Anaheim to tell a discouraged competitor not to give up, Kakehashi was still working to remove barriers between the companies that made instruments. It was not a business strategy. It was a belief about how the world should work. In two thousand eighteen, Dave Smith Instruments became simply Sequential again. The prophet had come home.

In January of two thousand seventeen, just months before his death, Kakehashi published a second book. An Age Without Samples, subtitled Originality and Creativity in the Digital World. The title was another declaration of philosophy. Even at eighty seven, after a life that spanned wartime devastation, tuberculosis, an industrial revolution, and the invention of a global standard, Kakehashi was thinking about the future of creative expression.

The Grammy and the Goodbye

On February ninth, two thousand thirteen, at the Wilshire Ebell Theatre in Los Angeles, Ikutaro Kakehashi and Dave Smith received a Technical Grammy Award for the invention of MIDI. The ceremony was part of the Recording Academy's Special Merit Awards during Grammy Week, a separate event from the main broadcast the following evening. It was the first time either man had been recognized by the Recording Academy. Kakehashi was eighty two. Smith was sixty two. Thirty years had passed since that first cable connected a Prophet-600 to a Jupiter-6.

Kakehashi's son, Ikuo, accepted the award on his father's behalf. Dave Stewart, the musician and producer best known for his work with Eurythmics, presented the tribute. His speech captured what MIDI meant to an entire generation of musicians.

Dave Smith and Ikutaro Kakehashi, you turned my world upside down and in doing so gave birth to a revolution that will never end.

Stewart told the audience about the day he walked into a music shop in Camden Town and saw a crowd gathered around a Sequential Circuits Prophet-600 with MIDI. Once he grasped what they were talking about, he said, he felt quite faint, his head spinning with the possibilities.

Kakehashi sent a statement to be read at the ceremony. It was formal and gracious, as was his way. But even in ceremony, his philosophy came through.

This year's Technical Grammy Award is the result of the cooperation by the companies who worked towards the same dream, growth of electronic musical instruments. I would like to share this acknowledgement with Mister Tadao Kikumoto, former research and development manager of Roland Corporation, the engineering staff, and all who were involved in creating and designing the MIDI concept and system.

Even in his moment of personal recognition, Kakehashi's instinct was to share the credit. Not to claim the invention as his own, but to name the engineers, to point at the team, to insist that this was cooperation, not individual genius. That instinct was the same one that had sent Sakai to Smith's hotel room three decades earlier. The same one that would, two years later, prompt him to call Yamaha's president about returning a trademark that was not his to return, simply because unnecessary conflict offended his sense of how the music industry should work.

Four years after the Grammy, on April first, two thousand seventeen, Ikutaro Kakehashi died at the age of eighty seven. He was still working. ATV Corporation, the company he had founded at eighty three, was producing electronic instruments and planning new products. The orphaned watchmaker from Osaka, the tuberculosis survivor, the man who had built Roland into a global empire and then walked away from it, was designing instruments until the end.

The tributes poured in from every corner of the music world. Marc Almond of Soft Cell called him the man who changed music. Martyn Ware of the Human League told the BBC that Roland was central to everything they did for the whole of their first two albums. Moog Music described him as a model of resilience and a genuine trailblazer.

Dave Smith wrote his own tribute to his old rival.

He was just an amazing man. A good friend. A very good competitor, of course. And just innovative continually all that time.

Five years after Kakehashi's death, on May thirty first, two thousand twenty two, Dave Smith suffered a heart attack in Detroit. He was there for the Movement electronic music festival, a celebration of the very kind of music that MIDI had made possible. He died at a nearby hospital at the age of seventy two. His wife Denise and their children Haley and Campbell were with him.

Sequential's statement said they were heartbroken but took some small solace in knowing he was on the road doing what he loved best, in the company of family, friends, and artists.

Both men died doing what they had spent their lives doing. Kakehashi was still building instruments. Smith was surrounded by the music his protocol made possible. Neither man ever made a cent from MIDI itself. One of tech's most successful inventors never made a cent, Fortune magazine wrote in a headline that captured the paradox. Smith could have been wealthy beyond measure if he had patented the interface and charged every manufacturer a licensing fee. Instead, he and Kakehashi chose to give it away. And that choice is exactly why it won.

Thirty Seven Years of Version One Point Zero

Here is a number that should make every software engineer pause. MIDI one point zero was published in August of nineteen eighty three. MIDI two point zero was announced at the winter NAMM show on January seventeenth, two thousand twenty. That is thirty seven years on a single version. In a technology industry that ships major updates annually and considers anything more than five years old to be legacy, MIDI one point zero was the protocol that refused to die. Not because no one wanted to improve it, but because no improvement was worth breaking what already worked.

Research on a potential MIDI update began as early as two thousand five. The work was led by a group of engineers within the MIDI Manufacturers Association, with Mike Kent, the principal architect of what would become MIDI two point zero, proposing the concept of MIDI Capability Inquiry in two thousand sixteen. Capability Inquiry was the foundational insight. Instead of replacing the old protocol, let new devices ask each other, do you speak MIDI two point zero? If the answer was yes, they could upgrade their conversation to the new protocol with higher resolution, more channels, and bidirectional communication. If the answer was silence, because the other device only knew MIDI one point zero, the new device would simply fall back to the old protocol. Every MIDI one point zero device in the world, billions of them, would keep working exactly as before.

MIDI two point zero finally offered what musicians had been requesting since the nineteen eighties. Sixteen-bit resolution for velocity, meaning sixty five thousand five hundred and thirty six levels of how hard you hit a key, instead of the original one hundred and twenty eight. Thirty two-bit resolution for continuous controllers, meaning over four billion steps on a knob or slider instead of one hundred and twenty eight, enough to make the staircase of discrete values vanish into a smooth curve. Bidirectional communication so devices could negotiate capabilities. Per-note controllers so you could bend the pitch of one note in a chord without affecting the others. Property exchange so devices could describe themselves. And all of it backwards compatible with every MIDI instrument ever made.

The announcement at NAMM two thousand twenty was made by representatives from Yamaha, ROLI, Microsoft, Google, and the MIDI Association. Roland, the company Kakehashi had founded, was there. Smith was alive and present. The specification was adopted unanimously. And the first requirement, the absolute non-negotiable demand that every participant agreed on from the very beginning, was the same principle that Smith and Kakehashi had built into the original specification four decades earlier. It must be free. It must be open. And it must not break what already exists.

The Esperanto of Instruments

There is something almost absurd about MIDI's reach today. Open any digital audio workstation, Ableton Live, Logic Pro, Pro Tools, FL Studio, Reaper, and the internal language it uses to represent musical performance is MIDI. Every virtual instrument, every software synthesizer, every drum plugin communicates through MIDI messages. When a producer draws notes on a piano roll, those notes are MIDI events. When a DJ triggers samples from a pad controller, those triggers are MIDI messages. When a film composer writes an orchestral score using sampled instruments, every note of every violin, every swell of every French horn, is a MIDI event driving a software sampler.

But MIDI escaped music long ago. Stage lighting can be triggered by MIDI signals, synchronizing light shows with musical performances in real time. Art installations use MIDI to connect sensors to sound generators, turning physical movement into musical expression. Drake Music, a UK charity, builds custom MIDI instruments for disabled musicians, transforming breath, head movement, and eye gaze into musical control signals. A breath controlled synthesizer. A head mounted digital baton. Hobbyists have wired everything from Arduino boards to homemade button grids to send MIDI notes, turning the protocol designed for synthesizers in nineteen eighty three into a universal language for translating human intention into machine response.

If MIDI disappeared tomorrow, every recording studio on earth would go silent. Every live electronic performance would stop. Every DAW would lose the ability to communicate with external instruments. Every MIDI controller, from a two hundred dollar pad to a ten thousand dollar keyboard, would become a paperweight. The protocol is so deeply embedded in the infrastructure of music creation that removing it would be like removing TCP/IP from the internet. You would not just lose a feature. You would lose the network.

And that network exists because two competitors chose cooperation over lock-in. Kakehashi was a man who had lost everything as a child, who had rebuilt his life from a watch repair shop, who had survived the disease that killed his parents, and who had built a company that changed how humanity makes music. He did not see competitors as enemies. He saw them as collaborators in a project bigger than any single company. And when his tribute was written, the word that kept appearing was not genius, or inventor, or visionary. The word was resilience.

The next time you open a music application on your computer, or press a key on a MIDI controller, or hear a beat that was programmed on a grid, you are hearing the echo of a knock on a hotel door in Anaheim. Two rivals. A five-pin cable. One hundred and twenty eight notes. And the decision that the music mattered more than the market share.

This has been a pilot episode of Ancient Tech. The series about old technologies that still run the world, told through the decisions that made them survive.