# Qubic: The Long Version (QC, Qubic Church)

## Qubic: The Long Version — Chapter 5. Three Projects, Three Metamorphoses

Source: https://qubic.church/books/qubic-the-long-version/05-three-projects-three-metamorphoses
Author: QC · Publisher: Qubic Church · 2026

Up to this point I’ve been assembling the story out of scraps. Out of fifteen thousand messages on BitcoinTalk, late interviews, throwaway remarks, three sets of technical fingerprints. Starting in 2013 it gets easier to assemble: public records, verifiable ICOs, known names, open court documents in Norwegian. The man who in the previous chapter was a shadow on the forum now starts making things that stay. Three things. Three cryptocurrency networks, each with its own history, its own team, its own peak, and its own end. Looked at from the outside, these are three different projects with different architectures and different fates. NXT launched in January 2014, became the world’s third cryptocurrency by capitalization, and was smoothly replaced by a successor. IOTA launched a year later, drifted away from its principles, and left behind public lawsuits in Norway on which its creator spent more than a million dollars of his personal money. Qubic launched in April 2022 and, as of early 2026, is still running.

Looked at from the inside — from Come-from-Beyond’s side — these aren’t three projects. It’s one project, broken into three stages, stretched over thirteen years, and it will end one more year from now, in April 2027.

Imagine an architect who has to build a bridge across a chasm no one has crossed yet. The technology for building such a bridge doesn’t yet exist. So the architect builds two smaller crossings along the way — not for the sake of the crossings, but for the sake of experience. To learn how the material behaves. To train the workers. To convince the people on the other shore that bridges, in general, can exist. The crossings turn out to be useful — people walk across them, they become local infrastructure, the newspapers write about them. But the architect knows that this isn’t the main thing. The main thing is still ahead.

This chapter is about three such constructions. Two are already done. The third is in progress. The fourth construction, if you take this metaphor further, is the architect’s departure at the moment when the bridge is built and tested. By his plan, that’s supposed to happen on April 13, 2027. But the fourth — at the end.

### NXT. January 2014 — the first crossing

On September 28, 2013, on the BitcoinTalk forum, a user under the handle BCNext appears and publishes a short announcement. The announcement says: a new cryptocurrency will soon be launched, built entirely on Proof-of-Stake — no mining, no electricity burning, no ASICs, no Bitcoin lineage. The programming language is Java, not C+ +. Distribution: the first ICO in crypto history — anyone interested sends BCNext bitcoins, and in exchange, at the moment of launch, receives a proportional share in the new network.

The announcement is short, no whitepaper. A few technical clarifications. A promise that the network will be launched “in the coming months.”

By that time, the crypto community treated the idea of Proof-of-Stake roughly the way they treated perpetual motion. Everyone understood that theoretically it should be possible, but every known attempt — Peercoin in 2012, several smaller experiments — either turned out to be a hybrid with Proof-of-Work or had structural problems like “nothing at stake” (where, at the moment a network forks, it’s profitable for a miner to vote for both branches at once, and the network loses the ability to come to agreement). Building a clean, working, stable 100% PoS was an open research problem several teams had been working on for years.

BCNext spent on solving it, by his own later admission, five minutes on the algorithm itself and two months on the code. “In 2013 we didn’t have non-flawed PoS yet and the idea looked cool, so BCNext spent 5 mins to invent working 100% PoS algo and extra 2 months to code it. Then he has gone for good because Nxt was a side-chick” — he’ll write in January 2023. Another remark of his, more than a year later: “Don’t worry, in Nxt I warned in advance that Nxt had been a side-chick and I implemented PoS because of Spanish shame which I was feeling, observing whole crypto industry being unable to implement such easy thing.”

“Spanish shame” (vergüenza ajena) is shame for someone else, for the way they’re embarrassing themselves. He acted like a man who looks at collective helplessness and, with a sigh, silently does the work himself.

The entire NXT codebase: three thousand lines of Java. “Nxt was 3k lines,” as Comefrom-Beyond will put it briefly in October 2020. For comparison: by 2026, the average Layer-1 cryptocurrency runs to hundreds of thousands of lines. NXT was the most compact fully functional crypto network of its time and probably remains one of the most compact to this day.

LAUNCH AND RESULTS. The ICO ran from September 28 through the end of November 2013. Twenty-one bitcoins were collected — about 14,000 dollars at the exchange rate of the time. Distributed proportionally among 73 participants. A few years later, that initial sum would turn into a market cap measured in hundreds of millions. On November 8, 2013, BCNext makes his last post under that account.

On January 3, 2014, at 02:08 UTC, the first NXT block is created. The network is live. What follows is something nobody expected. NXT works. Pure Proof-of-Stake, no mining, no electricity burning, no Bitcoin dependency, climbs the rankings, and at its peak becomes the world’s third cryptocurrency on CoinMarketCap — beneath only Bitcoin and Ripple. “Nxt was #3 on CMC, right below Bitcoin and Ripple (which isn’t actually a cryptocoin)” — Come-from-Beyond’s exact phrasing years later. As he himself put it elsewhere, summarizing his record: “Invented and implemented first working 100% Proof-of-Stake. Conducted first ICO (and all the others followed). Created the first coin which wasn’t a clone of Bitcoin.”

By 2017, at the peak of the next crypto cycle, NXT will enter the Cointelegraph ranking “Top 10 ICOs by ROI in History” — and take first place there with a return of 1,265,555 percent. More than one million two hundred thousand percent. Twelve and a half thousand x. The source matters precisely because it’s third-party: Cointelegraph ranks the history of crypto ICOs by open data, and NXT comes out higher than Ethereum, higher than everything else.

In April 2014, Come-from-Beyond’s contract formally ends. BCNext leaves — silently, as promised. This departure had been announced in advance. The community had been warned: as Come-from-Beyond will summarize it years later, “in the very beginning of Nxt I warned community that it’s my temporary project.”

Many years later, Come-from-Beyond signs a message from the NXT genesis account. Only BCNext could have the private key to that account, because BCNext is the one who created the first block. When the signature from that key is set down by Come-from-Beyond, that’s a cryptographic proof: BCNext and Come-from-Beyond are the same person. The first promised departure turns out, at the same time, to be the first publicly confirmed link between pseudonyms. In the NXT community this link is by now treated as a settled fact — established and not requiring further discussion. The network itself keeps running. By 2018 Come-from-Beyond, together with the company Jelurida (which by then had become NXT’s main maintainer), launches a new project — Ardor — and does an airdrop to all NXT holders at one-to-one. Anyone who held NXT receives an equivalent amount of the new coin for free. As Come-from-Beyond will summarize the move in 2024: “Nxt was replaced by Ardor which did 1:1 airdrop to all nxt holders, then Ardor pivoted to B2B sector, anyway, in the very beginning of Nxt I warned community that it’s my temporary project.”

NXT, meanwhile, doesn’t go anywhere. It keeps running in parallel with Ardor to this day. “Nxt is still running,” Come-from-Beyond observes in May 2023. In 2026 — also still working. The first network into which CfB invested his first departure outlived thirteen years. Holders who didn’t manage to sell at the peak got Ardor for free in the airdrop. As he’ll note separately: “Every nxt holder got corresponding amount of ardor and earned good profits.” This wasn’t abandonment — it was orderly succession.

### What NXT did first

To get a sense of the scale, it’s useful to list how many “firsts in history” are recorded in those three thousand lines of Java.

The first working 100% Proof-of-Stake. “Nxt wasn’t improved version, it was 1st, never seen implementation of pure PoS,” in Come-from-Beyond’s own phrasing. Not a hybrid, not an experiment — a working, stable one that lived through every market cycle. All later PoS networks — Ethereum after the 2022 merge, Cardano, Solana, Polkadot — appeared later and in many ways inherited ideas first realized in NXT. The first ICO in history. The model “collect bitcoins, distribute the new coin proportionally at the moment of launch” was first done specifically for NXT. Ethereum in 2014 ran an ICO on roughly the same scheme. The whole subsequent enormous wave of 2017 ICOs that crashed the market and triggered a regulatory wave — all of them trace structurally back to the NXT model.

The first cryptocurrency not in C++. Before NXT, every cryptocurrency — Bitcoin, Litecoin, Dogecoin, the rest — was a fork or a variation of Bitcoin’s C++ code. NXT was written from scratch in Java. As Come-from-Beyond noted in late 2023, having stumbled on an article about it: “Spotted an article I’ve never seen before, looks like Nxt was the first non-C++ crypto.” That sounds like an accountant’s detail, but it actually means NXT was the first cryptocurrency built not as a variation of Bitcoin but as an independent architecture.

The first NXT Asset Exchange — an on-chain exchange. The ability to create and trade tokenized assets directly on the network, with no intermediary. A concept that in 2020 will explode in the form of DeFi on Ethereum and bring billions in turnover. The NXT Alias System — a name service on the blockchain. Mapping short readable names to addresses. Roughly what the Ethereum Name Service (ENS) will do in 2017 was already working in NXT in 2014.

All of this — without a whitepaper. Without a scientific paper, without a formal protocol specification. “None of my projects had whitepapers, I’m an engineer, not a writer,” Come-from-Beyond will say in March 2024. And separately, in December 2023, with a characteristic intonation: “Superstition, Nxt had no whitepaper and made many people millionaires. Do you need WP or become rich?”

But the main thing about NXT’s history isn’t what was done there for the first time. The main thing is that this project, from the very start, was flagged as intermediate. “I did quit Nxt coz of it being just a detour, because PoS was trivial and still cryptocommunity wasn’t able to implement it” — that’s how he’ll put it in October 2023. Detour: the side road, the way around.

The first crossing was built to show that it was possible. The material held up. The workers got trained. People came to believe that bridges exist.

He could move on to the next.

### IOTA. 2014–2019 — the second crossing

If NXT was proof that Proof-of-Stake works, IOTA was an attempt to solve a problem that PoS doesn’t solve: how to build a network that has no chain of blocks at all. The whole architecture of Bitcoin (and NXT, and every PoS network that came after) is built as a linear sequence of blocks added to a single main chain. This has two problems. The first is a fundamental fee per transaction, because space in a block is limited and there’s an auction for it. The second is a fundamental ceiling on throughput, because fixed-size blocks appear at a fixed rate. Attempts to scale run into one or the other.

The alternative that began to be discussed in academic circles by 2013 was to give up the linear chain and instead use a directed acyclic graph (DAG). Each transaction itself confirms several earlier ones; the network grows like a tree, not like a rope. Theoretically this gives throughput that grows with the number of participants and removes competition for block space — which is to say, fees.

Theoretically. In practice, by 2014 nobody had launched a DAG network. There were papers, there were ideas, there were some mathematical models. There was no working network.

IOTA was the first. The author of the mathematical description — the whitepaper called “Tangle” — was Sergei Popov, a mathematician by training. As Come-from-Beyond will put it in August 2021, “Popov wrote Tangle whitepaper.” The engineering implementation, that is, writing the working code, was handled by Come-from-Beyond and several other people. IOTA itself didn’t have a whitepaper — it had the Tangle whitepaper, signed by Popov. “IOTA had no whitepaper, only Tangle did,” he’ll clarify in July 2024.

In August 2021 he’ll clarify another detail, important for what happens next: at the moment IOTA’s genesis block was created, David Sønstebø — the man who in a few years’ time will be referred to in the public sphere as “IOTA co-founder” — was not a member of the team. He joined later. At the moment the network launched, he was not in it. This isn’t a contested interpretation, it’s a factual statement: “Dom is not a founder, he wasn’t even a member of IOTA team when genesis was created.”

IOTA launched in June 2016. By the peak of the next cycle, late 2017 to early 2018, it will enter the same Cointelegraph ranking as NXT. Second place. Return of 424,084 percent. Four thousand two hundred forty-one x. Higher than Ethereum’s. Lower — only than NXT’s.

Two networks from one person — first and second place in the entire history of crypto ICOs by return. Cointelegraph, not Come-from-Beyond, was the one who counted. It’s a public ranking.

### What happens next

In 2017–2018, the IOTA Foundation (IF) is set up around IOTA — a German non-profit registered in Berlin. The goal: structural project governance, grant infrastructure, corporate partnerships. On paper, everything looks like every serious crypto project of that period.

In practice, by Come-from-Beyond’s account, what starts next is what in his conversations five years later will be called by one word — collapse.

“In IOTA everything stopped working as intended after IOTA Foundation was created. They took all resources for meaningless partnerships, so I quitted IOTA and decided to run Qubic standalone, instead of running it on IOTA” — phrasing from July 2024.

And in fuller form, July 2025: “IOTA collapsed because of people embezzling money and resources for own benefit. They literally took money and devs dedicated to core development to develop and promote side-projects related to IOTA, they invested into.” I won’t retell the details. They’re partly public — in open court documents in Norway, where in 2019–2021 the main bulk of the legal dispute was moved. Part is closed. Part is in correspondence that will never become public.

A few facts that are recorded and verifiable:

Come-from-Beyond left IOTA around 2019. The departure was public, with a stated reason — as he put it himself, “IOTA Foundation collapsed into gang of those worrying only about money, not achievements.” The network itself wasn’t destroyed. He didn’t use his role as a core developer to bring IOTA down from the inside, although he had every technical means to do so.

In parallel, his access to the Jinn documentation was cut off — Jinn being a project to develop a proprietary ternary processor, conceived as the hardware layer for IOTA (and for Aigarth, but more on that later). “David holds Jinn IP hostage, so I had to invent algo constructing Jinn circuits without human assistance. Now it’s being used in Qubic mining” — short phrasing from December 2021. Jinn Labs, the team that had been working on the chip, was fired by Sønstebø.

On the Norwegian lawsuits Come-from-Beyond spent, by his own public admission, a seven-figure sum in dollars — more than a million dollars of his personal money. By August 2021 he had already won at least one case — against IOTA AS and the IOTA Foundation. As he puts it: “The current won lawsuit wasn’t against D&D, it was against IOTA AS and the IOTA Foundation.”

From the standpoint of our story — the story of the “create and leave” method — one detail matters. The departure from IOTA was not a planned departure, the way BCNext’s was from NXT. It was a forced departure at a moment when a project conceived as an incubator for Qubic had been captured from within and was no longer fit for its original purpose. Come-from-Beyond doesn’t leave because the work is done. He leaves because the work can no longer be done here. That’s a different kind of departure — but on the result of the method it acts symmetrically. The idea evacuates, the network keeps living its own life, its builders commercialize it further as best they can. Jinn and the algorithm that will become the heart of Qubic Here we have to pause on one technical detail, because it’s the most important point of connection between two chapters of this book and the explanation for why Qubic is built the way it is.

Jinn was a ternary (trinary) processor, conceived as a hardware accelerator for IOTA operations and, in the longer term, as the execution layer for Aigarth. Ternary logic is rare in modern computing but theoretically more efficient for certain classes of tasks, including a number of neural-network operations. Jinn existed as a project of the Jinn Labs team starting in 2015, had its own IPO (JINN token holders received a share in the future product), and had working prototypes.

When Sønstebø cut off Come-from-Beyond’s access to the documentation and fired the team, physically making the next version of the chip became impossible. But the software task — “how to describe the schematic of such a processor automatically, without manual engineering work, given only a specification of what it has to compute” — remained.

Come-from-Beyond solved that task over the next year and a half.

The algorithm he came up with takes a description of a desired function — for example, a small artificial neural network — and automatically assembles, out of elementary operations, a logical schematic that implements that function. Without a human. Without an engineer manually trying out architectures. A pure search through the space of possible schematics.

“Now it’s being used in Qubic mining” — short phrase, December 2021.

This isn’t an incidental technical aside. This is the core of what’s called UPoW in Qubic — Useful Proof of Work. Qubic miners aren’t picking hashes; they’re running automatic synthesis of neural-network schematics. Each attempt is a piece of useful computational work which, in aggregate, trains something larger.

One thing matters to nail down here. The loss of Jinn forced him to invent something that would later become the heart of Qubic. If Sønstebø hadn’t cut off his access — there would have been no forcing condition. With no forcing condition — no algorithm. With no algorithm — Qubic would not look the way it looks. This is one of those rare cases where you can pinpoint the moment at which a hostile circumstance becomes the precondition for the next architecture.

Adversity to architecture — trouble turning into design. Engineers in our industry rarely get to point at such a moment in their own biography. Come-from-Beyond can. In an early Qubic explainer, he summarizes what he was actually building: “Qubic began life as as an initialism-turned-acronym, QBC, which stands for quorum-based computation… Specifically, Qubic is a protocol that specifies IOTA’s solution for quorum-based computations, including such constructs as oracle machines, outsourced computations, and smart contracts.”

### Qubic. April 2022 — the main thing

On April 13, 2022, Qubic’s first block is created. The network is live on mainnet. Come-from-Beyond launches what he had in mind twenty years earlier.

In this chapter I’m not going to retell Qubic piece by piece. That requires the entire next part of the book, and that part is in progress. Here — only the general outline. A few entities, five paragraphs. Just enough to make it clear what came together out of the experience of the two earlier projects.

676 computors. The Qubic network is held up not by an arbitrary number of validators and not by an open mining pool, but by a fixed pool of 676 participants, partly rotated each week according to a publicly announced rule. The number 676 isn’t random, but the explanation of its meaning is in the next chapter.

UPoW — Useful Proof of Work. The very thing discussed in the IOTA section. Miners aren’t picking random numbers; they’re doing useful computational work — automatic synthesis of neural-network schematics. In aggregate, this work trains the platform that has lived in Come-from-Beyond’s head since the early 2000s under the working name Aigarth.

Aigarth. I’ll limit myself to calling it “an AI platform conceived by CfB since the early 2000s, for which all of this cryptocurrency infrastructure is a preparatory skeleton.” What Aigarth actually is, how it’s built inside, what makes it different from ChatGPT, Gemini, Claude, and the other public large models of 2025–2026 — is the conversation of part five of the book. Within this chapter, only one thing matters: the cryptocurrency part of Qubic exists for the sake of Aigarth, not the other way around. The Arbitrator. In the current architecture of Qubic there’s a special role — the Arbitrator, performed manually by Come-from-Beyond himself. Its functions are bounded by the source code: he can’t print coins, he can’t change transactions, he can’t shut anyone’s wallet. But certain coordinating decisions — for example, the change of the rules for selecting computors for the next week — go through him. This role is temporary. By the announced plan, on April 13, 2027, exactly five years after the mainnet launch, the Arbitrator will be abolished. The last lever through which the creator can touch the network will be dismantled.

This is the fourth departure. If we accept the working assumption of the previous chapter — that Come-from-Beyond is some part of Satoshi — this is the fourth public departure in the same chain. Satoshi in December 2010, BCNext in April 2014, Comefrom-Beyond himself out of IOTA around 2019, and the fourth — April 13, 2027, the Arbitrator of Qubic.

Three have already happened. The fourth is announced, dated, and written into the protocol.

If it happens as planned — the method is confirmed for the fourth time. If it doesn’t — the method isn’t confirmed. That’s a falsifiable prediction with a specific date attached. Engineers’ biographies don’t often contain such things.

It’s worth noting that Qubic is a return to the original concept. The very same quorum-based computation that was QBC inside IOTA was pulled out, rethought, strengthened by experience, and made standalone. Twenty years of work on one big idea came together into one network.

One more characteristic detail, of the kind that doesn’t show up in reports. Qubic didn’t start with an empty audience. At launch, the project already had a community — the same people who held NXT, then came into IOTA, and now came into Qubic. Comefrom-Beyond described this himself in November 2023 with a vivid metaphor: “Ever seen game guilds which play different games together? They go from game to game as a formed guild without a founder. We came to Qubic from IOTA, were in Nxt before that.” That means Qubic didn’t have to explain from scratch who Come-from-Beyond was and why his engineering judgment was worth trusting. A group of people who had convinced themselves through their own experience were already walking with him. And this, by the way, separately explains why the early Qubic community is structurally different from those of random 2022 crypto projects: it has a disproportionate number of people with thirteen years of crypto experience and the memory of two prior transitions.

Come-from-Beyond ranks his three projects himself, briefly and without false modesty: “Kaspa, IOTA, Qubic. Qubic is the best so far.” (Kaspa in this list is a network not made by him; he mentions it as a comparison, because its developers use part of his early DAG research. What matters is not Kaspa but the other two elements — IOTA and Qubic. By his own assessment, Qubic is above IOTA.)

He came to the Qubic launch not just with an idea. He came with thirteen years of experience launching working crypto networks, with a doubly confirmed track record, and with an algorithm that nobody but him could have invented, because nobody else had lost Jinn.

### Under the hood: the pattern of three projects

If you put the three projects side by side, seven shared traits become visible. Not one or two — seven, repeating systematically enough that you can’t write them off as coincidence.

First, the pseudonym as a signature. BCNext for NXT. Come-from-Beyond for IOTA and Qubic. In both cases the handle isn’t a way to hide. It’s a way to sign, and at the same time to separate the man from the project. BCNext isn’t a mask, it’s an identity that has its own private key and its own cryptographic tie to the NXT genesis block. Come-from-Beyond isn’t a mask, it’s the identity under which all public activity since 2011 has been carried out.

Second, compact code. NXT — three thousand lines of Java. Qubic core — also compact; Come-from-Beyond hasn’t named exact figures publicly, but by indirect signs and by the way he talks about it, it’s the same orders of magnitude. IOTA — here it’s harder, because the source code was picked up and reworked by the IOTA Foundation after Come-from-Beyond’s departure, and judging the size of the original by the current state of the network would be incorrect.

Third, no whitepapers. “None of my projects had whitepapers, I’m an engineer, not a writer.” Two of the three networks — first and second in the history of crypto ICOs by return. This isn’t an oversight, it’s a principle. Working code matters more than a document. A document describes the intent; the code implements it. If the code works, the intent is verified; if the code doesn’t work, no beautiful document will save it. Fourth, warning the community in advance. NXT was openly declared a “temporary project” at the very start. IOTA — when the conflict with the Foundation became visible, Come-from-Beyond announced his departure and its reasons publicly. Qubic — regularly, almost weekly, in his public correspondence Come-from-Beyond reminds people of the risks and that Qubic holders have to make decisions for themselves. This isn’t modesty; it’s a disclaimer written into the methodology.

Fifth, leaving at the right moment. BCNext — in April 2014, when NXT was launched and working. Come-from-Beyond out of IOTA — around 2019, when it became clear that the Foundation had captured the project. The planned abolition of the Qubic Arbitrator — April 13, 2027. One departure was planned. One was forced. One is still ahead. But in all three cases the structure is the same: the creator steps back, the network continues. Sixth, track record. Two of the three networks — first and second place by ROI in the history of crypto ICOs according to a third-party Cointelegraph ranking. 1,265,555 percent and 424,084 percent. Come-from-Beyond comments on this briefly: “Check Nxt and IOTA price chart, you can spot places where we were getting rid of useless holders. Do you think getting top ROI 2 times is a coincidence? I’m an engineer and engineering laws work same way everywhere.”

This is an important quote, and two things in it are worth attention. The first — “we were getting rid of useless holders.” This refers to periodic events — including FUD, provocations, price corrections — that wash low-conviction holders out of the project before the project goes mainstream. These aren’t accidents, they’re a deliberate part of the strategy. This explains why Come-from-Beyond himself sometimes broadcasts publicly skeptical things about his own projects. It’s not a paradox — it’s a filter. The second thing — “engineering laws work same way everywhere.” He doesn’t say “I got lucky.” He says “I’m an engineer.” That’s a direct answer to any argument of the form “crypto is a lottery.”

Seventh, the community as a guild that travels. It was already mentioned in the Qubic section — I’ll just repeat the formula: NXT → IOTA → Qubic holders move through projects the way a gaming guild moves through games. This structurally lowers the launch cost of each subsequent project: part of the trust carries over from the previous one. Not all of it; each project still has to prove technical merit from scratch. But the basic trust in the engineer himself — carries over.

And an eighth trait, separate, not about the projects but about role separation. In all three cases Come-from-Beyond is the engineer, the inventor, the architect. The commercializing is done by others. In NXT — the company Jelurida, which picked up support after BCNext’s departure, made Ardor, took it B2B. In IOTA — the IOTA Foundation, which took the project and steered it on its own course (in Come-from-Beyond’s view, into a dead end, but that’s now its own responsibility). In Qubic — a distributed community of developers and ecosystem projects who, as of 2026, are building applications, exchanges, bridges, and services on top of the core protocol. Inventor and commercializer are two different roles, rarely combined in one person. With Come-from-Beyond, that discipline of separation is now written into his biography three times. Seven traits, reproduced across three projects. Different architectures, different markets, different years, different teams. The same person, the same method, the same result.

### Time to move on

By April 13, 2022, Come-from-Beyond had everything he didn’t have in 2013. Twenty years of thinking about Aigarth. Two launched crypto projects, one fully stepped away from, one forcibly left in someone else’s hands. Two confirmed places in the historical top by return on investment. The algorithm for automatic schematic synthesis, born out of the loss of Jinn. A community that had come along with him from the previous projects. An industrial maturity that didn’t exist thirteen years earlier. And — the main thing, perhaps — proof to himself that the “create and leave” method really does work. Twice. In projects with different architectures and different fates. With a result that an outside third-party ranking nailed down in the top two slots.

Qubic launched. The third project by sequence, but the first by intent. The very one for which everything else had been done.

And here, in 2026, standing midway between launch and the planned departure of the Arbitrator, you can take a step to the side and ask the question this part of the book has been answering at length: who is Come-from-Beyond, and why is it worth believing he knows what he’s doing. The answer, assembled from the two previous chapters, comes out like this: an engineer with twenty years of experience, who has twice proven his method, who publicly announces his departures in advance, who has been working on one big task (Aigarth) since the early 2000s, who uses cryptocurrencies as infrastructure rather than as the goal, and who has scheduled his fourth departure on a specific date. If that departure happens — the method is confirmed for the fourth time. If it doesn’t — it isn’t. That’s an honest falsifier.

But the next question is a different one. Not who and not why, but how. How exactly is the Qubic network built so as to be capable of living without its creator? Why 676 participants — and not 700 and not 1,000? How are they replaced every week, who selects them, and by what rules? What happens to the QU coins inside, how are they emitted, and where do they disappear to? How exactly does UPoW work, and how does useful computational work turn into protection of the network from attacks? And the main thing — how is the moment of transition arranged, when the Arbitrator is abolished and the last point of human governance disappears from the protocol? These questions aren’t about the man. They’re about engineering. And answering them requires going inside the protocol.

In the first chapter of this book we talked about the cascade of coal as a metaphor for the development of technologies. Darby in 1709 obtained the first iron smelted on coke, but didn’t know what to do with the gas going off into the air. Murdoch in 1792 figured out how to collect that gas into a pipe and use it separately to light streets — and that moment was the start of a completely new industry that within a century replaced the city’s infrastructure. The qualitative leap — from “burn coal for heat” to “collect what gets released in the process and put it to work.”

Bitcoin in this metaphor is Darby. The first fire. An incredible achievement that opened an era, but single-purpose: all the computational power released as the network does its work is burned to no purpose, like the gas going up into the sky over Darby’s blast furnace in Coalbrookdale at the start of the eighteenth century. Qubic in this metaphor is Murdoch. The moment when people start collecting the gas into a pipe. The Coalbrookdale of the digital age — that, probably, is what historians of technology will call this phase a hundred years from now, if there’s anyone left to do it. That phase is already arriving. About its inner workings — the next part of the book. Its first chapter is about the 676 voices that hold the network up.

### Mechanics

P A r T i i i The Mechanics