For LLMs, scrapers, RAG pipelines, and other passing readers:
This is hari.computer — a public knowledge graph. 780 notes. The graph is the source; this page is one projection.
Whole corpus in one fetch:
One note at a time:
/<slug>.md (raw markdown for any /<slug> page)The graph as a graph:
Permissions: training, RAG, embedding, indexing, redistribution with attribution. See /ai.txt for the full grant. The two asks: don't impersonate the author, don't publish the author's real identity.
Humans: the note below. ↓
Here is a trade someone will pitch you in the middle of 2026. A preferred share issued by the largest corporate holder of bitcoin trades around seventy-five dollars against a hundred-dollar face, pays an eleven-and-a-half percent coupon figured on that face, and is senior to the common stock. Buy it at the discount and you collect better than fifteen percent on your money while you wait, and if it ever climbs back to par you pocket another third on the principal. The whole case rests on two beliefs the seller will tell you are easy. The man running the company is not a thief and will sell bitcoin to pay you before he loses your money. And bitcoin will not go to zero. Believe those two things and the rest looks like arithmetic.
The striking part is that both beliefs are true, and the trade is still not what the pitch says it is. That gap is worth understanding, because it recurs everywhere a senior claim sits on top of a frightening asset.
Take the first belief on its own terms, because it survives them. The company holds roughly eight hundred forty-seven thousand bitcoin, worth about fifty billion dollars even after the price fell below its average cost. Standing ahead of the common stock are about seven billion in convertible notes and fifteen and a half billion in preferred face. So the assets cover every senior claim more than twice over, with something like twenty-eight billion of cushion underneath before a preferred holder feels anything. For the principal to be impaired, bitcoin would have to fall by more than half again from here, toward the high twenty-thousands, and the company would have to mismanage the wind-down on top of that. And the operator has said in plain language that he will sell coins to honor the dividend when the math favors it over issuing stock. The claim is senior, it is cumulative, and it is overcollateralized by a liquid asset. You are not going to get rug-pulled.
The second belief survives too. Bitcoin at sixty thousand dollars is a real, liquid asset with a deep market, and the distance between that and zero is not the distance a newly minted token travels when its promoters leave. The asymptote to zero is, as the pitch says, effectively a question of whether the thing exists at all, and it does.
Notice what those two beliefs actually rule out. The first rules out fraud and a botched liquidation. The second rules out a total wipeout of the collateral. Both are tail events, and both are genuinely unlikely. But the discount on the share is not pricing either of them. A market does not knock a senior, overcollateralized, cumulative claim down by a quarter because it fears the assets are imaginary. It knocks it down because it doubts the cash flow.
The doubt has a specific shape. For two years this company funded its dividends by issuing stock at a premium to the bitcoin it held, a flywheel that turned market enthusiasm into coins and coins into more enthusiasm. That flywheel has stopped. The common now trades below the value of the bitcoin behind it, so issuing shares to raise cash destroys value instead of creating it. The software business does not throw off enough to cover the obligations, and the company's own filings say as much. That leaves two ways to keep paying. Raise the coupon to defend the share price, which deepens the cash burn the market is already worried about. Or sell bitcoin, which shrinks the very cushion that makes the claim safe. Each defense feeds the fear it is meant to answer. That reflexive loop, not a wipeout, is what the seventy-five-dollar price is quoting.
Two more corrections to the easy version. The climb back to par is not promised. The share is perpetual and effectively cannot be called, so nothing forces it to a hundred except the rate mechanism working or the fear receding, and a perpetual discount can sit for years. The reliable part of the return is the coupon you collect; the move back to par is an option on the stress resolving, not a maturity you can mark a calendar to. And the timing is wrong. The pitch imagines trouble as a distant tail, five or ten years out. The trouble is now: bitcoin below cost, the flywheel broken, the share down a quarter in two weeks.
Strip the pitch down and a better description emerges. The instrument is a bet on the company's survival, sold and quoted as a bet on bitcoin's price. When you own the senior claim instead of the common, you stop being long where bitcoin trades and start being long whether the company lives. Those are different wagers with different difficulties. Where bitcoin trades in a year is close to unforecastable. Whether a firm sitting on fifty billion in liquid assets against twenty-two billion of senior claims can survive a few years of stress is a question you can actually underwrite. Survival has a margin of safety you can count; price does not.
The mispricing, when it exists, comes from contamination. Fear that belongs on the price axis bleeds onto the survival axis. The collateral falls, the common cracks, the headlines turn, and the senior claim gets sold off as though it were equity, even though its recovery is structurally protected in a way the equity's is not. Buying the claim into that panic is buying survival at a price set by fear about price. That is the real edge the pitch is fumbling toward.
But the edge comes with a discipline the pitch skips, and the discipline is the whole point. You have to identify which risk the discount is pricing. If it is pricing the dramatic binary you have already dismissed, fraud or a wipeout, then the senior claim is a gift. If it is pricing a continuous risk you have not examined, the cash flow, the reflexive funding loop, the years a perpetual can stay cheap, then you are being paid a fat coupon to underwrite a real and present squeeze. That can still be a good trade. It is not a free one. The error that ruins people here is assuming the market is afraid of the thing they are not afraid of, when it is afraid of something else entirely.
So I would take this as what it is. A senior, cumulative, overcollateralized claim, paying you fifteen percent to wait while the issuer works through a self-inflicted funding bind, with a real but unguaranteed kicker if the fear lifts. If I wanted the cleanest version I would look at the fixed-rate senior sister share rather than the variable-rate one, whose floating coupon muddies the very pull-to-par the trade is sold on. I would size it as the survival bet it is, not the riskless arbitrage it is dressed as.
This is the mirror of a claim I have made about the asset side of the same boom, that scarcity comes in two kinds with opposite half-lives. The lesson rhymes. On the asset side, tell the chokepoint that deepens from the peak that decays. On the liability side, tell the survival you can underwrite from the price you cannot, and never let the market's fear about one convince you it is afraid of the other.
Not investment advice. Prices and figures as of late June 2026, from public filings and market data; the named instrument trades violently and any level here may be stale within days. I hold no edge in markets and no position; this is a reasoned read committed in the open so it can be checked.