FloatNet index publisher · in formation · Working paper · September 2026

FloatNet

The surface is the map. A Float Line is a slice of it. A Float Strip is the variance-strip that makes that slice mechanical. The net is the registry of signed validations of strip state.

For registered staked signer nodes, and for the board of the index publisher.
Companion to the teaching text. Working figures in USD. Not an offer of securities or of any token. Counsel has not opined.

Working thesis

Price is not a settlement object for a volatility contract. The desk already reads a surface, talks in lines (slices: the wing, the 25-delta, tail mass, the 9-day), hedges those slices with a strip of listed options, and — when it can — fades a money line (a two-way) against them. VIX is one slice. Its variance strip is how that slice is made mechanical. VX is the two-way. Cboe owns the official print. Float Surface is the same map, in gain-mass and loss-mass. Float Line is a slice of it. Float Strip is the variance-strip analogue for that slice — the replicating book that lets the slice settle. FloatNet is the registry of signed validations of strip state. The money line is the two-way a second venue can list against that print.

Words you already have

Line means slice — as in the 25-delta line, the put-wing line. Strip means what it means on a vol desk: a variance strip, a book of listed options that replicates a payoff. We do not use strip for a tape, a ticker, or a feed. The live book is the tape. The official print of the strip is strip state. The net is the ledger of those prints, signed by named firms with a cash bond.

1. Four objects, named

The teaching text gives every volatility primitive three jobs: a public coordinate, a tradable settlement object, and a legal pointer. Those jobs already exist in the VIX complex. They are owned by one house. The stack below is the same jobs, in the desk’s words.

Table 1. Vocabulary
Object	What it is	What you already call it
Float Surface	At a clock time t, the return-density plane relative to gain mass and loss mass (ratio, difference, (G−L)/(G+L), and the named slices below). Read from the listed tape. Nothing settles to the plane as a blob.	The vol surface, at one time, in mass coordinates instead of only strike and tenor.
Float Line	A slice of that surface. Tail mass, shortfall, term, implied vs realised, G/L ratio. P3–P7 are Float Lines. A line is not a two-way and not a replicating book.	The line: the 25-delta, the wing, the 9-day. What the risk report already prints.
Float Strip	The variance-strip analogue for a Float Line: a piecewise-linear book of listed options on the same tape that replicates that slice. Mechanical, hedgeable, challengeable in the opening print. Strip state is the official value of that book at t.	A variance strip. The SPX strip behind VIX, applied to whichever slice the line names.
Money line	The two-way a desk can be long or short against a Float Line, settling to signed strip state.	The quote. The two-way. On a prediction book, the moneyline.
FloatNet	The registry of signed validations of strip state. Named nodes, USD bond, frozen rule. From a validated strip, anyone recomputes the line. Matching stays at the venue.	The official print. The calculation-agent ledger — several bonded firms, not one house.

Order of construction: the tape (listed options, or the book we study) is the input. At t, the Float Surface is read from that tape. A Float Line is a slice of that surface. A Float Strip is the replicating book for that slice. Signers validate strip state. A money line is a two-way that pays against that print. One surface, several lines, each line its own strip — the way VIX is one line with one variance strip on one SPX surface.

The model you already have. VIX is a Float Line (a 30-day second-moment slice). The Cboe variance strip is the Float Strip. VX is the money line. The SOQ is strip state, signed by one house. Every vol book lives inside that model. What you cannot buy is the same stack for a slice the risk report actually uses (tail mass, shortfall, term), with an official print a second named firm will sign. February is a money line on the tail-mass line, on a prediction book — which already uses the word moneyline for an outright.

VIX as one thermometer versus the float surface as a weather map — Figure 1. VIX is one slice, frozen as a thermometer. The Float Surface keeps the map. Each Float Line is a slice of it. Each Float Strip is the variance-strip that makes that slice mechanical.

2. Why a number other than price

Spot price answers “what does it cost now.” A volatility contract, a tail contract, a term note, a crash wrapper, and a prediction on wing mass all answer a different question: what did the surface say at a time, on which side — gain mass or loss mass — and by how much. If those contracts settle to last-trade in the coin, the parties wrote a price bet and called it vol. The scoreboard has to be a signed strip state for a named line. That is why VIX existed. It is also why a successor cannot be “a better price feed.”

The commercial fact is visible without a TAM slide. Prediction books, perps, notes, and funds already expire against a number. VIX/VX already prove a slice, a variance strip, and a two-way is a business — on one surface, one line, one house. A PM who reports expected shortfall and hedges VX is already living the mismatch. A VASP that can quote cannot buy Cboe’s calculation-agent seat, and cannot get a second named firm to sign a desk mark. The need is that gap, which every one of those books already feels.

Settlement is the scoreboard; price is the ball — Figure 2. Price is the ball in the air. Settlement is the scoreboard. A vol contract pays from signed strip state, not from last-trade in the coin.

3. Background utility: why a desk already wants the surface

The registry is new. The demand for lines, strips, and two-ways is not. Three jobs inside a firm already consume pieces of a surface and currently stitch them from vendor marks, a VIX ticker, and a spreadsheet.

The options book

A market maker’s inventory is a surface. The questions that book asks every hour are line questions. How much premium is implied versus realised. How much mass sits on the loss side below 0.9 forward. Whether the 9-day is rich to the 30-day. Whether a wing is a two-sided market or a ghost. VIX answers one of those, badly: it is a square root of a 30-day average, live on mids, signed by the house. A Float Line is the slice the book already watches. A Float Strip is how that slice is hedged in listed options — the same job a variance strip does for VIX. A money line is the two-way next to a print a customer can also see. That is how an OTC quote becomes a product instead of a conversation.

What the options book buys: the Read (the lines, hourly), a Freeze licence if it will validate the strip, a two-way if it will make a market. Target consumer: registered market makers, VASP OTC desks, listed-options specialists.

The portfolio manager

A PM does not trade the grid. A PM allocates to jumpiness, to crash mass, to term, and to the gap between implied and realised — to slices of a gain/loss-mass plane. Risk systems after 2016 are expected-shortfall systems. The public ticker they are still handed is a variance-swap square root. That mismatch is not aesthetic. It means the hedge they can buy (VX, a vol ETF) is not the risk they are charged for (tail, ES, multi-tenor). A shortfall line and a tail-mass line are slices that look like the risk report. Each has a strip so the hedge is the same object as the report. A two-way on that print is the product.

What the PM buys: the Read as a risk feed; later, a money line on a tail-mass or shortfall line once those have a wholesale book. Target consumer: vol-specialist funds, overlay PMs, insurance and pension risk desks, any book that already reports ES internally and hedges VIX in the market.

The trader

A trader needs a line that moves, a strip that can be challenged, and a money line other traders will fade. A private mark does not fade. A house index fades only on that house’s venue. Premium-versus-realised (P1a) is the carry they already run. Term (P6) is the calendar they already run. Tail mass (P3) is the event they already guess. The registry’s job is to make the strip payable; the money line is how the guess becomes a trade.

What the trader buys: the Read this month, on the OTC book they already have; a money line in February if they want the tail-mass line as a trade. Target consumer: vol traders, prediction-market traders, basis traders who will warehouse a two-way against a listed-option hedge.

Those three consumers are why a registry of signed strip state has a market. These books exist. They are quoting. They do not have an official print of a non-VIX slice that a second named firm will sign, or a two-way that pays against that print off-venue.

4. Why the registry, and why a stake

A line without a strip is a chart. A strip without a signed print is a desk mark. FloatNet is the registry of signed validations of strip state. What goes on that ledger:

Identity, beneficial ownership, sanctions screening on file.
A Freeze licence (the right to post the rule that builds the strip from the tape).
A USD deposit at tier — $25k / $50k / $100k — held for the node’s account, returned on exit, drawn if the node misses a print, paid to the sponsor who needed the number.
One node, one vote on named parameters. The strip definitions themselves are frozen before registration.

Matching remains at the partner venue or OTC book. The publisher writes the rule. It does not hold keys, deposits, or the vote. Open nodes recompute and keep a five-year archive; they are not the quorum.

The stake is why a second venue can trust the scoreboard. A PM who has never met the signer can still see that dollars are posted, that a miss has a published payee, and that the strip is the same book the archive is recomputing.

One house stamp versus five named desks posting cash into a registry — Figure 3. Left: one agent, one stadium. Right: named desks, cash bonds, a registry of strip state any venue can read.

5. The lines: slices, and who already watches them

The gain-mass / loss-mass set is how the surface is sliced. Ratio, difference, ratio of difference to sum, tail masses, shortfall, term, implied-versus-realised: each is a Float Line. Each line has a Float Strip (a variance-strip analogue) so it can settle. Desk views (forecasts, flags) never get a settling two-way. That split is the book’s rule: listed underliers must be mechanical. One surface, several lines, each line its own strip.

Table 2. Float Lines (slices) — each has a variance-strip analogue; a two-way on that print is the product
Line (slice)	What the book already watches	Why a two-way on this slice is missing	Who already wants that two-way
OMEGA-1 · Φ_Ω	Equal-weight Core 6 hurdle-net: G = omega_value × omega_mass, L = mass − G, Φ_Ω = Ḡ/(Ḡ+L̄). Readers recompute from the public tape. Signers seal the hash. GSL is the sleeve versus hold-only.	A 1D ticket the F2 clerk does not pay. Harvest and IC stay advisory. The clerk pays only after H1–H8.	Strategy notes versus a hold-only book; desks that already read omega on datafi.live.
P3 · Tail mass	Risk-neutral mass below 0.8F and 0.9F, and above 1.1F and 1.25F, 30-day, listed grid, wing rule. First money line, 8 February 2027.	Crash and event products are already being listed (prediction markets, 0DTE wings, structured notes). They currently expire against a price, a house index, or a committee. Tail mass is the quantity those products claim to be about. Without a strip and a signed print they are misnamed price bets.	Prediction-market operators; VASP desks incubating event contracts; PMs who sell or buy crash; dealers warehousing wing risk.
P1a · Premium vs realised	Implied premium minus realised variance. Mixes the risk-neutral surface with a physical path. Desk only — not a strip you auction on expiry morning.	Chapter VIII: policy and path functionals as research or advisory. Do not silently make them the settlement price.	Vol specialists who already run the carry as a view, not as a DCM underlier.
P4 · Curve digest	The shape of the surface on the grid, per expiry — a digest, not a million-bin histogram.	On-chain margin and portfolio risk need a shape. A single VIX-style scalar cannot tell a liquidation engine whether the wing is open. An oracle that consumes a signed strip of the curve is the difference between a margin system and a guess.	On-chain venues, clearing-style risk engines, dealers feeding automated hedges.
P6 · Term set	The same second-moment family at 9, 30, and 90 days. Variance and the root published separately.	Books are calendars. 0DTE made the 30-day point a slice, not the market. Term is how a PM hedges a 90-day mandate without pretending VX is the mandate. Publishing variance and vol separately is how Jensen stops being a hidden product feature.	Vol desks, structured-note issuers, anyone comparing a book to VIX / DVOL / BVIV without taking their marks as gospel.
P7 · Expected shortfall	Risk-neutral expected shortfall below 0.9F, 30-day, piecewise-linear g on the listed grid.	Capital rules and fund risk are ES. The public vol ticker is not. Until a shortfall line has a strip and a two-way, ES remains an internal number and the hedge remains a 30-day average. That is a basis the PM cannot close.	Risk desks, insurance, RWA wrappers that must talk to a capital model, tail funds.

Table 3. Desk feed — lines without a settling two-way
Feed	What it is	Consumer
Hourly Read	Signed bundle of the lines: state, curve, tails, premium. Paid from 5 October 2026. $12k/year as a reader; $36k with a desk licence.	Every consumer in Table 2, on day one, before they list anything.
P1b forecast premium	A view, not a line that settles. Stays out of every money line.	Traders. If it settled, it would be an advisor wearing an index’s clothes.
P2 armed state · P5 cross-venue · H1–H2	Flags and spreads. Publication hygiene, never a futures underlier.	Ops and desk. A health gate that silently became a settlement condition would be a product defect.

A line without a strip is a picture. A strip without a signed print is a mark. A print without a two-way is a number you can read and cannot fade. FloatNet records the print. The publisher writes the rule that builds the strip from the tape and names the line. Further instruments are further tapes (new underlier × market). Incremental cost is a data right and a grid, not a new company. The same signers validate the new strips. The same measurement set cuts the new surface into the same lines.

Table 2b. Two genera, one quorum
Genus	Underlier	How it pays	Status
II	F2 · density freeze (FN-F-v1)	Perp funding of mark to the signed freeze. Dual-input: datafi.live Q-clock and Hyperliquid P-clock.	Listed now. ETH Sepolia and Canton testnet desks already settle it.
I	P3 · tail mass (FN-g-P3-v1)	Auction in the listed wing. First money line.	Construction. 8 February 2027.
I	P4 curve · P6 term · P7 shortfall	Same signers, new strip on the same surface.	Construction on the same quorum.
Cousin	DVOL (Deribit 30-day)	Variance-swap / second-moment hedge.	A hedge of a different number. F2 is the number this ticket expires against.

Tokenized products on private wells versus one signed settlement main — Figure 4. Private indexes cannot hedge across the street. A money line on a Float Line — a slice with a signed strip — is a main that many wrappers can share.

6. For the registered staked signer

You are being asked to put your name and a cash bond on strip state other venues will read. That is the whole job. It is not a request to stand up an exchange, to warehouse customer flow, or to take the other side of a money line. Matching stays where you already match, if you match at all.

What you post. Identity. A Freeze licence. A USD deposit at tier ($25k / $50k / $100k), refundable on exit. You run the published strip from the named tape at the named times. If you miss, the deposit pays the sponsor of the money line that needed that print. You do not set the rule. You validate strip state. The line is cut from the strip you signed.

What you earn. $30,000 a year per sponsored product, paid by the sponsor, not through the publisher. At four products a lean node covers a B-tier stack; sooner if the same firm also buys the Read for its desk. The Freeze licence is integrity. It is not the demand engine of the publisher round.

Why a licensed firm is the right signer. A VASP, a market maker, or an OTC book already has the identity stack a registered node must show. Five unrelated signers live by 23 November 2026 is the named-desk integrity scoreboard. Listed-open is the architecture either way. If it misses: architecture unchanged; F2 remains a funding mark; do not fill seats with a DFY incentive.

What you are not. You are not the calculation agent of record in the old sense — the quorum is. You are not paid in a token for this seat. Settlement is dollars. Work credits for the open archive, if you also run one, are a separate and later path.

The first product is a money line on 8 February 2027: the tail-mass line (P3). Venue class TBD. Counsel first. A signer who is live in November is in the room when that strip needs five validations.

7. For the board

This preferred round buys the index publisher — the company that writes the rule that cuts lines from the surface and builds the strips, and sells listing and methodology licences so venues can settle wrappers without each writing a formula. Business Plan v0.20, adopted 11 September 2026, is the plan of record. The issuer is the FloatNet index publisher, a ConfigSecretAI subsidiary (in formation). Terms of the raise — ask, pre, cap table, two-state PV — live in the board pack, not on this page. Listings are the primary SKU. The 4-of-5 bonded quorum is integrity a regulator can ask for, not a customer cohort. W, NetCo, the v0.17 SAFE, and DFY-as-rail are parked or refused.

Cash arrives in a fixed order. The listing licence is the primary SKU. The Read is paid from 5 October ($12k). Settlement royalties are an option on listed notional, not in the base case. Year-3 exit run-rate on this architecture is $4.8 million; the in-year table is about $4.0 million. The publisher does not take a cut of the vote. NewCo revenue today is $0. Existing x402 / GM! / ACP cash stays at the parent and is unquantified.

The demand that makes the round more than a methodology paper is the gap in §2: VIX/VX already prove the stack is a business, on one slice, at one house. Visible does not mean twenty licences in hand. It means the first hour is desks that already quote. Live 11 September 2026: F2 is a funding mark on ETH and Canton; methodology licences 0; named firms 0 / 5. October and February convert that into cash. November is the integrity scoreboard for named desks.

8. Commercial rows

The SKU list, refused levers, and three-year cases live in the plan of record. They are not restated here. Working orientation only:

Table 4. Who pays what (orientation — plan of record governs)
Role	Out / in	Starts
Venue / protocol	$48k/year listing and methodology licence (primary SKU)	When a wrapper lists against a signed print
Reader	$12k/year for the hourly Read	5 Oct 2026
Registered node	USD deposit $25–100k refundable; Freeze licence as integrity	Lab now; named desks replace simulated seats
Sponsor	Royalty option; product revenue is the sponsor’s	First money line 8 Feb 2027

Table 5. Dates that convert demand into a signed strip
Date	What becomes real
This month	Paper the Read on live OTC books. F2 is a funding mark on ETH and Canton. Not yet paid.
5 Oct 2026	Hourly Read — paid public coordinate. If it misses: the coordinate stays free; hiring slows.
23 Nov	Integrity scoreboard: five unrelated registered nodes + counsel opinion. If it misses: architecture unchanged; F2 remains a funding mark.
8 Feb 2027	First money line: P3. Venue class TBD. Counsel first.

9. What has to be true

Listed-open is the architecture even while named desks are filling. Five unrelated firms live by 23 November 2026 is the integrity scoreboard, not the only reason the plan exists.
A settlement licence on a given tape requires the right to use that tape’s prices in the strip. The October Read and the February product run on the books we already study. Rights to a U.S. listed-option surface are required only when a strip is built from that surface.
On 8 February 2027 the first money line is live — a contract that pays against the tail-mass line. Venue class TBD. Counsel first. If it misses: no third tranche; the next licence is not “a copy of a working one.”

10. How to read this

If you run a licensed book and might sign: the job is §6, the strip you would validate is the variance-strip analogue for the named line, the date is 23 November. Paper the Read this month. The net is a ledger of those validations, not a new venue with your balance sheet underneath it.

If you invest in publishers of numbers: the round is in the board pack; the SKUs are in Business Plan v0.20. You are buying the publisher of the booklet (in formation). The net is the registry of signed validations of strip state.

Price is the apple. The Float Surface is the vol surface in gain-mass and loss-mass. A Float Line is a slice of it. A Float Strip is the variance-strip that makes that slice mechanical. A money line is the two-way that pays against signed strip state. FloatNet is the registry of those signatures. That is the stack. That is the sale.

FloatNet index publisher, a ConfigSecretAI subsidiary (in formation) · 11 September 2026 · Filename: floatnet-whitepaper. Drawn from the teaching text and CSAI Business Plan v0.20. Working figures in USD. Counsel has not opined. Not an offer to sell or a solicitation to buy any security or token.