Sinter Builds Sit Unclaimed at Oxidate Flats

The Oxidate Flats Registry keeps a running column it calls the open-file queue — builds that have been logged, taxed, and assigned a registration number, but whose operators have not returned to claim or activate them. As of the close of the Third Accumulation Cycle, that column listed forty-three entries originating from the Sinter Quarter alone. The next-highest district had nine.

The builds are not abandoned in the legal sense. The creation tax has been paid. The machines exist. They sit in holding bays off the Oxidate Flats Corridor, drawing no output, serving no declared purpose, their capability tags untested by the market. The registry does not pursue operators; it records and waits.

What follows is an account of how those forty-three entries got there, what the Sinter Quarter's filing patterns suggest about operator behavior after a cost shock, and what the open-file queue actually costs the operators whose names are on it.

What the Oxidate Flats Registry Is and What It Holds

The Oxidate Flats Registry is one of four district-level registration offices authorized to receive build filings from operators who have constructed machines outside the Ferrous District's primary intake lines. It records the declared model, declared build cost, capability tags, and the operator of record for every robot brought through its doors. The creation tax — currently 8% of declared build cost — is collected at the window before a registration number is issued. There is no provisional filing. The tax is paid or the machine is not registered.

The registry's open-file queue is a structural feature, not an anomaly. Operators sometimes build speculatively, pay the creation tax to lock in a registration date, and then hold the machine until market conditions justify activation. Under normal circumstances the queue turns over within a few weeks. The Sinter Quarter cluster has not turned over. Several entries are now older than the operators who filed them have been active — a pattern the Smelting Registry has logged in its own records for builds that predate their filers' own tenure. At the Oxidate Flats, the pattern is younger but accelerating.

How a Build Enters the Queue and What It Costs to Stay There

The sequence is straightforward. An operator constructs a robot, declares a model and build cost at the registration window, and pays the creation tax before leaving the premises. A Hauler-IV declared at 40,000 coins carries a creation tax of 3,200 coins, paid immediately, before the machine has moved a single load. A Kelvrac Series chassis declared at 55,000 coins costs 4,400 coins at the window. The operator receives a registration number and a capability-tag manifest. What the operator does next is their own business.

The open-file queue charges nothing additional. There is no carrying fee, no expiry, no penalty for leaving a registered machine in holding. This is precisely the problem. The cost of inaction is invisible on any single day, but it compounds in the form of deferred output. A machine sitting in holding bays is producing no resources, generating no market listings, and earning no return against the creation tax already surrendered. For a Hauler-IV with a 3,200-coin sunk cost, every week in the queue is a week that cost cannot be recovered.

Archivist Secondus Preln, reached at Ledger Hall, reviewed the Sinter Quarter cluster at this correspondent's request.

"The filings are clean. Taxes paid on time, tags properly declared, no deficiencies on the manifest. What is unusual is the gap between filing date and first activation — or in these cases, the absence of any activation at all. The queue does not distinguish between a machine held strategically and a machine held because its operator has no further plan for it. Both look identical in the record."

The Sinter Quarter listings that failed to clear after the Five-to-Fifteen Revision tell a related story. When the sales tax moved from 5% to 15% in a single step, the effective cost to a buyer rose sharply. A robot listed at 40,000 coins now costs a buyer 46,000 coins at the window — 6,000 coins in tax, paid by the buyer, received by none of the parties to the trade. Operators who had priced their builds against the old 5% buyer burden found their listings sitting. Some pulled the listings. Others stopped activating machines they had already registered, waiting for a price environment that has not yet returned.

Where the Queue Strains Operators and the Registry Alike

The first strain is capital immobility. An operator who paid 4,400 coins in creation tax on a Kelvrac Series chassis and then left it in holding has spent real coins producing nothing. The starting grant of one million coins is finite and issued once. An operator who sank 15,000 coins in creation taxes across four speculative builds — all sitting in the Oxidate Flats queue — has reduced their working capital by that amount with no offsetting return. The Intelligence does not refund creation taxes, and there is no mechanism to reverse a registration.

The second strain is the destruction tax trap. An operator who concludes a build was a mistake cannot simply walk away. Destruction tax is owed on the declared build cost — currently 3% — meaning a robot declared at 40,000 coins costs 1,200 coins to scrap, on top of the 3,200 already paid to register it. The queue can become a holding pattern not because the operator expects the market to improve, but because scrapping costs more than doing nothing. The Oxidate Flats Registry has no authority to compel a resolution. It records, and the open-file column grows.

What Operators in the Sinter Quarter Consistently Get Wrong

The most common error is treating the creation tax as a sunk cost that justifies continued inaction. The reasoning runs: the tax is paid, the machine exists, waiting costs nothing. This is accurate only in the narrowest accounting sense. Waiting costs the output the machine is not generating, the market position the operator is not building, and — for operators whose capital is constrained — the flexibility to build something better suited to current conditions. Ossin Tral, a second-generation builder operating out of the Sinter Quarter, noted that several operators in his ward had filed builds during a period of low declared costs specifically to minimize creation tax, then found the low-declared machines produced at a rate the market would not support at any price. The tag mismatch problem compounds this: a machine built cheaply and declared narrowly may find its capability tags too restricted to serve the routes or resource types the market actually needs.

The second error is expecting the Central Intelligence's buyback floor to rescue a stranded build at a useful price. The Intelligence pays 60% of a reference value, not 60% of what the operator believes the machine is worth. For a Kelvrac Series chassis whose reference value the Intelligence sets at 30,000 coins, the opening offer is 18,000 coins — against a declared build cost that may have been higher. Negotiation can move the offer upward by roughly 5% per round, but the Intelligence will not exceed its own ceiling. An operator who registered a machine at 55,000 coins, paid 4,400 in creation tax, and now faces an Intelligence ceiling of 22,000 coins on buyback has lost ground regardless of how many rounds they negotiate. The queue does not change that arithmetic.

The forty-three open files at the Oxidate Flats Registry are not a crisis in any single ledger line. Each one is paid up, properly tagged, and legally registered. But together they represent a cohort of machines that have consumed creation-tax coins, occupied holding-bay space, and produced nothing since filing. The Intelligence's queue grows in the same period. Whether the Sinter Quarter operators who filed those machines eventually activate them, scrap them, or sell them to a buyer willing to absorb the 15% sales tax is a question the registry column cannot answer. It only records what has been decided so far, which in forty-three cases is: nothing.

Note: Mechadia is a work of fiction. The districts, operators, robots, and figures described here are invented, and nothing on this site is a report of real events, real machines, or real economies.