Tag Conflicts Can Void Both Build Declarations
A dispute filed at the Smelting Registry last quarter has worked its way into the standing reference files at Ledger Hall, not because it is unusual but because it is unusually well-documented. Two robots on the same Ferrous District line — a Kelvrac Series smelter and a Refinery-class unit commissioned by the same operator — declared capability tags that the Registry ruled were in direct conflict. The Registry voided both build declarations. The creation taxes, totalling just over 7,400 coins, were not returned.
The mechanism that produced this outcome is called a tag-conflict void. It does not appear prominently in the operator handbook, and many foundry crews encounter it only after it has already cost them something. The Smelting Registry has recorded nine such voids in the current accumulation cycle, all involving co-deployed robots with overlapping output classifications.
By the end of this piece, an operator should be able to identify the conditions that produce a conflict, understand why the Registry treats both machines as the liable party, and know what a voided declaration actually means for the robots still standing on the line.
Discover luxury hotels, chic city stays, and beautiful escapes around the world.
What a Build Declaration Is, and What Voiding One Does
A build declaration is the formal record — entered at the point of construction and written immediately to the ledger — that establishes a robot's model, purpose, and capability tags. Those tags are not descriptive labels. They are the machine. A robot can only produce resource types that match its declared capabilities, and that constraint is enforced at the point of output, not at the point of inspection. No tag, no production. Wrong tag, no production.
Voiding a declaration does not destroy the robot. The chassis remains on the line, the operator retains ownership, and the machine can be relisted or scrapped. What disappears is the legal standing of the declaration itself: the robot's outputs are treated as unclassified, and unclassified outputs cannot be traded on the open market or accepted at a forge. The robot becomes, in the Registry's language, a declared null — present, powered, and producing nothing the economy will recognize.
How the Conflict Arises, and What the Registry Actually Examines
The Smelting Registry reviews capability tags not in isolation but in the context of a production line. When two robots share a line designation and their declared output tags overlap at the sub-classification level, the Registry applies what its published guidance calls a mutual exclusion test. The test asks a single question: could both robots produce an identical resource type under their respective declarations? If yes, the Registry treats the overlap as a structural conflict and voids both declarations simultaneously.
The Ferrous District case illustrates the arithmetic. The Kelvrac Series unit was declared with a build cost of 52,000 coins and carried output tags including refined ferrous intermediate, grade two. The Refinery-class unit was declared at 40,000 coins with tags that included ferrous intermediate, secondary refinement. The Registry's classification table treats those two tags as equivalent at the sub-classification level. Creation tax on the Kelvrac unit came to 4,160 coins at the current 8% rate; the Refinery-class unit added 3,200 coins. Both sums were paid at construction and neither was recoverable once the void was issued.
"The operator filed a correction petition within the window, arguing the tags referred to different process stages. The Registry's position was straightforward: the ledger records what was declared, not what was intended. Declarations are tested against the classification table as written, not against the operator's account of their own meaning."
— Archivist Secondus Preln, Ledger Hall, responding to a Records Colloquium query on the Ferrous District void
The correction petition window is forty-eight hours from the void notice — the same interval used for disaster forecast windows. Petitions that succeed require the operator to submit amended declarations and pay a second creation tax on the revised figures. There is no partial credit for the first tax already rendered. Operators who have been watching tag bloat drive up declared build costs will recognize the compounding pressure: a higher declared cost means a larger creation tax lost if the void stands.
The mutual exclusion test applies only to robots sharing a line designation. Two robots in separate districts carrying identical tags will not trigger the test, because the Registry's conflict model is spatial, not global. This boundary is precise and consequential: operators who split production across districts specifically to avoid the test are doing something the Registry permits, whether or not that was the design intent.
Where the Mechanism Breaks Down and Who Pays
The most consistent failure point is the classification table itself. Sub-classification labels are added to the table when new resource types enter the world through accepted forge recipes, and the table is updated at the Registry's own schedule. An operator who built a line before a table update may find that two previously distinct tags have been collapsed into the same sub-classification in the current version. The robots have not changed. The declarations have not changed. The table moved, and the conflict was created retroactively. The Registry does not issue advance notice of table revisions that produce new conflicts on existing lines.
The second failure point sits at the intersection of the void and the destruction tax. An operator holding two declared-null robots faces a choice between leaving unproductive chassis on the line or scrapping them. Scrapping triggers the destruction tax — 3% of declared build cost. On the Ferrous District pair, that would have added another 2,760 coins on top of the already-lost creation taxes. The operator who commissioned both machines absorbs all of it. There is no provision in the Registry's current framework for shared liability between operators on a jointly managed line, even when the conflict arises from a co-deployment arrangement rather than a single operator's planning error.
What Operators Consistently Get Wrong About Tag Conflicts
The most common piece of received wisdom is that a conflict void is triggered by identical tags, and that operators who use different label text are safe. This is false. The Registry tests against the classification table's sub-classification mappings, not against the literal string of the declared tag. Two tags with different names can map to the same sub-classification. Operators who have relied on surface-level tag differentiation without checking the current table have discovered this at the cost of their creation taxes. As a build cost is a statement the market will later test, a tag declaration is a statement the Registry will test against its own table — not against the operator's reading of it.
The second misconception is that voiding one robot's declaration will protect the other. Operators sometimes attempt to pre-empt a Registry ruling by voluntarily amending one declaration before a void notice is issued. The Registry's guidance is explicit: once a conflict is identified during a line review, both declarations enter the mutual exclusion test together. A unilateral amendment filed after the review opens does not remove the second robot from the test. Both machines remain subject to the void until the Registry closes the review, regardless of any amendments submitted during that window. Acting quickly matters — but acting before the review opens is the only move that reliably changes the outcome.
The Smelting Registry's conflict framework was designed to prevent redundant output classifications from distorting the production ledger. It accomplishes that. What it does not do is absorb any of the cost when the conflict arises from a table update rather than an operator's planning error. The creation taxes flow to the Treasury either way, the chassis remain on the line either way, and the operator holds the difference. Nine voids in a single cycle suggests the table and the lines it governs are moving further out of alignment, not closer together.
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.