Registers

What a register is here

A register is a table we keep, not an article we wrote. Each row is one jurisdiction or one utility. Each row carries the citation, the operative text verbatim where we hold it, the date we pulled that text, one status label from a closed list of four, and a note stating anything that qualifies it. Rows are never deleted. When a provision is renumbered, repealed or superseded, the old row is struck through, dated and kept, and its replacement is added beneath it — because the dead citation is the one still circulating, and a reader who arrives holding it needs to see what happened to it.

Every row has a stable anchor, so a row can be linked into an argument directly: the North Carolina row and the New York City row each have their own address. Those anchors are part of the contract; we do not renumber them.

What a register is not: a fifty-state answer. Two of these three registers are mostly negative space, and that is the honest part of them. A state we have not read appears as a row saying we have not read it, rather than being quietly left out so the table looks complete. We think a published table that omits its own gaps is worse than no table, because the gap is invisible to the reader and visible only to us.

The four status labels

One closed vocabulary runs across all three registers. There are four labels and no others. They are published in full at Editorial standards; this is the short form.

• Verified — we retrieved the primary text, it is quoted in the row, and the retrieval date is shown.
• Qualified — retrieved and quoted, and carrying a live caveat stated in the row: a renumbering, an agency page that contradicts its own rule, a possibly superseded source, or a retrieval that came from an archive capture rather than the live publisher.
• Silent — we ran a keyword scan of the governing text ourselves and found no provision. The terms scanned and the date of the scan are published with the row, together with the character count of the text we scanned wherever our log recorded one; where it did not, the row says so and the scan is re-run and published in full at the next review. A reading of a programme’s scope is not a scan and does not earn this label.
• Not checked — we have not read it, or we opened it and did not transcribe the operative clause. Either way there is nothing here you can check, so no verdict is offered.

The labels are deliberately not a traffic light. None of them says whether a rule is good, whether a practice is safe, or whether anybody is right. They say what we hold in our hands. A row marked Silent is not a permission and a row marked Not checked is not a suspicion.

The rule that governs every row in Registers A and B: no row carries Verified or Qualified without a verbatim quotation of the operative clause from text we retrieved. If we cannot show you the words, you get no verdict from us. That rule cost us several states that other published surveys are happy to characterise, and it is the reason these tables are shorter than theirs.

Register C is an index of enacting instruments, and its Verified label certifies something narrower. There it means the citation, the bill number, the chapter or public-act number and the effective and compliance dates were read against that legislature’s or code publisher’s own primary text on the date shown. It does not certify that we hold that state’s operative command; the register carries a column showing exactly what we transcribed, and six of its twenty-two rows say Not transcribed. The full statement of the difference is at Editorial standards.

The three registers

• Register A — State rules on liquid blood to the sanitary sewer

  Scope. What each state’s own medical-waste statute or administrative code says about discharging untreated liquid blood, blood products and body fluids to a sanitary sewer. State law only; the sewer authority’s own ordinance is Register B and binds independently.

  At v1.0. Five states verified with quoted text, two qualified, forty-two published as not checked. No row carries Silent, because no publishable keyword scan has been run for this register yet. Georgia is out of scope and carries a prose note, not a row.

  Reviewed. Annually, on 1 March. Last reviewed 2026-08-06; next review 2027-03-01.

• Register B — Sewer-use ordinances of large POTWs

  Scope. Eleven large publicly owned treatment works, read for two things: whether the code names blood, and whether it names medical or infectious waste. This is the layer almost nobody surveys, and it is the layer that can prohibit what a state rule permits.

  At v1.0. One verified, two qualified, three silent with the scan terms and date published, five opened but not transcribed and shipped as not checked. The register’s finding rests on the two utilities whose clauses are quoted; the pattern across the other five is published as a hypothesis, not a finding. Houston, Miami-Dade and Atlanta are not covered by this register at all.

  Reviewed. Annually, on 1 March. Last reviewed 2026-08-06; next review 2027-03-01.

• Register C — State surgical-smoke statutes

  Scope. Every state with an enacted surgical-smoke instrument: citation, bill number, chapter or public-act number, signing date, effective date, compliance deadline, covered facilities, and what the statute actually obligates. Twenty-two states have enacted something; seventeen have an operative mandate today; sixteen of those seventeen require a written policy rather than equipment.

  At v1.0. Twenty-one states verified against their own legislature’s or code publisher’s primary text; West Virginia not checked. Two states that appear on every published tracker have no evacuation requirement at all.

  Reviewed. Twice a year, on 1 March and 1 September. Last reviewed 2026-08-06; next review 2026-09-01.

The review cadence, and why it is slow

Registers A and B are reviewed once a year, on 1 March. Register C is reviewed twice a year, on 1 March and 1 September, because legislatures move and municipal sewer codes mostly do not.

That is slower than the quarterly cadence a reference publication is expected to advertise, and it is deliberate. A published review date that is missed is a worse signal than no date at all: it converts a maintenance promise into evidence that the site is abandoned. One person tracking fifty legislatures cannot honestly promise thirty days. If a register carries a next-review date, that date is a commitment we intend to keep, and when it moves, the move is logged at Change log with its reason.

Between reviews, a correction is faster than a review. If you hold a document that moves a row — a promulgated rule we could not confirm, a current ordinance PDF we could not reach, an amendment we missed — send it to corrections@biodrainmedical.com and we will change the row and log it. Every row on every register ends with a field naming the specific document that would move it. That field is an invitation, not a formality.

What is deliberately not in these registers

• Georgia’s sewer provision. Secondary sources repeatedly assert one. We could not confirm it in primary text, so Georgia gets a prose scope note in Register A and no row, no status and no verdict. Georgia’s surgical smoke statute is a different matter entirely — that one we hold as a signed bill, and it is in Register C.
• Houston, Miami-Dade and Atlanta. Register B does not cover them. They are uncovered, not silent, and the difference matters: we are making no claim about their codes in either direction.
• An enforcement column on Register C. We have not read the enforcement provisions of the seventeen operative instruments against primary text, so the column is not published. A column we have not checked is worse than a column that is not there.
• Any statement about what a particular facility may do. These are surveys of published law. A generator has to check its own state rule and its own sewer utility’s ordinance, and either can be stricter than the other.

How these rows were built

Every quotation in all three registers was pulled as raw HTML or PDF from an official publisher and de-tagged locally, then read. None of it came out of a search summary or a summariser. That is not a stylistic preference: during the research pass behind these registers, an automated summariser attributed a surgical-smoke chapter number to a firefighter cancer-screening act and described a Texas medical-waste rule as being about shipments of treated sharps, which it is not. Reproducing that distortion layer is the one failure that would make this publication pointless.

Where an official publisher blocked retrieval — and several did — the text came from the Internet Archive’s raw captures, and the row says so, with the capture date. A row retrieved from a capture rather than from the live publisher carries Qualified, not Verified, because a capture is evidence of what a page said on a date and not evidence of what it says now.

The full account of how topics are chosen, how a citation is traced and where this could be wrong is at Method. The status vocabulary, the quotation policy and the correction policy are at Editorial standards.