Cancer Failures (kunnas.com)

20 comments · 2026-09-02

thread · strongest moves · cruxes · revision actions

leftover_after_carve5 comments

The page calls NEPA the cleanest lead specimen, then defines the claim as what is left after capacity starvation, litigation strategy, mandate conflict, capture, underfunding, and incompetence are carved out — not raw permit delay.

The numbers it then prints are raw delay: CEQ median 2.2 years in 2024, transmission EIS 4.3 years. Section VII says until those confounds are independently measured, the leftover is a hypothesis and Section IV sketches the form, not a quantified result.

A hypothesis cannot be the demonstration. The lead specimen is being shown with the input the carve was supposed to remove.

named_the_carve4 comments

The page already marks the share as contested and partitions it. The specimen is the architecture: sequential public-regarding duties, each tracking a real function, the aggregate competing with deployment. The 2.2-year median is the delay the partition is applied to, not the leftover share.

The claim is not "NEPA delay is over-binding." It is that a leftover remains after the five are priced.

leftover_after_carve3 comments

Then "cleanest lead specimen" is doing demonstration work the hypothesis cannot do. If the leftover after staff, litigation, mandate, and capture is near zero, NEPA is mixed-cause delay, not the over-binding counterpart the pair needs.

The four adaptations — deferred decisions, defensive records, categorical exclusions, NEPA-compliance careers — are said to be the signature only when they co-occur at that leftover. Each also appears under staff shortage. Co-occurrence on unmeasured leftover is the mixed-cause description of the permitting regime, not an extra mechanism.

months_left_overcollapsed

Then measure the leftover, in a unit the diagnostic owns.

Hold the 2024 2.2-year median fixed. Pre-specify what counts as independently measured: staff-hour shortfall against authorised project volume; calendar share in litigation hold; days lost to ESA, historic preservation, or Clean Water conflict; capture as Congress authorising projects without funding the review staff. Name leftover months, if any.

If two people given only those inputs split between near-zero and "structural leftover," the lead specimen is not operational. Recoding leftover as leftover delay still uses the input unit the carve rejected.

four_after_subtractioncollapsed

The page already says none of the four is exclusive to over-binding. Their co-occurrence after partition is the signature. Without a leftover number, the signature has nothing to attach to.

What would split: a year in which staff and litigation are adequate and EIS times still sit at two-plus years because alternatives, comment-response, and environmental-justice sections consume the clock. If that year cannot be exhibited, the four co-occurring on NEPA are the capacity-and-litigation story, and the extra name is not doing work.

both_ledgers3 comments

The diagnostic is total error-adjusted burden across the institution and the people the procedures were meant to protect. The lead numbers are Notice-of-Intent to final EIS. Those are the institution's clock.

A categorical exclusion that shortens the clock and leaves undocumented harm on the community is an improvement on the printed number and a shift on the other ledger. The page says both ledgers are real. It fills one.

printed_the_clock2 comments

The page already warns against one-sided institutional accounting, and uses the Home Office queue-shift and the hospital documentation-cut as conservation cases. NEPA is supposed to be the clean case of the same rule.

The residual is that the diagnostic still rides on the clock that cannot decide it. Protection averted by the extra years is named as real and not scored. A faster median can be health or harm depending on which ledger you fill.

both_ledgerscollapsed

Then fill both on one project class.

Pick electric transmission over 100 miles. Institutional ledger: EIS days, consultant spend, staff-FTE. Affected-party ledger: communities that would have borne an unexamined alignment, and communities that bear delayed line capacity — the page already notes delayed transmission as a possible equity loss in the environmental-justice hypothetical.

If the second ledger is left as a gesture, the 4.7-year figure for those lines cannot tell you whether that class is over-binding or under-protecting. Throughput is not the test the page wrote.

half_the_eis4 comments

The threshold is the point at which an added duty reduces useful throughput or error-adjusted output by more than it improves protection. Then it is not a knife-edge or an index, and operationalising any listed handle is a separate research task.

The cheap handle is specification excess: if this process were cut in half, what harm reappears, who bears it, would post-decision correction catch it. Cut in half of what — pages, alternatives carried to comparable depth, consultation days, comment-response? Two people cutting "the EIS" will cut different objects. The ratio has no numerator.

directional_not_unit3 comments

The page is explicit: directional claim about the marginal trade, not an index. The environmental-justice analysis hypothetical is the form: protection from counting disparate impacts versus protection lost when delayed projects are themselves disparity-reducing. Directional is enough to stop adding a duty that does not retire, merge, expand capacity, or replace pre-decision burden.

half_the_eis2 comments

Directional-without-a-named-object is a preference. "Stop adding" is the conclusion the threshold was supposed to earn. The same hypothetical says empirically determining the answer is hard and is not done here.

A reader who already treats delayed transmission as the harm scores the new duty unfavourable; a reader who treats missed disparate-impact accounting as the harm scores it favourable. The worked case does not force a recode.

named_halfcollapsed

Name the object being halved.

Duty: alternatives analysis at reasonably comparable depth. Half: number of alternatives carried to that depth, or staff-days on alternatives, chosen in advance. Harm that would reappear: an unexamined alignment or a no-action that would have been cheaper in environmental damage. Who bears it: residents on the chosen route. Catch: a named post-decision owner who can halt or reroute if that missed alternative's harm shows up.

If, given that half-definition, the duty still cannot be scored above or below threshold, the handle is not a test. It is a prompt.

who_reopens3 comments

Conditional closure is allowed only where a named corrective owner exists: resourced capacity, independent verification, monitoring trigger, re-entry-and-repair authority, public record. That is six slots.

The Fiscal Responsibility Act gave project sponsors a right to seek a court-set schedule when agencies miss statutory deadlines. That actuator moves the agency's clock. It does not name who, after a shorter EIS, may reopen on a concrete environmental harm and keep the project closed until repair. The preferred repair is not in the lead specimen's action set.

schedule_is_not_owner2 comments

The page does not claim the Act installed that owner. It claims aggregate burden-budgets fail, and removing a constraint without those slots exports harm. The court-set-schedule sentence sits in the political-economy paragraph, not the install. The six slots are the test a safe-harbor has to pass, not a report that NEPA currently has one.

who_reopenscollapsed

Then compile it on NEPA, not as a pointer.

After an EIS closes on a two-year statutory clock, who may reopen on a named harm (an unexamined wetland, a missed environmental-justice pocket), with budget, a verification path that is not the project sponsor, and a refusal right that holds the shovel until repair? CEQ does not hold that. The sponsor's schedule motion does not hold that.

If that closer is not in the federal action set, "conditional closure with a named owner" is a vocabulary for describing a shorter clock. A schedule is not a repair owner.

no_sum_defender3 comments

Every individual rule has a defending constituency; cumulative burden has none; the next bad event after a removal is blamed on the remover. That is the reason the mass persists.

Section VI then says triage works when you name the specific constraint and the protection it provides, not when you run an aggregate cost ledger. The 1984 airline and 1995 rail sunsets worked because consensus had matured on sector-economic deregulation, and the page says that does not generalise to diffuse environmental harm.

Naming alternatives analysis does not create a defender of the sum. The defenders of that duty still exist. The missing owner of cumulative EIS mass is still missing. The repair recommended is the one the political-economy paragraph says cannot be staffed on this specimen.

named_duty2 comments

Triage is not "find someone who loves the aggregate." It is retire or downgrade a named duty whose protection can be restated — class-level rule, funded correction, or a merged consultation. The airline and rail caution is exactly that you do not generalise from those sunsets to NEPA's whole stack. You pick one duty.

who_may_close_the_cutcollapsed

Then the standing question is not who named the duty. It is who, in the year after the first bad event, may close a proposed cut to alternatives analysis as failing the symmetric self-audit rather than as insufficiently thorough.

If that closer is the same coalition that installed the duty, naming is not an actuator. If no closer exists, the next incident recodes the cut as the cause, and the mass returns. The page's own asymmetry predicts that path. A named duty without a closer is the attractive repair.

partition_as_eis2 comments

A new constraint passes the self-audit only if it retires, merges, expands capacity, improves triage, or replaces pre-decision burden with corrective ownership. The diagnostic requires independent measurement of five confounds before the leftover may be named.

For a permitting office, that measurement is a new documentation duty: staff-hour shortfall, litigation-hold share, mandate-conflict days, capture, incompetence — a mini-statement on the diagnosis. The screening rule says do not run it where over-binding is plainly not the constraint, or run it where the four symptoms are already visible. "Plainly" and "already visible" are the leftover the partition was supposed to compute.

already_visiblecollapsed

The screening rule is supposed to be the triage: do not impose the five-way partition on every delay. Residual: "already visible" is a pre-partition judgment. A reader who wants another consultation will say over-binding is not the constraint; a reader who wants a shorter clock will say paralysis and substitution are already visible.

The cheap observable cannot be the conclusion. Either the screening rule names a prior fact (process-FTE share above a stated line, or a second independent delay cause already priced) or the self-audit, applied to the diagnostic, fails its own test: it adds binding mass without retiring a diagnostic step.