idea-016 · feasibility deck
Every slide states a condition that would have to be true, then reports where it stands using the candidate file's own claim and its own confidence tag. A deck never upgrades a tag, invents a number, or recommends anything, and there is no ask slide.
The chair: engineering. The people who have to mould it, valve it, sterilise it, and produce the bench evidence that nobody outside the company is going to demand of them.
How to read this: every slide is a condition, not a conclusion. A bracketed claim reference points into knowledge-base/candidates/idea-016.md — the number is the claim's position in its ## Claims list, the tag is copied from it and never adjusted here. A [no claim] marker means nothing in the file speaks to this. See docs/deck-spec.md.
Would have to be true: The intended-use statement would have to reduce to a buildable spec: deliver sterile isotonic irrigation fluid into the canine external ear canal at a pressure capped at a fixed cracking value regardless of plunger force; hold a set insertion depth by graduated markings and an adjustable collar; aspirate through the outer lumen into a volume-graduated trap so instilled and recovered volumes can be read and compared; and do all of it through a soft, tapered, atraumatic tip in a size-matched family across canal diameters, with no energy source, software or electronics.
Where it stands: Three numbers the spec requires are not in the file. The cracking pressure has no numerical basis — the Generator identified no published irrigation pressure or flow threshold at which the canine tympanic membrane ruptures, and none informs the proposed valve [claim 9: unconfirmed]. The canal dimensions that would set the tip family and the maximum depth stop were in no claim and are now on record [claim 35: unverified]. The trap's reading only means something if recovered volume tracks residual volume, which is an untested hypothesis [claim 24: unconfirmed]. What is fixed is the label boundary: the fluid is sterile isotonic only, with no therapeutic claim, specifically to stay off the animal-drug route [claim 3: unconfirmed].
What would settle it: Canal length and diameter ranges are genuinely reachable with the tooling in this repo — literature / literature-detail index the veterinary anatomy and otology literature, which is the same asymmetry the file names for claims 6-9. The pressure and recovery numbers are bench items, not literature items.
If it's false: A spec with no acceptance criterion cannot be verified, and the bench programme has nothing to test against. That is exactly why Factor 6 sits at 2/5 with the mechanism called plausible and the central safety number called missing.
Would have to be true: There would have to exist a delivered-pressure window — below the canine tympanic membrane rupture threshold with a safety margin, and above the pressure required to clear ceruminous debris and exudate from the canal — and a fixed mechanical cracking pressure would have to sit inside it for every canal size in the tip family.
Where it stands: The window is an assumption this deck put on record, not a finding [claim 33: unverified]. Its upper bound does not exist in the literature: a targeted PubMed search this pass returned no canine irrigation-pressure rupture threshold, and the five near misses bound the gap rather than close it — porcine tympanic membrane pressures (PMID 12834126), eardrum-rupture physics (PMID 2497697), and canine tympanic assessment rather than threshold (PMID 2916310) [claim 9: unconfirmed]. Its lower bound has no claim and no literature at all — the sweep for canine ear-canal fluid-recovery and residual-volume work returned zero records [claim 24: unconfirmed]. The failure modes the cap is aimed at are documented, and the file notes the literature cuts both ways on their magnitude [claim 8: unconfirmed].
What would settle it: A cadaveric canine temporal-bone study measuring irrigation pressure to tympanic rupture across canal sizes — a funded bench programme, weeks and real money, and the file already names it as the minimum bench evidence needed before anything else. For the lower bound, a benchtop clearance study in a canine ear-canal model with synthetic cerumen. No connector in this repo reaches either.
If it's false: If the window is empty, or too narrow for a mechanical valve to hold across manufacturing tolerance, the product's entire differentiator is unbuildable at any price and the device reduces to a more expensive red rubber catheter [claim 16: unconfirmed].
Would have to be true: With no regulator writing the test plan, the company would have to impose one on itself — biocompatibility, sterilisation validation, shelf life and delivered-pressure verification — and would have to hold the label off the animal-drug boundary while doing it.
Where it stands: The absence of a pathway is the file's only settled ground. 21 CFR 807.65(b) exempts manufacturers of devices used solely for veterinary purposes from establishment registration, retrieved verbatim this pass [claim 1: verified]; 807.20(a), 807.81(a) and 814.1(a) each attach their duties to devices intended for human use, so there is no US premarket submission of any kind [claim 2: verified]. The self-imposed standards set that replaces it was in no claim and is now on record [claim 36: unverified]. The boundary that constrains the label is only half settled: 21 CFR part 514 and 21 U.S.C. 360b exist, but whether FDA would characterise a medicated ear-flush product as a drug is an agency interpretive judgement the eCFR connector cannot reach, and the statute itself was never retrieved because no connector here queries the US Code [claim 3: unconfirmed]. One row of the frontmatter is wrong and the file says so: the human-use counterfactual's part-874 locator is refuted — FDA's record for the human device type is product code OGQ, "Ear Irrigation Kit", Class 1, regulation 880.6960, and § 880.6960 "Irrigating syringe" is Class I and exempt from premarket notification under subpart E of part 807 [claim 4: refuted]. The refutation hits the NOT-LIVE counterfactual row only; it does not touch the launch route, which rests on claims 1-2. What it changes for engineering is the comparison: the human analog is a lighter pathway than the file assumed, not a heavier one.
What would settle it: An ISO 10993 biological-evaluation plan and a sterilisation-validation quote from a contract steriliser, against the fluid-path and mucosal-contact profile. For the label boundary, a regulatory-counsel read of the drug/device line for a veterinary irrigation product — explicitly out of the eCFR connector's reach, by that connector's own output.
If it's false: If a medicated variant is ever wanted commercially, the route changes from "no submission" to a new animal drug approval, and Factor 1's 4/5 would be describing a different product. If the self-imposed standards set is heavier than assumed, the Factor 7 inference that development is cheap and fast loses the thing holding it up.
Would have to be true: A single moulded inline mechanical relief valve would have to hold its cracking pressure within a stated tolerance across the full flow range produced by a hand on a syringe plunger and by a gravity IV line — and still be cheap enough to throw away after one dog.
Where it stands: On record as an assumption, with no bench data [claim 34: unverified]. The closest analog the file names is pressure-controlled pulsed lavage in human wound care (Stryker InterPulse), carried as a mechanism analog only and never as a predicate. There is no target pressure to design the valve against [claim 9: unconfirmed], [claim 33: unverified], and the file's own moat assessment concedes no technical barrier was identified that would stop a consumables manufacturer copying whatever design results [claim 21: unconfirmed].
What would settle it: A benchtop valve characterisation — cracking pressure versus flow rate across the manual-syringe and gravity ranges, with a calibrated transducer, across production tolerance — plus a moulder's quote at volume. Contract engineering; no connector.
If it's false: A valve whose cracking pressure drifts with flow rate makes the pressure ceiling a statement on the label rather than a property of the device. For a product whose entire safety argument is that ceiling, that is the worst available outcome, and it also invalidates the belief argument on the desirability deck's Slide 5.
Would have to be true: Before a live dog is involved at all, a bench and cadaveric rig would have to establish three things: the irrigation pressure at which the canine tympanic membrane ruptures; that the valve holds below it with margin across tolerance; and that recovered volume tracks residual volume closely enough for a veterinarian to act on the difference.
Where it stands: (a) is the file's named minimum bench item and does not exist in the literature [claim 9: unconfirmed]. (b) has no data at all [claim 34: unverified]. (c) is described by the file as the second-most important thing a bench programme would have to establish, with a PubMed sweep returning zero records — consistent with no data existing, and incapable of verifying an engineering hypothesis either way [claim 24: unconfirmed]. Whether the two bounds even admit a window between them is [claim 33: unverified].
What would settle it: A funded cadaveric canine temporal-bone programme for (a), benchtop valve characterisation for (b), and gravimetric or dye-tracer residual-volume measurement in cadaveric canine ear canals for (c). Weeks and real money; nothing short of measurement resolves any of the three, as claim 24's own note says.
If it's false: If the bench work cannot produce these three numbers, there is nothing to publish — which also removes lever (a) of the moat [claim 21: unconfirmed], the only proposed lever that could have become defensible. Whether any live-animal work follows this slide is a Tier 3 human decision, and nothing on this deck proposes one.
Would have to be true: A coaxial dual-lumen catheter with a soft tapered tip, an inline relief valve and a volume-graduated trap would have to be manufacturable, sterile-packaged and shelf-stable at disposable cost — and it would have to work on a wet table, on a sedated dog, through whatever suction the practice already owns.
Where it stands: The suction assumption was in no claim and is now on record: that ordinary US general-practice suction provides enough vacuum and flow to evacuate viscous ceruminous debris through the outer lumen at canine-canal sizing [claim 37: unverified]. Sterilisation and biocompatibility are likewise newly on record [claim 36: unverified]. The unit cost has to fit under a price that is itself only an assumption [claim 29: unverified], [claim 12: unconfirmed]. The human-factors half — whether a technician can set the collar, run the valve and read the trap correctly after one demonstration without adding anaesthetic time — is stated on the desirability deck's Slide 2, where the user chair does the work [claim 26: unverified].
What would settle it: A contract-manufacturer design-for-manufacture review and quote for a coaxial extrusion or moulding with an integrated valve; a bench trial of debris clearance against the vacuum and flow range of common veterinary suction units; timed usability sessions with technicians.
If it's false: If clinic suction cannot clear the outer lumen, the device needs its own vacuum source — which turns a disposable into capital equipment and puts it directly against the OtoPet-USA console it was deliberately positioned beneath on price and capital [claim 20: unconfirmed].
Would have to be true: The specific valve geometry, the coaxial tip and the graduated inline trap would each have to be designable around any live blocking right — and the workaround would have to still be the product, not a compromise of it.
Where it stands: Nothing is known, in either direction. No patent or freedom-to-operate search of any kind was performed for pressure-limited irrigation, coaxial dual-lumen irrigation-and-aspiration catheters, inline mechanical pressure-relief valving, depth-limiting collars or graduated inline traps; no number, family, assignee or expiry is named, and the patent connector resolves a specific patent number rather than searching prior art, so there was nothing to invoke [claim 18: unconfirmed]. The working position that the base mechanism is old art is explicitly an assumption, not a search result [claim 19: unconfirmed] — the one adjacent datapoint, a cleared human ear-irrigation set from 1989 (K884786), is recorded as context only and says nothing about any patent's scope or expiry. The veterinary lever gives no relief here: 35 U.S.C. 271(a) reaches making, using, offering to sell or selling without reference to species [claim 17: unconfirmed]. Named search targets are already in the file: OtoPet-USA, Nupur Technologies, Stryker/InterPulse, and ENT irrigation-suction handpieces.
What would settle it: A human-run USPTO/EPO prior-art search against those four targets — free analyst hours — and a counsel reading of 35 U.S.C. 271(a), which no connector in this repo can retrieve because the eCFR carries regulations, not the US Code.
If it's false: FTO blocked is an automatic kill in the rubric regardless of every other score, and a Scorer must not read the absence of a refuted FTO claim as clearance. As an engineering matter, a blocking claim over inline relief valving forces a different pressure-limiting mechanism entirely and re-opens Slides 2 and 4. The same question is the viability deck's load-bearing condition, read there as a business risk rather than a design constraint.
Every condition above with nothing verified behind it — the [no claim] markers, plus the unverified claims this deck itself put on record. Read this slide first.
literature today; the other two are bench work.
Two conditions on this deck do have verified claims behind them, and both are on Slide 3: that there is no US premarket submission to satisfy [claim 1: verified], [claim 2: verified]. That is permission to build without a reviewer — it is not evidence that the thing can be built. Note also the one refuted line in the file, [claim 4: refuted], which corrects the human-use counterfactual to a Class I, 510(k)-exempt product code and therefore lowers the barrier for a human-side entrant rather than the candidate's own burden.
If only one thing from this chair could be checked: whether a delivered irrigation pressure window exists at all — below the canine tympanic membrane rupture threshold with a safety margin, and above the pressure that actually clears debris [claim 33: unverified], with its upper bound the number the literature does not contain [claim 9: unconfirmed].
Everything else on this deck is downstream. The valve (Slide 4) has no target without it; the spec (Slide 1) has no acceptance criterion; the bench programme (Slide 5) has nothing to measure against; and the safety argument the desirability deck depends on has no number to show a veterinarian. It is also the one condition here that cannot be shortened by a cheaper instrument — PubMed has already been asked this pass and returned only near misses in the wrong species, so the answer comes from cadaveric measurement or not at all.
Naming it is not a recommendation, a gate, or a kill.