Seven pulls on the line means two different things.
One is the standard line-pull code and one is the closed-circuit code, and both are printed in the same manual. If your flashcards only know one of them, they will teach you the wrong one, and they will be extremely convincing about it.
USN Diving Master is an Anki deck: 965 cards for eleven Navy diving watchstations, with a citation on every single one. $49 for the deck, $129 with the fourteen training drills. One payment, no subscription.
Both answers are correct. That is the problem.
Table 8-2 is the standard line-pull code, and in it seven pulls means “go on (or off) searching signals.”
Table 15-6 is the closed-circuit code, the one you are on when you dive an MK 16, and there seven pulls means “I have started, found, or completed work.”
Three signals collide between those two tables. Here is all three of them.
| Signal | Table 8-2 | Table 15-6 |
|---|---|---|
| 7 pulls | Go on (or off) searching signals | I have started, found, or completed work |
| 3-2 pulls | Ventilate, or “more air” | Breathing from EBS |
| 2-1 pulls | I understand, or “talk to me” | Unshackle from the lazy shot |
Both tables, USN Diving Manual Rev 7 Change A. Table 8-2 is in Volume 2, Table 15-6 in Volume 4.
Neither table is wrong. They are different codes for different rigs, and the manual prints both, in two different volumes of itself.
So which one is on the card you have been reviewing every night for three weeks?
If you cannot answer that, the trouble is not that you forgot. The trouble is that the card never told you which table it came from, and you have been practicing it anyway.
Anki has no opinion about whether a card is true.
It does one job, and it does it well: finding the facts you are shakiest on and showing them to you until you are not shaky any more.
That is the whole mechanism, and it works on wrong facts exactly as well as it works on right ones.
Spaced repetition takes whatever is on the card and installs it as certainty. A plausible wrong number here is worse than a missing card.
That sentence is the first rule in this project's working notes, written before a single card existed.
Here is why it matters at a board. A gap in your memory feels like a gap. You hesitate, you say you would pull the manual, and the Master Diver files that away and moves on. A wrong fact you have drilled four hundred times does not feel like anything at all. You answer immediately, and you answer wrong, and you have no idea, because certainty is exactly what the drilling was for.
Nobody sets out to write a wrong card.
Wrong cards happen because facts travel and leave their sources behind. Four ways it goes wrong, all of them ordinary.
The chapter numbers moved.
Revision 7 renumbered the manual. Treatment of decompression sickness and arterial gas embolism used to be Chapter 20. It is Chapter 17 now, and recompression chamber operation used to be 21 and is now 18. Any notes written before December 2016 still say 20 and 21, and those notes are still in circulation, still being photocopied, still being handed to the next guy. A citation to Chapter 20 cannot be checked against a current manual. It also does not look wrong, which is worse.
The job title changed.
Change A renamed the Diving Medical Officer to the Undersea Medical Officer. If a card says DMO, you have just learned roughly when it was written, and nothing else about it has been checked either.
Should and shall are not synonyms.
In a manual they are two different obligations, and a card that softens one into the other teaches you a decision you are not authorized to make.
A shared deck has no revision at all.
It has a filename. You cannot tell which manual it was built from, whether a paragraph it depends on was rewritten in 2018, or whether the person who made it had the manual open at the time.
The enemy here is not the guy who typed up the locker's notes. He did you a favor on his own time and he was probably right about most of it. The enemy is a fact with no source attached. You cannot check it, you cannot update it, and you cannot tell it apart from a fact you can.
The manual is free. Say it out loud.
All 987 pages of it, on the public internet, no login. So the honest objection goes like this: I already have the source document, why would I pay for something built out of it.
Fair. Here is the entire answer.
The manual is free. Reading it is not.
Nobody finishes the Diving Manual. Not because divers are lazy, because it is 987 pages of reference text that was written to be looked things up in, not read straight through, and because it sits behind a full working day of dive ops and maintenance and PQS chasing.
What you would be paying for is not the facts. The facts belong to the Navy and to the public. You would be paying for 965 decisions about which facts your watchstation actually gets asked, each one carrying a pointer back to the paragraph it came from so you can go argue with it.
How this deck is built
Four moving parts, none of them clever. All four are boring, which is exactly the point.
1. A citation on every card.
Not “the Diving Manual.” Paragraph 17-2.5. Table 17-1. Figure 17-1, Note 1. The citation is on the back of the card, in front of you, every time it comes up.
2. A program that checks the citations.
It resolves each one in the manual's own text and confirms the card's numbers appear inside that section. Last run, 31 August 2026: 1,049 of 1,087 units clean. 38 flagged. All 38 of those had already been read against the manual by hand and recorded as correct, with a reason, in a file that ships with the deck. Zero new and unreviewed.
3. A check on should against shall.
Those two words are different obligations in a manual, and a card that swaps one for the other teaches you a decision you are not authorized to make. A program compares every card's wording to the sentence it cites. That sweep found five cards that had quietly softened a requirement into a suggestion. All five were fixed.
4. A gate.
One command runs the schema check, the citation check, the modal check, and both builds, and it fails on any of them. It was run on 31 August 2026, before this page was written, and it printed GATE: green. Nothing ships red, ever.
PASS check (schema + vocabulary) PASS cite (citations, incl. drills) PASS modal (modal strength) PASS deck (.apkg builds) PASS drills (packets build) GATE: green
Underneath all of that: 22 source PDFs, about 4,860 pages. Every one was identified by opening it and reading its own title page, rather than trusting the filename or the link text. Two documents get confused constantly, and the manifest names both mistakes rather than quietly avoiding them. The Salvor's Handbook is S0300-A7-HBK-010, not S0300-A6-MAN-060, and it is Revision 2 from 2021, not the Revision 1 from 2004 that third-party mirrors still serve up as current.
There is one thing none of that machinery can do. It gets its own section further down, and you should read it before you buy.
What is actually in the deck
965 cards. 702 of them carry more than one watchstation tag, because gas laws belong to every station and an MK 16 canister limit belongs to exactly one.
| Watchstation | PQS | Cards |
|---|---|---|
| Air SSDS Diving Supervisor | 307 | 563 |
| Mixed Gas Diving Supervisor | 309 | 534 |
| SCUBA Diving Supervisor | 306 | 528 |
| Chamber Supervisor | 305 | 526 |
| MK 16 Diving Supervisor | 322 | 421 |
| MK 25 Diving Supervisor | 324 | 369 |
| Recompression Chamber Operator | 303 | 366 |
| Inside Tender | 304 | 308 |
| DLSS Reentry Control Supervisor | 313 | 217 |
| SNDL Chamber Operator | 319 | 126 |
| TRCS Operator | 320 | 126 |
Every card carries four kinds of tag at once: which watchstation needs it, what it is about, what kind of recall it demands, and which chapter it came from. So you can cut the deck down to exactly the session you have time for.
deck:"USN Diving Master" tag:qual::chamber-supervisor tag:qual::inside-tender tag:topic::treatment-tables tag:qual::air-supervisor tag:topic::emergency-procedures
By what the card asks of you: 269 recall, 220 numeric, 197 procedure, 195 judgment, 84 table. The 195 judgment cards are the ones that sound like a board question, because they are the ones where knowing the number is not the same as knowing what to do.
Here is one of them, exactly as it appears.
Must you consult an Undersea Medical Officer before starting a recompression treatment?
No. Treatment tables can be initiated without consulting a UMO. A UMO should always be contacted at the earliest possible opportunity, and must be contacted prior to releasing the treated individual.
USN Diving Manual Rev 7 Change A, para 17-1.2 and 17-2.5
Card identifiers are permanent. When an update ships, re-import it and Anki updates the notes in place. Your review history survives, which is the entire reason to build it that way and the reason a re-download from a shared drive costs you months.
That deck is $49, once, updates included. The next section is a different product for a different person, and it is why the second tier exists.
Fourteen drills, for the person who has to run training
The deck is for the diver studying. Drills are for whoever has to stand in front of the locker on a Thursday and run a scenario that holds up.
Each drill is an emergency with a timeline of injects, the actions expected at each one, a worked solution, and grading criteria. Every one of them builds into two packets, a facilitator version and a student version.
Here is the opening of Drill 001, verbatim.
MK 16 diver surfaces with omitted decompression, no chamber on station.
Hull inspection on a moored vessel. Two-diver MK 16 dive at 1.3 ata setpoint, working depth 90 fsw. No recompression chamber on scene. The nearest certified chamber is a 35-minute drive from the pier, and you are 15 minutes from the pier.
T+44. During ascent, Diver 1 reports his primary display has gone blank and he is experiencing a rapid, uncontrolled ascent. He arrives on the surface. His deepest omitted stop is the 30 fsw stop. He is conscious, alert, and reports no symptoms.
The facilitator packet says what the supervisor should establish in the first sixty seconds, which branch of Table 15-7 applies, and the moment the problem stops being a decompression problem and becomes a treatment problem. Every step carries its own citation, same rule as the cards.
Two things you should know before you weigh that. These drills have never been run with real divers, by me or by anyone. They were written against the manual and checked against the manual, and that is the entire extent of it. And the whole of Drill 001, the facilitator packet, is published free, right here, no email required.
Read it before you decide anything else on this page. It is the fastest way to find out whether the work is any good, which is a better basis for spending $129 than my saying it is.
What I cannot prove
This is normally where a sales page stops. Here is the rest of it, because you are going to drill this material until you are sure of it, and you deserve to know the shape of what you are trusting.
No human has read all 965 cards against the manual. The verifier checks that a card's numbers appear near the passage it cites. It cannot check that the card teaches the right thing. A review tool exists for exactly that job and it is built and running. The verdict table is empty. Nobody has started.
A cleared flag is not evidence a card is right. One fix to the verifier widened the span of text it searched by 110.7 percent, and twenty-one flags disappeared. Not one card was validated by that. A bigger haystack finds more numbers by construction. One of those cards had actually been corrected by a person, and the tool took the credit until somebody went and checked. That is written down in the project notes so nobody makes the inference again, and I am not going to make it at you here.
Eighty-two cards put a number in the question. A diver at 80 fsw, a delay of three minutes. Those numbers are not claimed to be in the manual, so no program can ever flag them, and only a reader can tell you whether the scenario sits inside the rule it is testing. A person read all 82 and recorded the pass. That is one person, once.
Nine watchstations have no tag of their own. DLSS Maintenance Technician, DP Diving Supervisor, MK III LWDS, FADS III, Flyaway Mixed Gas Console Operator, MK 16 Maintenance Technician, MK 25 Closed Circuit Technician, Salvage Rigger, Salvage Supervisor. Their material either sits under the nearest tag above or is not in the deck yet.
Nobody has passed a board with this. There are no students, no pass rates, and no quotes on this page, because there is nothing true to put in them. You are the first.
If any of that disqualifies it, do not buy it. Losing the sale costs me less than having you drill a card I oversold.
Here is what you can do about all of it, though, and it is the reason the citations are on the front of the product instead of buried in a credits file. Every card names its paragraph. Pick any card, open the manual to that paragraph, and you will know inside about ten seconds whether the card is right. Nothing else in this category lets you do that, and the free drill above lets you do it before you pay anything.
What a week of this looks like
Import the file, then build one filtered deck for the watchstation you are actually sitting for. How many cards a day that turns into is your call, because it is your Anki settings and not mine.
Within a few days the recall cards start to settle and the numeric ones do not. That is normal, and it is also information: those are the facts you were going to blank on.
Then a card surprises you. You were sure of the answer, and the back of the card says something else. The citation is sitting right there underneath it, so you open the manual to that paragraph and find out which of you is wrong.
That loop is the actual product, and the cards are only what starts it.
Who this is not for
If you are a recreational or a commercial diver, this is not your deck, and it is better said here than after your money has changed hands.
Navy Diving Manual material circulates widely outside the Navy because it is the most detailed diving text anybody can download for free. But this deck teaches Navy procedure, Navy tables, and Navy obligations: who may modify a treatment table, what a Chamber Supervisor is required to do, which manning level applies. Studying that as a rec diver teaches you rules that do not govern you and answers you cannot act on.
Buy the manual's own free PDF instead. It is genuinely great reading.
$49 for the deck. $129 with the drills.
SINGLE DIVER · $49, ONCE
All 965 cards, all eleven watchstation tags, and every future revision free forever, with your review history surviving each one.
DIVE LOCKER · $129, ONCE
Everything above, plus all fourteen drills in both packets, licensed to run at one command's locker. This is the version for the Master Diver or the training petty officer who is buying it for other people.
No subscription, no login, and no app to install other than Anki, which is free.
What lands in your inbox, exactly:
usn-diving-master.apkg, about 1.1 MB, the file you import into Anki- The source manifest: all 22 documents, with revision, date, and where each came from
- The adjudication log: every card a program flagged, and the reason a person cleared it
- Dive locker only: 28 markdown packets, a facilitator and a student version of all fourteen drills, plus an index
No installer, no account, no DRM. It is a file.
The guarantee
Full refund, any time, no questions, no form.
A guarantee on a file is a strange thing, because you keep the file either way and we both know it. So here is the honest version: if you read the section above about what I cannot prove and then decide it was not worth $49, tell me and I will send the money back. I would rather have that email than your silence.
Questions
Is this official?
No. It is not a NAVSEA product, not a SUPSALV product, and not approved by anybody. The source manuals are public releases. What was built out of them is not endorsed and does not claim to be. Your PQS still signs off in a book, not in Anki.
The manuals are free. Is it legal to sell something built from them?
Yes. Every source carries Distribution Statement A, approved for public release with distribution unlimited, and every one was pulled from an official public host. The manifest lists all 22 of them with revision and date. Nothing came from a CAC-gated system, and nothing ever will. That is a standing rule on the project.
Will it work in the app I already have?
Only if that app is Anki: the desktop app, AnkiMobile, AnkiDroid, or AnkiWeb. Those four are the whole official family. AnkiApp, which now calls itself AlgoApp, is a different unaffiliated product that traded on the name, and this file will not import there. Neither will Anki Pro or Noji. Check which one is on your phone before you buy, because it is an easy mistake and an annoying one.
What happens when Change B or Rev 8 comes out?
A scheduled check watches SUPSALV every quarter for a new change, a new revision, and a newer ANU list. First run is 1 November 2026. When something lands, affected cards get rewritten against the new text and the update goes out to everyone who bought it. Because card identifiers never change, you re-import and keep your history.
Does it cover my watchstation?
Eleven of them, with the counts in the table above. Nine more have no tag of their own and are named in the section on what I cannot prove. If yours is on the second list, wait.
Can I run the drills with my whole locker?
That is what the $129 tier is for. One command, one locker, run them as often as you want.
What about salvage and seamanship?
78 cards, drawn from the Salvor's Handbook Revision 2. The Ship Salvage Manual volumes and the Towing Manual are in the source set and mostly not mined into cards yet. Treat salvage coverage as thin and everything else as the priority it was.
One more time
You are going to spend the next several weeks making yourself certain about a few hundred facts. That part is happening either way.
The only question on this page is whether the facts you make yourself certain about can be traced back to a paragraph in a current manual.
965 cards, with a citation on every one. A program checks those citations, a written record holds everything it flagged, and a printed list says what none of it proves.
P.S. Go find the line-pull card in whatever you are studying now. If it does not say which table it came from, you have a fifty-fifty chance on seven pulls and no way to find out which side you are on.