There was no great submarine stagnation

by on April 30, 2017 at 3:26 pm in History, Political Science, Science | Permalink

Recall the development of the Polaris nuclear-missile system in the late 1950s. The whole package—a nuclear submarine, a solid-fuel missile, an underwater launch system, a nuclear warhead and a guidance system—went from the drawing board to deployment in four years (and using slide rules).

Today, according to the Defense Business Board, the average development timeline for much less complex weapons is 22.5 years. A case in point is the Ford-class aircraft carrier. The program is two years delayed and $2.4 billion over budget.

That is from John Lehman, Secretary of the Navy under President Reagan.

1 wophugus April 30, 2017 at 3:33 pm

In what possible sense is the Ford-class carrier less complex than a Polaris sub.


2 Austin April 30, 2017 at 3:52 pm

I have to agree. Common sense seems to be left behind in that comparison. Anytime I read a statement like that which clearly disregards logic, I am forced to stop and ask “What is this author’s agenda?” It’s disappointing when intelligent people put an agenda before clear thinking and complete information. It’s just too hard to find good information anywhere these days.


3 Thiago Ribeiro April 30, 2017 at 4:04 pm

“Less complex” may not be the right way to put it, but I wonder if the Polaris system package wasn’t a bigger step forward than the Ford-class aircraft carrier is. I admit I have no idea.


4 Viking1 April 30, 2017 at 4:12 pm

Let’s see:

Nuclear launch system that can not easily be defended against: Big step forward. The greatly extended [submerged] range of a nuclear sub compared to a diesel electric means detection is much harder.

Ford aircraft carrier: Probably invulnerable to any conventional treat, but still can be destroyed by an ICBM. Same as previous generation.

Definitely +1, perhaps +10. Paraguay will be proud of you.


5 Thiago Ribeiro April 30, 2017 at 4:50 pm

Then, I may be right that the Polaris system was a bigger advance from the previously existing product.
The Paraguayan aggressor was a stubborn and worthy adversary even if it was not honorable.


6 mikea April 30, 2017 at 7:44 pm

You’re crazy if you think that a carrier is invulnerable to conventional threat. There are a lot of Russian and Chinese missiles right now which cost a fraction of the $15 billion dollar carrier that are specifically built to sink a U.S. carrier. Even if the carrier group can effectively stop 90 or 95% of these missiles, they are so cheap that firing dozens or scores is trivial for both China and Russia.

The effect is that in any shooting war American carriers will be kept thousands of miles from Russian or Chinese shores. This is also why China is building islands in the South China Sea. Because a few acres of land allows them to prevent American carrier access over thousands of square miles of ocean.

The US is building weapons for fighting the last war at great expense.


7 Viking1 May 1, 2017 at 12:46 pm

Don’t know if anybody is listening still, the key is that a carrier killing missile needs to be fast enough that existing missile defense systems fail. Agree on your points, also your linked article mentions the submarine threat.

My main point was that it is unlikely the next generation carrier is any better protected than the previous generation, whereas the polaris system was a substantial step forward.

Another point that is underappreciated in cold war history, is that a preemptive nuclear attack on the soviet union anytime until 1960 probably could take place without any soviet nukes reaching the American mainland.

8 mikea April 30, 2017 at 7:48 pm
9 TheAngryPhilosopher April 30, 2017 at 4:34 pm

In several ways, actually:

1. It’s less complex relative to the tools we have available to design it. We can simulate and plan and test all sorts of things with computers, while in the 1950s, as mentioned, they had to do everything by hand.

2. It’s less complex relative to existing hardware. The USS Gerald R. Ford, while frighteningly complex from any absolute standpoint, is basically a revamped and modernized Nimitz-class carrier. The missile subs had no obvious predecessors and many of the systems developed for it were the first of their kind (for the US). Designing from scratch is much more difficult than designing from an established predecessor.

It’s still a little silly to describe the Ford-class carriers as being broadly “less complex” than the submarines, but I think I see what Lehman was going for. I also think that his broad point is still valid. The state of the US military-industrial complex is in fact worse than it was in the 1950s – the carriers would be better, cheaper, and delivered faster if the institutions of the 1950s had designed them (assuming they had access to current tools and knowledge of course).


10 Thiago Ribeiro April 30, 2017 at 4:47 pm

What do you think has changed?


11 Andao April 30, 2017 at 10:08 pm

I think it’s safety. And/or redundant systems. A good analogue might be commercial aircraft, they crash a lot less often and can run on autopilot pretty much from start to finish. They also burn less fuel but that’s a moot point when you’ve got a nuclear reactor.

Perhaps the dollar value on a sailor’s life has increased dramatically since the 50s? So the expensive foolproof systems are worth it based on some accounting?

The aircraft carriers are still stupid. Like a floating bullseye that’s too expensive to actually use in battle.


12 Thiago Ribeiro May 1, 2017 at 5:01 am


13 Ray Lopez April 30, 2017 at 4:53 pm

Interesting. More evidence that we need to increase the prize via better patent laws.

Bonus trivia: Nuclear submarines can be detected using quantum gravity detectors that are so sensitive they can detect a heavy metal pipe thirty feet underground from the slight gravity change induced by iron over soil.

Bonus trivia II: patents were so disfavored historically that despite iron working discovered since 1000 BCE, the spiral spring was not invented until the 15th century, and the leaf spring was not invented until the 17th century. Some say medieval guilds (akin to modern vested interest corporations) held back inventions. Arabic numbers were not adopted in accounting until several hundred years they were known in Europe and 150 years after European treatises were written on how to use them. Source: M. Postan, Cambridge historian.


14 carlospln April 30, 2017 at 5:26 pm

You’re referring to magnetic anomaly detection, not ‘quantum gravity’:


15 Ray Lopez April 30, 2017 at 8:52 pm

Thanks carlospin. No, I am aware of magnetic anomaly detectors but I was referring to a speculative top-secret project the US Pentagon may be working on, as speculated by the Economist, involving quantum mechanics (entangled photon interferometer). Hence “can be” not “are detected”.

16 Alistair April 30, 2017 at 9:37 pm

Gravity-based detectors with this tech are certainly coming. The current lab work is fascinating; I just didn’t think they had quite this magnitude of resolution or were ready for field deployment. Give it another decade (or two) and it might be possible.

17 Pete April 30, 2017 at 7:33 pm

Submarines are always kept within a fraction of a percent of neutrally buoyant. I am very skeptical that gravity would ever be a good way to detect them.


18 Sure April 30, 2017 at 8:56 pm

Actually, having to deal with pre-existing infrastructure may well be a net detriment on time concerns.

Because you will share a significant number of components with other users you get a discount on price, on the other hand your procurement needs may be deferred for some other project. Having to share certain supply chains with the Nimitz class means that you may have weight behind a ship that needs part of production run. This can lead to exponential delays. When you are the sole user of some widget, you just need to make the widget.

Likewise, designing things to pre-existing specs can add massive amounts of time to design. If you are designing some component from the ground up, then you basically say what performance you want, price it, and go. If you have to use a pre-existing part you cannot in the middle of the design process just change (slightly) its performance (e.g. add a bit more carbon to the steel). Again you save money through this sort of constraint, but you often will need to jigger around far more design work than just being able to change your component specs on the fly.

Polaris cost a much higher percentage of GDP for a comparable build out. While I doubt it is definitive, having a whole ecosystem in place may well offset a lot of the time savings you might expect merely from “updating” a design.


19 prior_test2 May 1, 2017 at 3:37 am

‘is basically a revamped and modernized Nimitz-class carrier’

Including the revamping of the entire aircraft launching system –

And since the Ford class generates a lot more electric power, it is quite possible that the Navy will be able to use systems along these lines – Not as a long range weapon particularly, but imagine a carrier that is able to put out multiple thousands of what one hopes will turn out to be at least one golden BB in terms of missile defense at a rate that ensures 100,000s of such pellets being in the path of incoming warheads or aircraft.


20 Viking1 April 30, 2017 at 4:07 pm

Perhaps the comparison is apples and oranges, but there is no doubt that we get way less bridges for the money after adjusting for inflation, and the four years to develop the polaris system using slide rules is impressive, a decade to replace the world trade center, and multiple decades for the F-22 and F-35 programs is not impressive.

In a sense, we are venturing into a bureaucratically driven mini dark age, especially with respect to energy.


21 Amigo April 30, 2017 at 6:00 pm

For thoughts on this, it’s interesting to read the book on Elon Musk by Ashlee Vance on bringing cost out of rocket launches at Space X. One detail that’s emphasized is bringing down cost and increasing developmental speed required going in house for much of the development, including making many of their own parts that were readily available on the market at much higher prices. Vertical integration had huge benefits in terms of price, logistics, and speed of development. I don’t think Musk has interest in these military type programs, but it’d be interesting to see what a skunk works today based on the way he built space-X would be able to produce.


22 Albigensian May 1, 2017 at 10:46 am

And here I was thinking that vertical integration went out of fashion with the death of the first Henry Ford?

In any case, Polaris’ engineers presumably had access to digital computers (such as the IBM 700/704/709) as well as slide rules. The real change in weapons systems is the huge role of electronics (hardware and software) in the systems themselves.

Software development, and the development of custom electronics hardware, surely is subject to cost disease as the explosion in what’s possible has produced vast increases in the complexity of weapons systems even as security-driven compartmentalism (e.g., developers are permitted to know only as much about a system as they need to know) hobbles productivity.

It seems to be a great paradox that programmability, which one would think would make a system more flexible and easier to modify, seems to have done just the opposite.

Yet cost-and-schedule bloat seems to afflict all big projects (and especially public projects); thus, it’s not just military weapons systems that suffer schedule and cost bloat but more prosaic big-project technologies, such as New York City’s Second Avenue subway.

Perhaps the true singularity won’t be some fantastic future but just a “machine-runs-down” apocalypse in which we become unable to build anything for any purpose within any imaginable time and budget constraints?


23 purple water fountain April 30, 2017 at 6:05 pm

Multitudes of internal & external DOD critics have been loudly moaning about the outrageous time/money acquisition costs of Defense systems… for the past half century.
The critics are correct. The problem causes are well known, but the situation steadily worsens.

Primarily, the Pentagon lacks personal or financial accountability for its acquisition excesses. Military fantasy wish-lists become hard requirements and are routinely embellished during the development cycle… when they should be stable. The generals/admirals/SES pay no personal penalty for failure or vast waste of resources (except perhaps for a comfortable early retirement). The taxpayers are endlessly fleeced. The Pentagon has no effective accounting system overall, has never passed an independent audit, and routinely can not account for billions/trillions of Dollars.

The fault ultimately resides with a dysfunctional Congress, which now cannot even perform its most basic duty of passing an annual budget… or formally declaring war when the Federal Government takes the nation into very real war(s).


24 mavery April 30, 2017 at 4:55 pm

Cost disease is a real thing, and it is clearly worth looking into w.r.t. DoD acquisitions, but this is an awful comparison.


25 DK April 30, 2017 at 5:06 pm

Construction of the Empire State Building took 15 months from A to Z. More than 80 years later, an ordinary office building takes as much. That, and the fact that we no longer possess an ability to bring a man to the Moon. But we have smartphones and disparate impact laws, so that balances thing out – right?


26 Nicholas Marsh April 30, 2017 at 5:14 pm

Comparing a ship or aircraft with one from the 1960s is a bit like comparing a typewriter to a word processor. They are both used for a similar task but the modern machine can do far more. Likewise the electronic equipment in modern military ships and aircraft accounts for most of the reason why they cost so much more than in the 1960s.


27 Cliff April 30, 2017 at 5:22 pm

Sorry. Any evidence?


28 A Definite Beta Guy April 30, 2017 at 7:58 pm

Take a look at your own household appliances.


29 BC April 30, 2017 at 8:11 pm

Are household appliances more expensive now and do they take longer to bring to market than in the past?


30 Pete May 1, 2017 at 10:08 am

Trident D5, the modern SLBM, has a range of 4000 nautical miles, the Polaris was 1200. Take a look at the size of the earth and you’ll see that this means way less submarines must be made, manned, and deployed to achieve the same effect.


31 Steve Sailer April 30, 2017 at 8:25 pm

You could compare aircraft from the 1950s-1960s in terms of maximum speed and altitude to today’s versions. Not really all that much has changed.

Lockheed Skunk Works spy planes were famously fast to be developed and delivered.

The XF-104 reached Mach 1.8 in 1955.

The U-2 spy plane was remarkably fast in development: “Lockheed received a $22.5 million contract in March 1955 for the first 20 aircraft, with the first $1.26 million mailed to Johnson’s home in February 1955 to keep work going during negotiations. The company agreed to deliver the first aircraft by July of that year and the last by November 1956. It did so, and for $3.5 million under budget.[13]”

The A-12 / SR-71 recon jet was flying 2000 mph by 1963, that 54 years ago. It was insanely expensive to operate, but it was fast.

The F-104 and SR-71 had problems that made them troublesome in regular use, but the U-2 was an extremely successful design that is still more or less in service.


32 Adrian Turcu May 1, 2017 at 3:56 pm

“You could compare aircraft from the 1950s-1960s in terms of maximum speed and altitude to today’s versions. Not really all that much has changed. ” -Military aircraft are not built just to fly high and fast: the main objective is to detect and destroy enemy aircraft and land targets. Today’s aircraft are an order of magnitude more capable in doing that. An F-22 can detect multiple targets at 400km distance and engage them at 150km. An F-89 had to get 10-12km close to a Russian bomber to be able to fire. An F-104 could engage at maybe 30km.


33 adam May 2, 2017 at 2:08 pm

Do I hear you volunteering to take a 1960s-era fighter into a live fire dog fight with an F-22 or F-35? Better make sure you have your will drafted before you head up. You’d be in a ball of flames before you even saw the F-22.


34 dearieme April 30, 2017 at 5:22 pm

Proposition: from the time of Wilson (if not before) the history of the USA is a tale of the federal government being run more and more in the interests of the big corporations.

It explains why the USA was prepared to provoke Japan to war: her corporations wanted to be able to trade into China in a way that Japan’s invasion would impede. It explains why the USA prioritised the war with Germany (once Hitler had declared war on the US): the corporations didn’t like the idea of not being able to sell to Europe, from which they would be excluded had Hitler won. And now: need I say any more than F-35?


35 So Much For Subtlety April 30, 2017 at 6:40 pm

Those corporations would have lost more money in the war than the trade with China was worth. I doubt there is any evidence of any conflict being caused by, or even influenced by, a major corporation in modern times. Not even the Guatemalan coup.

It is more likely that significant lobbies in America – mainly on the Left – were concerned about Europe. Roosevelt wanted to find a way to get Congress to support the war. He prodded the Japanese until they provided. Notice I am not saying it is true, just more likely.


36 Troll Me April 30, 2017 at 7:25 pm

Iraq? The VP had previously sat on the board of a major arms firm, and stood to profit greatly from their profit.

I’m not saying he committed crimes. I’m saying that what was plain as day in front of our face was plain as day in front of our face. (The VP had been on the board and was close with one of the main corporate beneficiaries of various Iraq-related contracts.)

Also, the bar you set of not even having influenced conflict is too high a bar to pass. You’d be hard pressed to mention any conflict whatsoever where no corporation even “influenced” it. As for being the primary instigator of cause, that’s going too far.

What is known, however, is that there are shareholders of arms firms who like to make money. And if people behave in the way that econ 101 predicts them to, they might just like to tilt things a little more in the direction of buying more bullets.


37 So Much For Subtlety May 1, 2017 at 3:49 am

Well, pretty much what I expected from Nathan. Halliburton was not and is not a weapons company. It did get some Iraq-related contracts. So what? Again they probably cost America and hence Halliburton more than the contracts were worth.

You would have to demonstrate any influence. I doubt there is any. Leftists have been allegeding for a long time so they have not really been looking. Still, government papers are fairly public. You can look up, for instance, the talks around Vietnam or Korea. I haven’t heard of anyone finding a single corporate fingerprint on any of them.


38 Troll Me May 1, 2017 at 12:16 pm

The VP had been on the board.

No influence? Please don’t call us dumb. But was it unreasonable in the situation? He would obviously have been predisposed towards the course of action of promoted for the fact of his previous position(s).

39 carlospln April 30, 2017 at 5:24 pm

“They are both used for a similar task but the modern machine can do far more”

They do the same bloody thing-a platform for projecting air power.


40 CMOT April 30, 2017 at 5:29 pm

Davy Jones’ Cost Disease?

Neptune’s Cost Disease?

This needs a catchy name.


41 Michael Cain April 30, 2017 at 5:43 pm

How many more tens of millions of lines of software run on the new carriers compared to the Polaris subs? The costs and schedules for DOD’s big new toys are dominated by the software these days. Not just DOD. The IEEE ran an interesting piece a few years back about GM’s development of a new transmission for a hybrid SUV. Two-thirds of the engineer-hours for the transmission were writing and debugging the software that ran on the embedded processor.


42 So Much For Subtlety April 30, 2017 at 7:01 pm

Sure but how many hours of engineering time was spent drawing blueprints and working with those slide rules? Computers have made a lot of tasks even easier.

Still America should count itself lucky. Just 22 years for an aircraft carrier? Brazil launched its nuclear submarine program in 1978. It launched its first nuclear submarine ….. well they are still working on it.


43 BC April 30, 2017 at 8:27 pm

Did the transmission development take 5.5 times longer than similar development in the past? I would think that software-based development would actually be faster than hardware and physical/mechanical development.


44 egl April 30, 2017 at 7:33 pm

The Ford-class carrier couldn’t have been built in the 1950’s, so the comparison is as useful as comparing baseball teams of different eras. If you want to spend time thinking about high-end military procurement, look at F-15 vs F-16 or F-22 vs F-35. You can also look at program cancellation and survivorship bias in estimating program development times.


45 Steve Sailer April 30, 2017 at 8:42 pm

The really prodigious military hardware development was done either during wartime or under threat of nuclear war during the early Cold War (e.g., the Moon Landing was sort of a peaceful demonstration of nuclear war fighting capabilities).

My father’s more talented friends, for example, designed the Mach 2 F-104 in the mid-1950s as a more or less kamikaze interceptor to get from ground level to the stratosphere unbelievably fast to shoot down Soviet nuclear bombers before they reached Seattle.

My father then spent much of the 1960s trying to make the F-104 safe enough for pilots of NATO allies to use as their all-purpose fighter-bomber, which it was never intended to be in the first place.

I would imagine Lockheed devoted far fewer more man-hours to the original design of the F-104 than to later trying to redesign it to stop killing so many West German pilots.


46 Steve Sailer April 30, 2017 at 9:03 pm

“(and using slide rules)”

My father, a Lockheed engineer, broke down and bought himself an electro-magnetic adding machine around 1969 and an electronic calculator around 1973.

But, it’s worth mentioning that Lockheed itself bought its first IBM mainframe, of the 700/7000 family, in 1953 or 1954. I’ve seen a list of IBM’s first customers for this computer and it’s pretty clear that they were being allocated in order of importance to developing America’s nuclear war-fighting capability. For example, Lockheed got the 18th 701 ever built.

So, the hardest computations done by defense contractors were being given to computers by the mid-1950s, although individual engineers continued to use slide rules into the 1970s.


47 Steve Sailer April 30, 2017 at 9:10 pm

Okay, here’s IBM’s list of the first 19 installation of its 701 mainframe. You can see that national defense (especially nuclear war division) was dominant in determining who had priority: UC Berkeley got 3 of the first 19, with #2 and #13 going to Los Alamos and #16 going to the bomb shop at Lawrence Livermore labs. Lockheed got #3 and #18.

701 Customers

Machine number Shipped to Date Note
1 IBM World Headquarters, New York, N.Y. Dec. 20, 1952
2 University of California., Los Alamos, N.M. Mar. 23, 1953 (a)
3 Lockheed Aircraft Company, Glendale, Cal. Apr. 24, 1953 (b)
4 National Security Agency, Washington, D.C. Apr. 28, 1953
5 Douglas Aircraft Company, Santa Monica, Cal. May 20, 1953 (c)
6 General Electric Company., Lockland, Ohio May 27, 1953
7 Convair, Fort Worth, Tex. Jul. 22, 1953
8 U.S. Navy, Inyokern, Cal. Aug. 27, 1953 (d)
9 United Aircraft, East Hartford, Conn. Sep. 18, 1953
10 North American Aviation, Santa Monica, Cal. Oct. 9, 1953 (e)
11 Rand Corporation., Santa Monica, Cal. Oct. 30, 1953 (f)
12 Boeing Corporation, Seattle, Wash. Nov. 20, 1953 (g)
13 University of California, Los Alamos, N.M. Dec. 19, 1953
14 Douglas Aircraft Company, El Segundo, Cal. Jan. 8, 1954 (h)
15 Naval Aviation Supply, Philadelphia, Pa. Feb. 19, 1954
16 University of California, Livermore, Cal. Apr. 9, 1954
17 General Motors Corporation, Detroit, Mich. Apr. 23, 1954
18 Lockheed Aircraft Company, Glendale, Cal. Jun. 30, 1954 (b)
19 U.S. Weather Bureau, Washington, D.C. Feb. 28, 1955 (i)

(a) Used for hydrodynamics calculations.
(b) Mathematics Analysis Department used the 701 for problems in aircraft design, such as aerodynamic performance and stability, thermal dynamics and structural and flight dynamics. Production data handled on 701s by Lockheed’s Factory Data Processing Group included project base schedule preparation, parts scheduling, shop order writing, direct labor hour forecasting and parts activity ledgers.
(c) Arrived on May 23, 1953, aboard a DC-6A aircraft. The 701 was used to get the DC-7 into production months ahead of schedule. It solved engineering and scientific problems on all Douglas commercial aircraft, including the DC-6B, DC-7, DC-7C and the development of DC-8.
(d) Used to calculate rocket and missile performance and to simulate flight conditions of these devices at the U.S. Naval Ordnance Test Station China Lake.
(e) Handled engineering problems, from basic configuration selection through aerodynamic and structural design to the analysis of flight test data.
(f) Used to solve wide variety of problems in economics, mathematics, aircraft, missiles, electronics, nuclear energy and social sciences. Later moved to West Los Angeles.
(g) Used to assist engineers and designers in solving problems in aerodynamics, stress and structural development, and flight testing of supersonic and jet aircraft and guided missiles.
(h) Solved engineering problems on U.S. Navy A3D Skywarrior, A4D Skyhawk and F4D Skyray programs, and USAF C-133 and RB-66 programs.
(i) Produced from spare parts.


48 Anooj Pakvasa April 30, 2017 at 9:14 pm

9 11-1=1950


49 April 30, 2017 at 9:40 pm

For the submarine, mechanical and electronic silence is very important. For the carrier?


50 buddyglass April 30, 2017 at 10:15 pm

Yeah, but in the 50s they were just repurposing alien technology from crashed UFOs. These days we have to come up with stuff on our own. Whole other ball game.


51 Amigo May 1, 2017 at 12:27 am

Caught me off guard. LOL.


52 Andao April 30, 2017 at 10:26 pm

I would guess the bulk of these advances come from making the weapon systems more automated, less prone to failure, less prone to user error, and less likely to kill the operator. All of these are boring upgrades and might not be worth the cost, but there’s probably some Pentagon accountant who thinks the numbers add up.

Planes and cars haven’t gotten any faster in decades, but they are much safer and easier to use. Still, this doesn’t explain stupid strategic decisions, like investing so much in fighter jets when drones and guided missiles can effectively replace them, or these aircraft carriers that will never be used in active combat because they’re too expensive.


53 Steve-O May 1, 2017 at 9:59 am

The median car has gotten much faster, hasn’t it?


54 Patrick Laske May 1, 2017 at 12:22 am

An apples to apples comparison would be the Midway class carrier and the America Class “Amphibious Assault Ship”. Both ships are about the same size, though obviously the America class has a fourth the crew of an old Midway ship (half if you count the marines). Both are basically upgrades to previous models, the Essex and the Wasp class respectively. The Midway took 5-6 years, under wartime conditions, to get out the door from initial talk. It took about 3 years to get to laying down the first ship. While initial talk for the America class began in 2001, real work didn’t begin until 2006. The first ship was laid down in 2008, and while it was delayed a couple years, due partly to political factors like the sequester, the first ship launched in 2014.

“We’re going to make these” to “Laid Down” on Apples to Apples carriers, you’re looking at about a 1/3rd time reduction even in ‘peace time’. Ship construction took a few more years, but US Shipbuilding capacity is considerably different than what it was in 1944. The great shipyards of New York City, for example, are now hipster apartments and Wegmans.

The Ford class is an experimental platform, designed to showcase new technologies, such as EMALS, which launches planes using electromagnets. Sure the polaris missiles were experiments as well, but those were partly based on the Jupiter Missiles that preceded them and the George Washington class were modified Skipjack class attack submarines. The real advances where the computers and the physics and math that went into programming them, along with fantastic advances in miniaturization. The Ford Class is designed to be an experimental platform, so it can do 25% more sorties with 25% less crew. Unlike the Nimitz, it is designed to be future proof, with much more powerful reactors and space for future weapons and systems. It could last up to 90 years, and will save billions of dollars in ongoing costs (which is the bulk of the cost of a carrier, not the design and construction), and be simpler and easier to maintain. There were mistakes made in constructing and designing it, but I’m not sure how long such a platform should take to design or build. I have a hunch, based on converting the savings of the Ford Class into an annuity, but I won’t embarrass myself in posting a number.

There actually are places where the US military is making mistakes in ship design and building, the LCS and now FCS. The US is spending too much on the small stuff, taking too long to do the small stuff, and getting too little return from it. If you really want to get angry sometime, look up the GPS upgrade. Big stuff has too many eyeballs on it for it to get too far out of hand.


55 Thanatos Savehn May 1, 2017 at 12:50 am

NASA became a jobs program long ago. Most “research” sponsored by our government is just a p-hacking jobs program for the otherwise unemployable. Increasingly Defense is becoming just another dig-holes-and-fill-them-back-in jobs program. It’s actually a very old problem. When the poop hits the fan next time we’ll probably just hire mercenaries – and they’ll look like Brin and Musk. The interesting question is “who’ll play the role of the Visigoths in this our unfolding tragedy?”


56 A.G.McDowell May 1, 2017 at 12:50 am

At least the last two chapters of “Skunk Works” by Rich and Janos contain laments about growing government bureaucracy, with some acknowledgement that this was introduced at least partly due to worries about cost overruns and security. (Economists might enjoy setting up models to flesh out his words about wishing to rely on reputation instead).

Attempts to replicate Skunks Works development success in computer projects was one of the streams feeding into the Agile movement, which attempts to increase development speed in computer project development, for instance by substituting test use of working prototypes for reading and writing large volumes of detailed documentation. The success of this has led to principles being published at Attempts to introduce this to large bureaucratic organizations such as governments has led to


57 Jon May 1, 2017 at 1:18 am

Moore’s law is often viewed purely as a cornucopia, but it also means engineered complexity has been growing exponentially.

Imagine a city with each street as a wire in a computer chip (200m between blocks). In 1963, designing a computer chip was similar to designing a city 4sqkm in size. In 2000, this task was similar to designing a city that covered the entire Earth. In 2010, this task was similar to designing a city that covered all of Jupiter.

Engineering cost has been been much less than doubling every 18 months. This is a productivity miracle.

Unfortunately it is a miracle that for example gets used so that when you tilt your iPhone, you see a parallax effect.


58 Steve Sailer May 1, 2017 at 4:44 am

I’m interested in what the Oroville Dam spillway rebuild is going to turn out like.

They built a new spillway from scratch nearby at the Folsom dam and it’s just now being finished after about a decade of work.

They really need to get Oroville fixed up faster than that. It will be interesting to see how fast they can do it and what the costs of going faster will turn out to be.


59 David Khoo May 1, 2017 at 6:44 am

This is a truly egregious case of cherry picking and survivorship bias. If you have the degrees of freedom to pick one of the luckiest and most successful programs of the past and compare it to a troubled program of today, you can come to any conclusion you want.


60 The Engineer May 1, 2017 at 8:57 am

The Polaris guidance system development was probably the most difficult development. Control systems engineering was in its infancy then, it was literally developed right alongside these guidance systems. And, yes, computers themselves were developed to be able to develop these guidance systems. Most of the early computers were used in the SAGE system, the continental wide radar tracking system guarding against Russian nuclear bombers.


61 Type III Error May 1, 2017 at 10:19 am

Office of Naval Research already has a tried and tested frigate type ship (Sea Fighter (FSF 1) Fast Sea Frame that cost only $250M, half of an LCS. How nobody ever brings this up boggles my mind.


62 John Mansfield May 1, 2017 at 11:29 am

The Ford-class aircraft carrier is a replacement for already existing carriers. It’s future need was known with decades of anticipation based on the expected life of the existing carriers, so design work was methodically initiated decades ago, because it could be and should be.


63 Donald Pretari May 1, 2017 at 2:04 pm

How much damage could the nuclear missiles on one U.S. sub do to North Korea? There are 1,100 or so cities and towns in Russia. We once had 30,000 nuclear warheads. We could have targeted down to chicken coops in those days. What would happen to the rest of the planet if we simply set off 4000 warheads here? Liberal Puget Sound is basically one large military base. Do you see anybody local protesting that? We’re using missiles to target individuals now.

The military budget is the mother of all boondoggles. Nothing rational explains it except crony capitalism/government and the irrational fears of U.S. citizens regarding our destruction. Do not bother to try and make sense of it based upon any serious criteria.


64 Rafael R May 4, 2017 at 7:07 pm

Now that’s an example why humans respond to incentives: in the 1950’s with the Cold War tensions were sky high and development was super fast, otherwise the USSR would be able to open a gap.

Now, the US’s military leadership is not disputed (China is not even remotely interested at the present in engaging in any sort of arms race) nobody has incentives to improve their weapons. Ford, for instance, is just a marginal improvement over existing designs so why rush it into deployment? It’s marginal gain is small.


Leave a Comment

Previous post:

Next post: