It can't...
If the Pentagon tries to go with try/fail/learn/try again like SpaceX, then they'll get roasted.
It's like NASA. They're under the Congressional microscope with Congressional micromanagement and "failure is not an option" literally.
The outgoing first chief architect officer of the US Air and Space Force urged the Pentagon to lay off wasting time building everything itself, and use commercial kit if available and appropriate to upgrade its technological capabilities quickly. "If the Department of Defense is going to compete on today's global stage and …
as Henry Ford said, "you can have any color you want, as long as it is black" (with reference to the model T).
Or he said something very similar to that. It made sense when he was trying to significantly lower the average cost of ownership, which was his primary goal at the time.
keep in mind that the Pentagon needs to consider a few things:
* Foreign made equipment is not under their control
* Foreign made equipment can have design features NOT conducive to national security (back doors, kill switches, things the CIA has ALREADY [allegedly] done to others I might add)
* If a war breaks out and they need MORE of this stuff, an embargo would SERIOUSLY screw things up [a single EMP might make that a necessity]
and stuff like that. "Not Invented Here" is a BAD thing for national defense.
A website I'm in the habit of reading every now and then, Tomdispatch, reminds us that the Pentagon faces no penalties for failures such as the War in Afghanistan, Iraq, etc. And indeed, likewise their suppliers face no penalties for being a decade behind, and a few billion over budget. Eg, Lockheed's Technical Learnings of Lockheed for Make Benefit Glorious Nation of America aka the F-35; the US Navy's Littoral Combat Ship
https://www.forbes.com/sites/davidaxe/2021/05/20/the-littoral-combat-ship-cant-fight-the-us-navy-is-finally-coming-to-terms-with-it/?sh=1202da452587
and other Technical Learnings for Make Benefit Glorious Nation of America aka overbudget monstrosities. What's a few billion between a congresscritter and his/her purchaser? At least said congresscritter isn't expected to pose for sale naked on the block like slaves fresh from Africa had to, back in the day. (A pity - a few more purchasers' suicides from the sheer Lovecraftian horror of the experience would do wonders to keep the rest in check.)
I suspect the entire problem is to do with the "easy money" they're getting in spite of lack of performance. And the resulting sense of entitlement.
eh, your list is less than satisfactory since OSs like Linux, being open source, can be forked and maintained domestically.
However, _I_ would focus on the HARDWARE. Think of this: how MUCH of a typical computer is MADE FROM OVERSEAS MANUFACTURED COMPONENTS?
And THAT is the crux of the problem, as I see it. In a war, your "friend" could become your ENEMY, and with the CCP [consider their current posturing] and the Russia/Ukraine war, this is even MORE evident.
FreeBSD (and the other BSDs) were at least FORKED from things invented in the USA. I thought I might just mention that. And, again, open source makes it possible for DoD to maintain it themselves.
Millions of LOC are almost impossible to secure. seL4 is like a frigate with 20 compartments - a single torpedo can only take out one compartment, but the ship stays afloat.
For example, a bug in the TCP stack of an seL4 system does not pwn the entire system. Your data stays secure, as it is decrypted in a different compartment. With Linux, it is Game Over.
Windows is even worse, as they run a c**pload of stuff in the Kernel, including font parsing+rendering, graphics rendering, several network stacks, a network file system and whatnot.
Likewise, Memory Safe Languages will contain the effects of programming errors in about 70% of cases. With C or C++, a single programming error typically hands the kingdom to the intruder. See http://sappeur.ddnss.de/Sappeur_Cyber_Security.pdf
Enough already!
Go peddle your annoying panacea du jour garbage somewhere else, goddamit! By now, we're all aware you're a paid troll, and are becoming as annoying as Google "targeted ads".
Bugger off. Maybe there will be more fertile grounds for you to farm in a Reddit forum or something.
You are correct that hardware sourcing is the real issue here.
Having designed quite a lot of equipment for fast jets (among other things) I can say that most modern equipment can use industrial / automotive grade parts; the key is at the PCB / box level to ensure that the bits are properly thermally managed.
In a safety critical design (I can think of quite a few) the designer has to show that a single component failure will not cause a failure of the system (there are actually metrics for this) but a very important consideration is that complex parts (microprocessors, microcontrollers, memory devices and so on) must have a track record, with no new errata for (preferably) years.
That means they will not be new parts (particularly in civil avionics).
The qualification process is quite arduous - vibration, shock, thermal shock, temp / altitude testing, gunfire vibration (particularly tough), lightning protection (it can be done), temp/alt/humidity (not quite the same as temp / alt) to name but a few of the tests.
There is, quite simply, no way to speed that cycle up if you need proof that the thing (whatever it is) will work, and continue to work in harsh environments.
There are a few application where the more specialised mil grade parts are required (the wing pylons where you can hang weapons and / or fuel tanks come to mind).
Nice one, Preston Dunlap/Katyanna Quach.
The UKGBNI Ministry of Defence thanks you for your service and is pleased to hear they are not alone, for that litany of US Department of Defense woes is surely equalled and mirrored by their own.
However, to imagine that such be allowed to continue has one realising there be a fundamental lack of practical radical active intelligence and vivid viable imaginanation in such bodies/institutions/projects and that is inevitably extremely self-destructive ...... which may be very attractive to those hitherto unknown or formerly overlooked and/or misunderstood by defence requirements and/or procurements with products specifically/particularly and peculiarly designed to make good on the deficits revealed by such current systemic apathetic performance.
And quite why national intelligence services would condone and support such a situation without the proposal of a solution is quite telling and equally disturbing and distressing.
> The DoD's internal culture has to change
The DoD culture is to funnel as much public money as possible to external contractors. A far-second objective is to provide offensive (very) and defensive capabilities - ones that do, sometimes, actually work. Eventually.
As one who had a couple of decades experience with military systems, let me opine that the US is probably well rid of this guy. Yes, military gear is often a decade behind the times and a bit klunky. That's because it's highly desirable that military gear work after, for example, falling off a truck in a rainstorm and being run over by the next truck in the convoy. And it has to work reliably in Alaska in the Winter and Death Valley in July and in tropical jungles. Do not try that with the stuff you buy at Best Buy or Amazon. As a result, designs are conservative. And in most cases, they have been tuned over many years/decades to be increasingly rugged and reliable.
And where cutting edge technology is really needed, the military is perfectly capable of fielding it although they mostly do not tell us common folk about it. Consider the U-2 and SR-71 spy planes. Where is the commercial equivalent of those? There isn't one. No market. Are surveillance satellites today superior to Google Earth? I don't know and probably couldn't tell you if I did know. But my guess is the government stuff is probably vastly superior to what's available commercially. And note that the military continues to launch classified satellites. Maybe that's just for show. But more likely they have sophisticated payloads that do things that commercial satellites don't do today and won't do for decades. If ever.
There is, BTW, a perfectly reasonable mechanism in place for buying commercial stuff where it meets military needs -- things like office PCs, or cleaning supplies. It's called Commercial Off The Shelf (COTS) and it's been around forever -- or at least since WWII. It can, and sometimes does, deal with more complex gear if there is an actual need.
Thanks for your reasonable words, if now only people could develop a more complex understanding of the world than just single-minded sloganeering.
For example, in electronics components there are different classes:
A) consumer-grade (phones, toys, TV, radio etc, dies in four years)
B) industrial grade ( a bit better than A)
C) Automotive grade (can handle climate cycles, defined voltage spikes, defined vibrations, will live 15 years)
D) Military grade (can handle even worse climate cycles, extreme voltages, worse vibrations, can live 30 years)
E) Special stuff for space applications, which needs to be radiation hardened.
Of course from A to D there is also a price scale of 100 or so.
Ah, but "slow and reliable" isn't sexy. Everything has to be an AI-generated 5G disruptive blockchain microservice written in Angular.
I'm sure that the truth is somewhere in the middle, but the way forward for the US DoD is probably not "move fast and break things" (unless the things they break are *insert joke that conforms to your political beliefs here*). I really do not want to trust most commercial software with my life.
Many, vtcodger, who may know more about virtually anything or everything than they are ever likely to tell you or many would ever eventually discover or uncover, would point out that in nearly all cases is the vastly superior government stuff that very few folk would know anything about, supplied and maintained by a very private and/or secretive pirate sector.
Whenever nothing is ever as it seems, is everything quite different from what is widely considered usual and normal.
back in 2010-ish i worked on a system that used drones and wifi and mesh networks for ground-air-ground battlefield communications. It was part of a proposal to the DoD. It did not get the contract, though. Point is, like many such DoD contracts, you have competing businesses offering their solutions, and as long as the specification is clear and the solutions meet the requirements, they pick the one to go with, etc. (usually the lowest bidder, but not always).
Did it ever occur to you than 2400Mhz is less than ideal for a real battlefield with trees, slopes, houses ?
SINCGARS operates at 30..80MHz and due to the nature of the wavelength can bend around said obstacles. Wifi much harder or not.
Next time, hire a radio engineer or a signal corps man.
SR-71 retired last millennium. U-2 is still flying but only because its specialised military drone replacement is over a decade late. In the mean time the US military are trying out COTS alternatives - and for satellites too. We know that the best satellite reconnaissance has higher resolution than Google Earth because the previous president thought he would get likes for posting secret pictures on Twitter. One possibility is a fleet of satellites based on Starlink but with a camera. The resolution would suffer for not being in an elliptical orbit but coverage and availability go way up and for far less money.
Well done all those in the US military involved in testing COTS to find applications that are useful and save tax-payers money.
>Consider the U-2 and SR-71 spy planes. Where is the commercial equivalent of those?
The Concorde didn't quite achieve the speed and height of the SR-71 but it did serve champagne and it didn't leak like a sieve when on the ground (and require in-flight refueling with a special fuel after takeoff). Its wasn't commercially successful because the US government had a fit and so severely restricted its use -- you just can't have airliners trucking around that outperform your air defense fleet (and outrun your SAMs). Especially foreign ones.
SpaceX has also completely upset the space program -- globe spanning rockets are now a commodity. Their constellation of near Earth satellites is also proving to be both a reliable form of communication and a valuable intelligence gathering tool. All done for pocket change compared to a comparable military program.
(UK readers may recall that the early experience with the MoD's all singing/ all dancing communications system called BOWMAN rapidly became known as "Better Off With Map and Nokia".)
Innovation in the US military moves with the speed of contracting and security. In other words very slow. In cases where troops in combat need something, procurement regulations are waived and things are much faster. Why this can't happen for regular procurement is that too many rice bowls would be broken.
The only saving grace is that this process is still faster and more efficient than our adversaries'.
Icon: The desired outcome of the process.
"achieving secure communication of classified info between devices via commercial satellites"
Surely something pretty basic in this day and age
"shooting down a cruise missile from the ground using AI and 5G"
Probably just a missile with a dodgy altimeter susceptible to 5G
"cutting kill chain timelines in a "critical defense mission" from 16 minutes to 16 seconds"
16mins is pretty good for organising a meeting, setting the agenda, outlining the idea, setting the objectives, 'blue sky thinking', debating the options, making a decision, answering objections, planning a timeline, planning for follow-ups, AOB, meeting adjourned...
I do find it amusing that any government activity is met with cries of "privatize it all, fire all the bureaucrats" and any government privatization effort is met with cries of "outsourcing firms are all crooks" accompanied by stories of their incompetence, contract overruns, nepotism, etc.
I fairly recently worked as a DoD developer contractor. The place I worked (which will remain nameless to protect the guilty) is a contracting company's dream. A contractor can hire programmers, then hire CMMI* process stewards who will slow down the programmers with mandates for useless documentation. Obviously, that will require hiring of more programmers to compensate for the negative work generated by the CMMI camp and the requests from other teams. Then there are the network team, deployment team, architecture team, and myriad more all which must be cared for with proper documentation which unsurprisingly leads to needing more developers. You see where this is going. All the contractor has to do is set back and let the government bureaucracy and infighting do the work of creating new positions for them. They don't have to slow down the system to make more money. The customer does that for them.
* If you've never heard of CMMI, count your blessings. Think of the Agile Manifesto but with the phrases on each side of the word "over" reversed, and you have the right idea.