Software updates delivered mid-flight to Typhoon fighter jets could see battlefield threats eliminated faster than ever before.
Given the average quality of first pass attempts a code that should eliminate Typhoons quite nicely. Good that, given the manner in which climate change is driving extreme weather events, no?
Depends on whether they inevitably outsource the job to India’s finest, cheapest, computer janitors who know how to copy and paste from GitHub while pretending to be coders.
I heard the South Koreans had a massive cyber security breach recently when they tried using ChatGPT to write their code.
Imagine what Pajeets will do with such technology.
It’s not new. I remember reading about Abrams tanks getting software updates in the middle of the desert during the first gulf war.
To be honest, I’d be happier getting an update as a tanker rather than as a pilot. “Software update downloaded, rebooting in 5…4…3…2…1” in the middle of a dogfight doesn’t sound like a winning strategy.
This doesn’t pass the sniff test. Big fighter plane. Encrypted comms back to base. Now send a mission critical (ie muck it up and the plane falls out of the sky) update to counter a threat that has been found and countered *since it took off*?
Nah. After landing software updates to the database of ‘stuff you might see’ I can believe though.
When are they going to go the whole hog and take the meat out of the aircraft? The first country to do this successfully will win every aerial battle and therefore every war – up until the moment that the losing side goes nuclear.
Interested, the Tempest project for a Typhoon replacement has an no-pilot option. I’m not aware of whether it has two variants or each may operate with or without a sky rupert.
I once saw an estimate that ~90% of the capital cost of a modern military aircraft is spent on measures to keep the pilot alive and functioning. Lose the meat and you get 10 aircraft rather than 1 and they’re more manoeuvrable because software doesn’t black out at high Gs. In extremis it’s easier to use them in kamikaze mode as well.
Interested – drone swarms are thinking small.
The future of warfare belongs to whoever is the first to develop self-replicating machines. Couple that with the fanciest machine learning algorithms and massively distributed computing power, and you’ve got yourself the next industrial revolution.
The attack package would then behave much like a virus, or invasive species. Seed ships, burrowing away deep in enemy territory. Constantly building, refining, and adapting in the sky, under the soil, in the water. A partly invisible army of machines, ranging in size from a mote to an atomic-powered genocide Zeppelin, all working together 24/7 for the destruction of the designated enemy.
Of course, this could well doom humanity, but it’d make a reasonably good Netflix series.
@BiW, I remember tales of a septic Aegis missile cruiser that was left defenceless for a while because the NT 4 (IIRC) system that ran it had blue screened…
Ah… they managed to update drones without crashing them more than usual…
That’s quite a difference from updating systems on the fly on a Typhoon.
I imagine the pilots themselves would have to say a choice thing or two about the notion as well…
Rhoda, IIRC the Tempest project initially had an autonomous vehicle named Mosquito as the wingman. They cancelled that and may have rolled the two vehicles/technologies together.
Steve
That’s one version of the grey goo scenario
I heard that one too. It had to be towed back to port IIRC.
To be fair, it was a testbed program:
https://en.wikipedia.org/wiki/USS_Yorktown_(CG-48)
USAustralia is working on a drone “wingman” that has a high degree of autonomy.
https://en.wikipedia.org/wiki/Boeing_MQ-28_Ghost_Bat
I remember reading a short story in the ’80s-ish where arms manufacturers triumphantly declared that their solution to the extreme cost of all the unmanned fighting machines being destroyed was to insert a cheap human to replace all the expensive electronics.
The future of warfare belongs to whoever is the first to develop self-replicating machines.
Only someone without the slightest clue about engineering could come up with this one.
You any idea how many supply chains come together to produce something simple like a kitchen food mixer?
It’s like people wittering on about nanotech.. Where’s the memory drives your eeny-teeny replicators? You’re going to need sufficient for a few million lines of code. What are you writing it on? Quarks?
They’ll still have jet jocks. And a dog, to bite them if they even think about touching any of the buttons.
It may be possible to update the software in flight via a real-time process. What is not possible is to get the update through Change Management* in less than three months.
*Change Management. Ideally a process to stop you fucking up. Otherwise just an obstacle. Trouble is you don’t know which on any given occasion.
Mid-flight software update use case from 2007, no casualties:
http://catless.ncl.ac.uk/Risks/24/58#subj1
(first two items):
As the Raptors reached the International Date
Line, the navigation computers locked up, so the aircraft returned to
Hickam until a software patch was readied. “Apparently we had built an
aircraft for the Western Hemisphere only,” says a senior U.S. Air Force
official.
and
[Long ago there was an urban legend about the F-16 flipping over and
flying upside down when it crossed over the equator. That report emerged
because a consequential software flaw had actually been DETECTED in
simulation, and had been FIXED before it could have happened in actual
flights.
And another near miss, 2006:
“Don’t Try to Program and Fly at the Same Time”
http://catless.ncl.ac.uk/Risks/24/50#subj3.1
RISKS Digest – recommended reading 🙂
“Steve
April 14, 2023 at 10:54 am
Depends on whether they inevitably outsource the job to India’s finest, cheapest, computer janitors who know how to copy and paste from GitHub while pretending to be coders.”
It doesn’t even depend – at best these will be the finest programmers you can find in defense contractor sinecures;)
Hope there’s an easy and quick process to rollback patches!
“bloke in spain
April 14, 2023 at 1:19 pm
. . .
Only someone without the slightest clue about engineering could come up with this one.
You any idea how many supply chains come together to produce something simple like a kitchen food mixer?
It’s like people wittering on about nanotech.. Where’s the memory drives your eeny-teeny replicators? You’re going to need sufficient for a few million lines of code. What are you writing it on? Quarks?”
I would just point out that *we* are nanotech machines.
BiTiN – no, this one is pink.
BiS – I’m tired of this lack of ambition.
We’re footering about with widgets and toys, when we should be dreaming the big, insane, Laudanum-haunted dreams of a feverish Isambard Brunel.
I want the space elevators, flying cars, O’Neill cylinders, sarcastic robots, and sassy space princesses we were promised. I want top-hatted Lunar aristocrats playing golf on Mimas. I want giant atomic death robots. I want a pillow that doesn’t get hot under my ear.
These are my demands. If they are not met within 48 hours, I plan on firing a laser beam at the Moon.
Candidly, you have been warned.
Agammamon – imagine if aeroplanes were as reliable as Windows 11 though. Exciting!
Better hurry up. When Elon gets there he’s building a solar powered rail gun that fires tons of molten aluminium at anyone who tries to take the Moon back off him.
He’s also bought a furry white cat and a grey Mao suit.
We can always dream, can’t we?
As someone who has written avionic software for high speed, customer-focussed, bomb delivery services, this one was ROFL.
“But if it’s written in Ada, it cannot have bugs, right? It’s a certified compiler, so no testing is needed.”
Yes, really.
On the upside, Ivan will be able to BSOD the sky-ruperts with much less trouble & expense.
@Interested – “When are they going to go the whole hog and take the meat out of the aircraft?”
Have you missed what’s going on in Ukraine? Both sides are making use of lots of drones. And lots of video taken from drones is available showing the results.
Yup. Must admit that I can’t see the point of the Russians building massive trenches if the drones just fly up, spot one or more Ivans and then drop bombs on them.
I mean trenches worked well enough a century ago when it was protecting them from forward pitched artillery weapons, but when you’ve got a drone dropping bombs on specific targets (rather than scattered randomly in the general direction of the enemy), then that seems both more efficient and harder to avoid unless Ivan is going to start hiding in hand dug caves off the trench (like WW1 soldiers used to sleep in), in which case what is the point of them being their except as targets.
The Russians seem to be reenacting WW1 for some reason.
A 2002 Michael Crichton “Prey” is a about a nanoswarm with distributed computing and using a flight pattern similar to the “murmuration” of birds.
. . . I can’t see the point of the Russians building massive trenches if the drones just fly up, spot one or more Ivans and then drop bombs on them.
The drones are still few in number compared to the extent of the battle. You’re far more likely to get taken out in the open by a sniper, machine gun or mortar round than killed in a trench by a drone bomblet. You see disproportionately more of the latter because the video feed is integral to targeting. Both sides make extensive use of trenches and other fortifications because they are still the best solution to relatively static warfare.
The Russians seem to be reenacting WW1 for some reason.
Strangely not. If they (quickly and discretely) massed tens of thousands of their newly mobilised cannon fodder and sent them over the top in combination with precision targeting of local Ukrainian defences, they would very likely be able to overwhelm the defenders and open up the rear areas for exploitation by mobile armoured forces. They may yet do so in the better weather. But so far they seem to think sending forward a dozen tanks and a hundred guys into mined and artillery-zeroed open approaches is a good idea (then immediately doing it again, and again) especially when repeating the same tactic in different places several times. Morons.
Prediction — in the next decade there will be extensive work done on anti-drone systems. So that by the next big war they won’t be half as effective.
We’re just learning how to use them effectively. Next logical step is how to stop them.
I would just point out that *we* are nanotech machines.
It is rather the point Agammamon. And it takes 30 trillion odd of them to make a gunsmith to disassemble a gun. And any nanotech system is going to require the same level of complexity to perform the same task.
That sounds A LOT like WW1. Especially The Second Battle of the Somme in August/September 1918.
First time sitting in one of Birminghams spiffy new (Spanish made) Trams on the way to work this morning. Got almost halfway there before it throws a wobbly and stops dead.
Overhearing the driver on the blower to the control center – “turn it off, then on again, see if that works”.
One restart later, no joy – it’s officially dead.
If thats the best they can do on something that essentially only goes backwards and forwards, I would not like to be a pilot trying to do an on-the-fly software/firmware update in mid air, on a fly by wire aircraft that is designed to be inherently unstable.
Good job they have ejection seats!
Bloke: Same thing when the grid went pear-shaped a few years ago – many modern trains needed rebooting. The joke is that a software update just before that event removed the capability for the driver to do that, so a man in a van with a laptop had to go round to reboot them.
Developing and deploying software is still not an engineering discipline.
Especially The Second Battle of the Somme in August/September 1918.
I don’t know how you can compare the two. The Second Battle of the Somme was a successful strategic level mass attack(s) that led to the end of the war. The bizarre serial piecemeal attacks made by the Russians (for instance at Vulhedar) just resulted in horrendous losses and no gains. If they’d sent in all at once a force the size of what they lost in dribs and drabs, they’d have probably taken the town.
The general responsible (or scapegoated) has finally been sacked, but the basic lesson doesn’t seem to have been learned. Everywhere the Russians seem unwilling to commit a force big enough to get the job done; perhaps overcautious after their disastrous assault on Kyiv at the start.
I noticed Saddam seemed to have similar problems during the Iran/Iraq war.
Logistics too poor to conduct a big offensive??
TG – About 20 years ago, the post-privatisation sliding door stock had a bit of an upgrade, so the doors would only open if the train was in the correct position according to GPS.
Bit of a problem at Tunbridge Wells, as there’s tunnels at both ends of the platforms, and it’s sort of half cut into a ridge. No reliable signal.
So the driver either had to move the train back and forth for a bit, or try a sequence of reboots, or the guard had to go to every door and take them off that control. Hours of fun.
“Logistics too poor to conduct a big offensive?”
Entirely possible. The thing we keep hearing about is the difficulty that Russia is having with supply.
Most of the conflicts out there are either small (guerrilla) or they have problems with supply. It’s the most difficult part to get right, possibly because when it goes right it’s not noticed.
Rather like infrastructure repair, even in the West. You notice that there’s lots of new construction announced (with the politician’s name on the bridge) but you still have lots of stuff falling apart (potholes or even the occasional bridge falling down).
Dubai has lots of very modern buildings where the toilets lead to holding tanks with a chain of trucks taking the overflow out to the desert to dump.
A new building or bridge has lots of publicity. Repair has none (at least until it falls down). It gets worse when you have graft sucking away the resources, but even without it’s still an unglamorous, difficult, utterly necessary job.
M – failing to repair potholes boosts GDP.