>Oh yeah, did anyone mention how long it takes to integrate a new system onto the F-35? Fracking years. All of which has to be done by LM, forever. Because the F-35 is not a jet, it's a Master Contract.
This is the new reality of military procurement and has been for years. Integrated Logistical Support contracts are preferred by senior leadership for lots of reasons that won't fit into an HN comment box, but the wave tops are that it's wastefully inefficient to have uniformed aerospace engineers, logisticians, project managers, etc. doing R&D work. Private industry does it faster, better, cheaper, and pays bigger salaries with better lifestyles which means they can attract better talent.
I've been an aerospace engineer both in-uniform and out, and I can assure you that uniformed service members (and their families!) sacrifice a lot that's hard to quantify and not always immediately apparent. It's not 1950 any more; the best and brightest mostly don't want to touch government with a 10 foot pole. There's more money and prestige elsewhere, in the private sector.
It's not just that uniformed (or DoD) engineers have no access, the subsystem vendor (e.g. Raytheon) also get no access. LM is incredibly obstructionist, even compared to the bastards at Boeing.
This cluster** has led directly to initiatives for open (Govt open) architectures and vendor agnostic interfaces for talking between vehicles and components, and between component, between jets and drones and C2 etc. That has a long way to go too, but at least we've broken with the idea that it can be a closed system.
I'm familiar with the problems of service careers. There is a lot that could be done to improve that, but that's a different discussion. I think it's extra important now that we have jet engineers who know about AI in the aviation context.