Consumables in the very least should be replaceable for consumer products, or the price for its replacement capped, otherwise we're throwing away electronics that hasn't reached even a third of its lifetime just because the battery's lifetime is over, it's like buying a new car everytime the tires are worn out. It's an incredible waste of resources.
Same thing for connectors like USB, audio jack, buttons and joysticks.
Most egregious is the chip signature-checking and exclusive buyer rights on same chip, that should be outlawed, heavily fined and easily reported and checked. It's like adding a chip with a serial number for each tire and if you install one not from your car manufacturer then your car refuses to move. It's just a scam.
The irony is that a lot of military equipment is designed to destroy and be destroyed, yet is probably designed with serviceability in mind more than consumer electronics.
The _actual_ irony here is that batteries are often not easily replaceable in lots of military equipment, and especially not always in weapons. It may require a very extensive tear down and replacement of entire sub assemblies to replace something as mundane as a coin cell battery. And don’t even start with thermal batteries…
The Israeli Iron Dome project did it differently. From the references in the Wikipedia article I found this gem:
"As scientists we dream to sit in our offices without limitations of time and budget and to develop perfect products. But the reality is different, and these constraints forced us to think hard. There are parts in the system forty times cheaper than the parts we buy normally. I can give you even a scoop—it contains the world's only missile components from Toys R Us... One day I brought to work my sons toy car. We Passed it among us, and we saw that there were actually components suitable for us. More than that I can not tell.".
https://phys.org/news/2010-12-air-playstation-3s-supercomput...
They also have been experimenting with using video game controllers to control military equipment
https://www.smithsonianmag.com/smart-news/military-contracto...
Isn't this the very problem? These kinds of contracts should be illegal, not just in the military, but across the whole public sector. Not getting full designs along with a custom product makes the public agency dependent on a commercial vendor. Might as well put a neon sign above should provision that says "a loophole for defrauding the public".
On a similar topic, a lot of LRUs (Line Replaceable Unit) in aircraft such as the F-35, the front-line maintainers, or even squadron/wing level avionics shops, can't even do basic diagnostics on the LRU. They either have to get a manufacturer tech representative to do it, or send the box back to the manufacture for repair. They are routinely returned "No fault found" with a significant cost attached.
https://www.nytimes.com/2019/11/20/opinion/military-right-to...
https://foxtrotalpha.jalopnik.com/even-the-american-military... is a later article that references back to the NYT opinion piece.
Anyway, vendor lock-in means we can’t compete subcomponents and our long term maintenance costs are really high because we always have to go back to the vendor for service. (Sounding familiar…?) The brass started dreaming up “modular weapons systems” where we maybe develop things like universal power modules, guidance modules, and other components, then we could compete for low cost production and long-term maintenance and achieve utopia.
I spent quite a while on this, and what we found is that it wouldn’t work (at least not for weapons). Universal components aren’t well-suited to different platforms (3” rocket vs 14” missile, etc.), the modular design adds weight, reduces efficiency for the electronics systems, increases part count to make them all interoperable, and so on. The engineering trades that you have to make for modular designs are incompatible with optimizing performance as measured by the end user. Ironically perhaps, one of the major case studies that helped argue that point was based around Apple products.
Which is perhaps a long way to say that most consumers don’t seem to care about modular computers, and in the worst case many of the changes that modularity would require are counterproductive to the things consumers _do_ seem to care about.
It’s one thing to argue that we should use governmental regulation to push negative externalities back onto firms so that they reconsider the design and engineering constraints they prioritize, but it’s not like firms are being irrational or irresponsible by giving people what they want—some combination of smaller, lighter, faster, longer battery life, and cheaper devices. There would be some negative impact on some (or all) of those characteristics if Apple, Samsung, or anyone else switched to fully modular designs—and I think most consumers would choose the cheaper, faster, smaller, lighter, or longer battery life models than to buy a modular one either on principle or because they expect to repair or upgrade it for significantly longer than they expect to own them today.