This, btw, is why CVE scores are insane at times.
The vulnerability is that when spawning a new process which is a bat file you need special treatment of the arguments to avoid spawning a second process.
So you need a rust program setup to spawn other processes which also somehow forwards unparsed user input into those processes and is executing a bat file.
There’s a reason nobody has fixed this, it’s because it’s an insane setup that affects basically no rust programs.
The cost has already been paid. Even small farming communities have rail line access that’s mostly been abandoned because the line owners switched business models.
As for flexibility, again, that’s mostly an issue with how rail line management has evolved. From shorter more frequent trains to ultra long infrequent trains. Mostly to cut the cost of staffing.
The solution is simple, nationalize the rail service. Put it under the USPS and have them figure out scheduling to optimize the speed of goods shipping.
The current state of the rail system is entirely due to the monopolistic nature of ownership. The incentive is to increase prices as much as possible while shipping to the fewest stops possible. Profit motives are in direct conflict with generalized shipping.
The reason trunking works today is the public nature of roads. Well, why shouldn’t rail lines be equally public? We practically gave the property away to the current rail owners with the notion it was for the public good… They’ve failed that.
I wasn’t fully aware of NVK and where it’s at. It’s actually pretty exciting. I wouldn’t mind dropping my current nvidia binary blob for fully open source drivers.
Not as bad as you might think. The nouveau drivers have come a long way for maxwell. You should give it a shot if you haven’t. But, unfortunately, if you are using anything new then nouveau sucks. It’s a fun game where you get to wait until nvidia no longer wants to support your GPU and hope by that point that nouveau has progressed far enough that you won’t be looking at noman’s land.
Graphics drivers are what matters. Your orange pi uses a mali GPU which is well supported by Linux (thanks ARM).
nVidia is just barely at the point where their most recent gpu drivers aren’t terrible under Wayland. It’s taken a while to get there.
GPUs with good open source drivers will fare fine.
It does mean you are more likely to get eyeballs on your driver from other people doing graphics card driver work. That usually results in higher quality and a higher likelihood of catching issues.
I using and deploying to kubernetes. Nerdctl has a docker API but it’s completely backed by k8s. So, for regular dev I’d just need a k8s cluster and not k8s + something else to build the images and push them into the k8s image repository.
WSL2 is Linux on a virtual machine. Docker for Windows is running in a VM.
I’m also a weirdo though, I’m using podman instead (and may switch to nerdctl).
It does not work like that.
The problem with such statements is the energy costs are nowhere near fixed. The amount of energy needed to play a song on my iPod shuffle through a wired headset is wildly different from the power needed to play that same song on my TV through my home theater equipment.
The same is true on the backend. The amount of power Google spends serving up a wildly popular band is way less than what they burn serving up an unknown Indy band’s video. That’s because the popular band’s music will have been pre-optimized by Google to save on bandwidth and computing resources. When something is popular, it’s in their best interests to reduce the computational costs (ie power consumption) associated with serving that content.