To say nothing about hardcoded repo URLs (not resolvable identifiers), transient dependencies, URL schemes, and trying to pretend github is the only package source.
391 karma · joined December 6, 2013
To say nothing about hardcoded repo URLs (not resolvable identifiers), transient dependencies, URL schemes, and trying to pretend github is the only package source.
Sure you want the source available, otherwise it's closed source, but ideally you never need to look at it unless you're a contributor.
git has really conflated the two concepts.
The race comparison doesn't hold in quite the same way, because there isn't automatically a heredity line. One group can just exploit the other indefinitely.
If you use git for packages, then your repo becomes the package boundary. You no longer have the option of producing multiple packages from one repo, or even one package from multiple repos.
Maybe it gets flagged because there's no chance of having an intellectually honest discussion if it involves Elon or Tesla? I certainly don't flag them.
As a consumer, I'm pissed off. I do feel conned.
But I'm fine explaining Musk's promises away as hubris. He made promises he should not have, and couldn't keep. He shouldn't have done it, but I do think he believed it. I don't think it was an intent to mislead. Incompetence before malice and so on.
He deserves credit where credit is due. He did push us into the EV era.
It's silly to call him a nazi because he doesn't fit the profile well at all. It only works if you redefine nazi to be whatever you don't agree with at the moment. This might even be harmful, as it's an obvious strawman that provides cover for his real faults.
He denies being a nazi. We can take his word on that. One thing about nazis is they are weren't shy about their beliefs.
My point was there's a huge anti-Tesla bias entirely as a reaction to Elon Musk. It's emotional, not rational. It's not objective criticism of Tesla. Sure, there are some awful moves he's made, vision only, FSD claims, but if it was another car company it wouldn't even be on HN. Wasn't so long ago that VW (a manufacturer literally started by the Nazis by the way!) were caught falsifying their emissions.
I ended up with the Tesla. It is hands down the better vehicle and I'd be very surprised if anybody seriously thought otherwise. There wasn't very much in it price wise so that wasn't a factor.
The BYD (Sealion 7) wasn't even a bad car. It's a good car. But it's inspired and just a little gaudy. It felt like a conventional SUV with an EV powertrain. The Tesla felt like the future.
On the whole, it's a great car, it really is. They've pretty much nailed the fundamentals. It's opinionated, not unlike Apple, but if the opinions work for you you'll enjoy the car.
But there are shortcomings, and they are jarring. The parking sensors basically don't work at all due to being vision only - and apparently can't be made to work properly. The lane change and reverse warnings are just crap and may as well not be there. My previous car implemented these to perfection, but I cannot trust the Tesla. The autopilot is a gimmick that offers you nothing but increased risk - and there's no way in hell I'd trust FSD for car that can't accurately detect the distance of my house when parking. The big touchscreen is great for passengers, but outright dangerous for drivers.
Having said all that, it seems strong emotions around Musk and Tesla cause people to want Tesla to fail. They want the car to be bad. There is so much motivated reasoning around this brand that it's hard to take any article like the above, or half the comments in this thread, seriously.
But there's an even better reason. Consistency, encapsulation of process, and a form of self documentation. This is the real goal - the time savings are a bonus.
It's also got nothing to do with Apple fanboyism. The same is true for all corporations in the US.
An ipv6 lan with default ingress deny is more secure than ipv4+nat
NAT is then unprotecting them a little by letting them punch out again. It's super easy for routers to implement this behaviour by default if your LAN is publicly addressable, and removes a whole class of exploits caused by applications making NAT hacks.
In the enterprise, RBAC is a royal pain. You give out a URL and it's hard to know if the consumer can fetch it.
URLs are absolute, there is no resolution by name. Compounded further if you want transient dependencies (maybe not needed in this instance though).
In your project, you end up hardcoding the https/ssh scheme.
If your config is turing complete and consumed as-is, then without a lot of discipline you can dig yourself into a hole, sure.
If you're producing YAML that is not turing complete, that constraint means you have to code in a way that produces deterministic output. It's actually very safe, and YAML maps 1:1 to types in something like Python.
My favourite go-to example is for AWS Cloudformation: