Veserv

61 ms·

811.
by Veserv 3y ago
Tesla and NHTSA both agree it is a recall. As to the definition they are agreeing on, from the NHTSA website [1]: “Initiated safety recalls require a manufacturer's action to announce and remedy the defects. A recall is issued when a m
812.
by Veserv 3y ago
You are right. People seem to think a recall means “taking it in to get fixed”. But, a recall actually means the product has a dangerous defect that requires the manufacturer to promptly fix it for free otherwise the vehicle must be removed
813.
by Veserv 3y ago
The cars are fundamentally unsafe until fixed. That is what a recall means: there is a safety defect requiring prompt removal of unfixed cars. The manufacturer can avoid refunds by fixing their defective product by creating a new, safe, f
814.
by Veserv 3y ago
Why does it give a false sense of security? It is not like >99% coverage is the finish line; it is the starting line, the bare minimum. That is like saying you get a false sense of security when a handgun does not pierce your newly deve
815.
by Veserv 3y ago
A quick search shows current prices are ~80 $/kg for Tellurium, ~10 $/kg for Antimony, and ~1,300 $/kg for Germanium. Unless pure ingots are not the preferred mode of delivery, the relative scarcity of metals in the entire Ea
816.
by Veserv 3y ago
That is not even comparing apples to oranges, that is comparing apples to steel. You are correct in that energy density does not matter very much for weight-insensitive generation such as grid-scale generation, but energy density matters fo
817.
by Veserv 3y ago
Tesla has that covered as well: https://casetext.com/case/diaz-v-tesla-inc-7 Some choice quotes from the decision based on facts proven in the case about endemic racism at the Fremont, California factory: "This wa
818.
by Veserv 3y ago
The failures of a tiny player like Tesla Solar are totally unrelated to US solar photovoltaic (PV) capacity. From 2021 to 2022 solar PV went from ~161,499 GWh to ~202,080 GWh of net generation (~25.1% increase) [1]. Utility scale went from
819.
by Veserv 3y ago
Are you seriously trying to state that killing 100,000 people through a coordinated clandestine bombing campaign in another country is even comparable to the difficulty of just spending a few million dollars to develop that same remote capa
820.
by Veserv 3y ago
Welcome to 2015? https://www.theverge.com/2015/7/21/9009213/chrysler-uconnect...
821.
by Veserv 3y ago
Because if you can do that you can spend a few million dollars to hack millions of Teslas causing hundreds of thousands of deaths. You are literally comparing a few bombs to Hiroshima. Except nuclear weapons cost billions to develop, billio
822.
by Veserv 3y ago
Yes. TCP connection establishment inherently forms a two-way channel between two entitys. There is little reason to duplicate the connection/routing (in the generic sense) establishment for each direction in the general case where you
823.
by Veserv 3y ago
Which is a problem of their own making. If they use two sockets, which would be needed if all three programs were on different machines, then it would work exactly the same as the pipe example. The problem is that they saw TCP sockets are t
824.
by Veserv 3y ago
Links to their Common Criteria EAL6+ certification for their microkernel: https://www.commoncriteriaportal.org/files/epfiles/nscib-cc-... https://www.commoncriteriaportal.org/files/epfiles&#x2
825.
by Veserv 3y ago
A Neopanamax vessel is allowed to transport 120,000 metric tons of cargo [1]. ~60 K$ would only amount to 0.50 $/t of extra cost. That is only low single-digit percentage increased shipping cost [2]. And again, that is the cost of desa
826.
by Veserv 3y ago
The Panama Canal Authority derives ~3 G$ in fees over ~14 K crossings per year for a average cost of ~200 K$ per crossing. 40 M gallons is ~150 K*m^3. Modern bulk desalination turns seawater into potable fresh water at ~0.40 $/m^3, so
827.
by Veserv 3y ago
Then they can put that value proposition into legally binding, non-revokable terms. Until then, I see no reason to believe a person on the internet over a trillion dollar company executives and lawyers who actively avoid making any contract
828.
by Veserv 3y ago
Jeez, I hope it did not require “security hardening” for Nvidia to do something this basic. If these other vendors missed some tiny corner resulting in state leakage, that would be understandable. But, forgetting to clear local memory is ju
829.
by Veserv 3y ago
tl;dr GPU drivers made by various vendors do not sanitize compute unit hardware scratch memory between uses, so you can just freely read whatever the last user left laying around when they stopped. Literally too incompetent to follow even b
830.
by Veserv 3y ago
You can use function entry/exit (FEE) tracing. A optimized implementation should only incur overhead around one hardware timestamp per function call which is generally around 10-100 ns. That is sufficient for benchmarking quantitative
831.
by Veserv 3y ago
Only if their intent is deception. Here is a good litmus test. If the executives think real hard and focus test how to rename something to trick people and hide their negative reputation, then that should be fraud.
832.
by Veserv 3y ago
That is only around 80 K$ per crossing using modern desalination plants. The internet says there are ~40 crossings per day resulting in ~14K crossings per year deriving ~3 G$ in fees for the Panama Canal Authority. At ~80 K$ per crossing th
833.
by Veserv 3y ago
I only know about it broadly, but in terms of what I have read about it, it is almost reasonable except that it overly encumbers FOSS and non-commercial software. Liability should lay with the provider of the final product to use components
834.
by Veserv 3y ago
Verification is significantly more important for patient safety than openness. Just look at your local homeopathic remedy section for evidence of that; plenty of openness, but not a lick of verification. As to the formulation of your medici
835.
by Veserv 3y ago
Hardly any that are frequently ingested are anywhere in the vicinity of the openness normally expected from open source. Openness is not and has not been sufficient, necessary, or even expected for useful medicine or basically anything else
836.
by Veserv 3y ago
That is just standard compulsory defect remediation. Like the hardware components they should be required to monitor and remediate any and all newly discovered defects in any components and the product. Again, this is something they already
837.
by Veserv 3y ago
You do not receive a full chemical makeup, structure, and formulation of the medicine you are ingesting. You do not receive the entire production process of the medicine you are ingesting. You do not receive the full quality control process
838.
by Veserv 3y ago
The archived page claims they started in 2013. Time travel debugging was commercially available (by other vendors [1]) for over a decade by that point. [1] https://jakob.engbloms.se/archives/1564
839.
by Veserv 3y ago
Are you worried that a medical device manufacturer will intentionally replace certified software with known uncertified software? That would be shockingly stupid. There is exactly zero plausible deniability if they did that. That is “slam-d
840.
by Veserv 3y ago
Actually, your pacemaker should be running code certified by unbiased, competent third party experts to be free of safety defects with the liability to match. That is what we demand of every other physical, electrical, chemical, and biologi

More ›