Re: What's the real function?
It's because TPM 1.1 and TPM 1.2 were borked. So TPM 2.0 was a re-think. Which turned out to be somewhat more subtly borked.
484 posts • joined 15 Mar 2016
Of course you can.
The typical operation is PaaK -- local bluetooth recognition. This requires no internet connectivity, but does require a working phone.
Backup operation is using a keycard or equivalent. Some have the transponder embedded in a ring or other jewelry.
An alternative is explicitly using the app, which contacts Tesla's servers and tells the connected vehicle to unlock. This obviously requires both the phone and car to have internet connectivity and Tesla's servers to be online.
Read the previous comments. A Tesla 3/Y and new S/X is usually set up with bluetooth PaaK (Phone as a Key), which requires no cell coverage for either the phone or car. You just need to be reasonably close to one of the doors.
The issue in the story only happens when you use the phone app to signal the car to remotely unlock. You can usually do this from anywhere, but it requires both you and the vehicle to have connectivity to Tesla's server.
If a Model 3/Y "house" (12V) battery has gone dead, there is a simple procedure to resolve the problem.
You pop open the tow hook cover by pressing on the 1 o'clock position. Inside you will find red and black wires, with the black wire holding on the tow hook cover. Apply 12V, perhaps from a charger or jump start box, to release the electric hood latch. Once the hood is open you can "jump start" or charge the 12V battery.
Note that applying 12V when the car is operating normally should be blocked, so this isn't a break-in method, and you can't practice beforehand.
If the traction battery still has a charge, it should only take a few seconds for the drivetrain controller to start up and activate the internal 12V charger. At that point the car should be ready to drive.
If the traction battery is fully discharged you'll need to leave the charger or jump start box attached for a minute or two to activate the charging system. If you aren't near a charger, it's tow truck or mobile charger time.
I've seen keys snapped off in automotive lock cylinders a few times.
It most often happens with aftermarket keys, which tend to be brass with a bright plating. Soft brass puts less wear on the key cutting machines than OEM stainless (e.g. manganese steel) or cupronickel, so that's what most key copying shops stock.
These brass keys are soft enough to easily bend, but then work harden so they easily crack.
The lock cylinders don't help either. The combination of old hardened grease, pocket lint, and ill-advised graphite application (automotive locks use waterproof grease) can make the lock cylinder bind up.
(Yes, I have my own vertical key cutting machine for "laser cut", dimple and other security keys. Because.. ahhhh.. well, I'm part of the demographic that reads The Reg)
Sure, you can turn it all off and not carry a keycard. But then you won't be able to get into the vehicle.
There are companies that remove the keycard transponder and install it into a more convenient enclosure, such as ring.
Tesla has a very reliable PaaK (Phone as a Key) implementation. It's significantly better than Ford's, which has been problem prone and requires a trip to the dealer to update the software if you want to roll the dice on a more recent version. Plus Ford originally only provided one key fob. They started providing two over the summer because of the PaaK failures, but ran out of chips. Now you get only one and a backorder wait if you want a second.
Why don't you talk about the details of the "less optimized algorithms", and how they directly relate to the language choice?
It's not that they didn't have the original algorithms to look at. The source code was right there. They had no qualms about directly copying/transcribing everything. They could benchmark and profile to see where the time was lost.
The coreutils programs are clever because they selective use mmap() and family when that's the most efficient way to read, process and write large objects. That's not a one-off trick. It's an efficient and effective programming model. C is flexible and expressive enough to support it, notably the ability to allow page aligned data to remain aligned throughout processing.
A similar challenge exists in the kernel. Many structures are carefully aligned within words, cache lines and pages. This is important for performance, but it often critical for correctness.
The correctness issue is a challenging one. It's often more subtle and pervasive than non-kernel programmers realize. You can't just "encapsulate it in 'unsafe'". Some objects must always be cache aligned, or avoid spanning pages. They can't have language-imposed headers, and they must be directly manipulated without allocating and copying to align them. Even if you solve every current problem, you need the flexibility to address new ones without a re-write when new hardware has an errata like "never allow an [interrupt] to occur when taking a triple page fault during the read of this structure in EL1 or higher", which effectively means the table cannot span a single physical page.
Don't think that they have those chips in stock.
Most sellers are simply playing the middleman game. They'll take your money and then look for a supplier that can deliver at a lower price. If they can't make a profit, they'll delay hoping that their price will drop. Eventually they'll refund your money. Or not.
Back in the MOSFET shortage days I repeatedly received counterfeit parts, re-marked lower spec parts that superficially appeared to work. There are plenty of stories of people getting 'factory sealed' reels of more complex parts that were perfectly marked but were something completely different, sometimes even having the wrong number of pins. That evolved into the first few parts on the reel being genuine to pass inspection, with the rest being bogus.
It's a bit off topic, but the residual oil film provides plenty of engine bearing protection during start-stop operation. As long as there is oil on the surfaces, the oil creates its own hydrodynamic pressure so there is essentially no bearing wear. Think of the oil pump as providing fresh, cool oil rather than creating the pressure that separates the surfaces.
Others have covered the need for the switch.
Oh, and the clutch probably has two switches, one for the start of travel and a second for clutch disengagement. The latter is used for engine load estimation for emissions control.
The switch itself is almost certainly using a Hall Effect device in place of mechanical contacts. That device is a complex chip with thousands of transistors. The same is true for dozens of other switches and sensors in the engine and body. The door locks, power windows, HVAC system, fluid level detection, etc all likely have Hall Effect devices. And this is just one aspect of chip use that people overlook, thinking that they are "just switches".
10,000 threaten turns into '34 refuse'
With numbers like that, you question the sincerity of even those 34.
That is about the number of people that would be expected to retire or quit every day. At a guess, most of those 34 were planning to leave anyway. This just provided them an excuse, and includes a lottery ticket for some extra cash if a lawyer is successful in court.
I expect that there are least as many that didn't actually get vaccinated and will soon be playing the more subtle 'suspended with time accumulating for retirement' game.
Looking closely at what was written, I'm concluding that that they are largely unsuccessful at RF fingerprinting.
Which is unsurprising.
Identifying transmitters by their unique signatures has a long history. A century ago radio operators could identify each other by their unique "fists", the specific way they sent Morse code. Experienced operator could also identify the CW transmitter by the tonal quality, how that specific transmitter varied in frequency and amplitude. That was easier than it sounds, since Morse code over radio is rapidly starting up and shutting off a transmitter. And a century ago, with valves and marginally stable crystals, those transitions were wildly sloppy.
Transmitter identification became more sophisticated with AM and FM voice transmissions, but the phase and amplitude imperfections at the start of transmission were still key fingerprint features. Discrete components varied, and even their exact position during assembly produced some variation straight from the production line.
Essentially all of that is gone with modern radios. Frequency-agile radios require a design the minimizes frequency deviation, and modern assembly techniques results in astonishing consistency. There might be characteristics of a chip type, but you aren't going to be identifying the specific device.
And that's pretty much what they found, even if the headline suggests otherwise.
In July 1987 I had just graduated and moved from Boston to my first full time job working for Harris Corporation in Melbourne FL. On July 12 it reached 95F with a dewpoint of 91F, a notable local record.
Since I was a new grad with college loans and little else, I was driving my college car, an old Beetle with no A/C and jammed-open heater boxes. The guys I worked with were oddly eager to volunteer to drive when we went out for lunch.
I'm observing considerable pent-up demand. I expect that it will result in full fab utilization through the end of 2022.
It's easy to predict the when inelastic and bounded demand will be sated. Automotive is a good example. Shortages stop production. Surpluses, no matter how excessive, don't expand production.
Another type of demand is out there. People that can't buy their next game console. Gamers that stuck with their old GPU, or bought low-memory GPUs. Students that bought crappy Chromebooks because they needed something for remote school, but better machines were sold out. Work-at-home types with a budget for laptop and screen upgrades.
Apple is claiming in court that they shouldn't have to release repair information because consumers going to third parties for repair is a theoretical privacy risk.
This story is about Apple's "in house" (which is really contracted out) repair being a proven privacy failure.
Does that tie it together for you?
CPU redundancy has been around almost since the beginning of electronic computing, but it largely disappeared in the early 1990s as caching and asynchronous interrupts made cycle-by-cycle comparison infeasible.
My expectation is that this will turn out to be another in a long history of misunderstanding faults. It's seeing a specific design error and mistaking it for a general technology limit.
My first encounter of this was when dynamic RAM was suffering from high fault rates. I read many stories on how the limit of feature size had been reached. The older generation had been reliable, so the speculation was that the new smaller memory capacitors had crossed the threshold where every cosmic ray would flip bits. I completely believed those stories. Then the next round of stories reported that the actual problem was the somewhat radioactive ceramic used for the chip packaging. Using to a different source for ceramic avoided the problem, and it was a motivation to simply change to less expensive plastic packages.
The same thing happened repeatedly over the years in supercomputing/HPC. Researchers thought that they spotted disturbing trends in the largest installed systems. What they found was always a specific solvable problem, not a general reliability limit to scaling.
It only takes a few pins for a serial flash, and that's the type of flash used on a management processor.
But there is no evidence that the modification ever took place!
I could have come up with a much more credible story than the Bloomberg story e.g. using direct editing of Gerbers to support a COB/DCA modification under an existing component.
But, again, there is no evidence that any modification ever took place!
Re-read the actual details of the message the Japanese ambassador was to deliver.
It wasn't a declaration of war. It was a passive statement about a unilateral cessation of negotiations. Which, it might be argued, implied that something else might happen ('war is a continuation of negotiations by other means'). But it wasn't an explicit, direct, immediate statement.
This ambiguity was the reason why both the interception and the ambassador's scheduled meeting were not considered top priorities. It was only in retrospect that they were accorded significance.
The idea of replaceable battery packs has been floated. Many, many times.
Tesla even had an on-stage demo.
But everyone concludes that the logistics don't work out. Even if you can get everything else right, such as standard form factors to support a variety of vehicle types and how to do automated quick swaps over a range of vehicle types, you still can't create a cost-competitive system.
First, you need to produce and pay for multiples of the most expensive part of an EV. That multiplier isn't going to be close to 1.0. Depending on your assumptions, it might be above 2.0. If your system relies on transportation to a central depot for charging, and buffer stock at heavily used stations, it could be higher. If you need to support a regional crisis, such as hurricane evacuations, it could be significantly higher than 2x.
Next you need to deal with how to pay for pack degradation, including aging, use and physical damage. If you are proposing battery-as-a-service, you are likely proposing a national utility-like monopoly. If you don't want a monopoly, you need to come up with a valuation formula that isn't trivially gamed. And keep changing the rules as new ways to cheat are found. And try to stabilize prices in an economic system where the rules constantly change. The mind reels at how complex the rules would need to be, and how expensive the system would ultimately become.
A subset of owners almost certainly did make money overall. And the parties with "losses" likely did pretty well.
If it is your goal, it's easy to make lots of revenue go out the side door, rather than show up as profit. Then the primary corporation folds with a giant paper loss, generating tax benefits for the shareholders and bad loan write-off tax benefits for the lenders.
Trump accumulated tax-credit 'losses' wildly in excess of his investments through the 1980s and 1990s. Those losses were part of his tax avoidance through at least 2014.
According the Washington Post story, the limit is no more than 7 days consecutively and no more than 21 days a year. It applies to any guest, not Trump specifically, but Trump's attorney at the planning meeting did specifically agree that Trump would adhere to the restrictions.
Curiously, it throws Trump's Florida mail-in vote into question. He arguably voted illegally, since his proper domicile should be either New York or Washington D.C. For 2017 and 2018 he spent a total of 69 days in Florida, which was not enough to establish new residency when it still had a stronger presence in New York.
I haven't seen a suggestion that Trump was invested in HP.
Highly leveraged real estate, especially apartments, hotels and golf courses, are his business. His tax accounting team seems especially skilled at writing off the expenses and losses several times over. But those skills aren't in line with technology investing.
What the original poster is likely suggesting is that the high levels of the US government seem readily swayed to do the bidding of private financial players. Although current set of pocket-liners is a completely different set of people from 2012.
I'm not sure if I consider this lawsuit reasonable.
But saying that YouTube isn't for kids under 13... that's obviously BS. There is plenty of content targeted to very young kids, and teachers that are requiring kids to watch YT videos for class. (That is a problem at home because we now need to enable YT access during the school day, which leads to a string of distractions.)
I'm curious at the characterization of Zamak alloys as "hard wearing".
I think of it as "easily melted". It is able to be cast with fine surface detail and minimal shrinkage, allowing die cast parts to be used without any machining. But beyond that, it doesn't have great properties. It's not strong against impact. It might bend before breaking, but really only enough to make the cracked-off piece difficult to repair. And if it's doesn't break, it's has work hardened so it's going to break when you try to bend it back. It can be plated, but there is a good chance the surface will degrade even with no exposure to moisture or chemicals.
Lock picking isn't difficult.
But it's less predictable than bolt cutters. Bolt cutters usually work, and work quickly. If the bolt cutters aren't going to work, you'll get that answer in a few seconds.
It's usually worth spending more money defeating bolt cutters than lock picking. Someone with bolt cutters is definitely up to no good and will cost you money. Someone that picks the lock hasn't yet demonstrated that they want to steal or destroy.
Know how the lock is lubricated before applying anything.
Automotive locks are typically lubricated with clear waterproof grease. Applying graphite will gum them up horribly. Lock de-icer works great to rinse out dirt and redistribute the grease, although you'll need to re-grease after a handful of uses.
Residential locks are typically brass with hardened pins. They are lubricated with a few percent of lead in the brass, not an external fluid. Graphite powder will help temporarily but lead to long-term clogging.
Back to the article: most locks above entry level have features on the driver and key pins that will defeat this approach. These features are a trivial additional cost.
I was bitten, hard, by a similar file system problem.
The standard Unix file system (BSD FFS, Berkeley Fast File System) put all of its effort into metadata consistency, and none in to file data consistency. When sometimes had the effect of "correcting" corrupted directory entries to point to zeroed blocks if they were invalid. While in some cases it wasn't entirely silent about doing this, it wasn't a fatal file system check error. The messages looked the same as and were buried among the many others in the boot log after an unexpected shutdown.
The result was that the data directory looked entirely correct, with no indication of a problem. File permissions, timestamps and sizes were correct. But some the file contents were zeros. Which were dutifully backed up, overwriting the incremental and then staggered backups. A few weeks later when this was discovered, there was no backup, on-site or off-site, that contained the data.
Ironically, this was in the middle of BSD Unix fans slagging the Linux file system designers on how "unreliable" it was by not implementing continuous metadata consistency. A Linux crash would often result many more metadata inconsistency problems, but almost never file data corruption. And especially not the horrendous silent data corruption when the directory structure appeared consistent (because the file system programmers cared about "their" directory structure and performance benchmarks, but not the user's data).
The point here is that you can have a sophisticated, responsible backup scheme and still lose data. Saying that "the harm wasn't our fault because they should have had backups" misses the point.
Film and record conversations?
That doesn't trump a paper contract. Which they had.
IBM claimed that was only a contract when it came to satisfying employment laws, not when it came to being paid. That they got a lawyer to make that argument, albeit in disconnected cases, shows that IBM cares cares more about per-quarter financials than long term business and reputation.
> Splunk spokesman Alex Harking claimed in a one-line statement to The Register that the company "does not tolerate discrimination in any form", adding: "We are committed to fostering an inclusive environment."
I'm a fan of Splunk, but the next time you see a statement from them remember that they are empty words.
You can say "we disagree with the tribunal and will be appealing", or "we are disappointed in the ruling" or "we are reviewing the decision to plan a constructive path forward".
Saying 'none of our internal rules or policies were broken' after a ruling that they broke the law suggests that they are fostering an inclusive environment for sociopaths.
I appreciate the completeness of the story. Just as I was questioning when the next Russian cosmonaut was scheduled to fly, I read the answer in the next paragraph.
My unanswered question is if the Russians can afford to fly to the ISS without NASA paying for everything plus a dacha or two. If not, and our trampoline doesn't break a spring, when and how will the replacement cosmonauts fly?
I don't agree about the U.S. not entering the war, but do agree with your main point: many large U.S. corporations engaged in world trade, and world trade meant selling to Axis countries. Many in the U.S. praised the German economic recovery from afar, with most of those not understanding the "ground truth" of what was happening.
Most here know of the IBM sales that helped Nazi Germany track undesirables, which was largely a commercial rather than ideological relationship. Few talk about Ford's outright praise and admiration for the Nazi leaders. (I realize there is a little bit of "what about.." there. My point isn't that IBM should get a pass, but that they weren't atypical. They were willing suppliers rather than active supporters.)
"Certainly you are in the USA, such a dryer is a kilowatt-devouring monster. And "modern" (i. e. about 20 years old) humidity control versus time control yields superb results in terms of drying speed and nearly-ironed clothes."
I've lived in many places where a gas or electric clothes dryer was the only reasonable option. Your situation does not represent the universal experience.
You are a few decades off for the advent of humidity/dryness sensors. They save power, but not quite as much as most people imagine. The simpler mechanism of increasing the timer speed when the outlet air reaches the target temperature is approximately as effective and less prone to false shut-off.
There are legitimate reasons for checking if you are running under validation, just as there are reasons to check if you are running in a virtual machine.
But both types of checks should expect strict scrutiny.
As for getting rid of a way to check: no, that shouldn't be done. Because there *might* be legitimate reasons, established the proper way to check and audit code that checks. If you find code that uses a different way to check, hit it with the over-size ban hammer.
New hires will get provisional positions so that they can be fired without additional expense once the immediate crisis has passed. That 'immediate crisis' is usually caused by laying off someone that was unexpectedly a key person, often someone that quietly did a difficult or knowledge-intensive job and made it look easy and easily-replaced.
I'll add an echo: "44% faster" on a cherry-picked subset sounds like pretty much the same performance.
And I don't see the value in integrating the WiFi. There is a well-defined abstraction to the network with moderate bandwidth required. It's the archetype for a functionality that should be on a separate chip with a vendor-neutral interface.
What you claim in a press release and the finances that you state to potential investors are legally very different.
Many start-ups make statements in conflict with reality in their press releases. The laws aren't written to infringe upon your right to write fiction.
The laws are written to protect investors from basing decisions on cooked books.
There are issues when you write a bogus press release and then show the investors the stories written because of those press releases as if the stories are independent confirmation. There
I agree: if they sue, they open themselves to discovery.
No doubt they are considering it right now, with the board members in their pocket agreeing to a quiet private settlement. But that runs of the risk of a change in board control undoing the plan and discovery happening. Or the CA attorney general coming around for another look.
It's better for them to lay low and try again later. Perhaps fewer people will be watching. Perhaps everyone will have gotten used to the idea.
More to lose than by making the sale?
This looks more like a strategic retreat so that they can try again later.
The lure of well over $1B had them repeatedly trying until it was quite clear that there was a chance of going to prison. And even then it was the "... and not getting the money" part that bothered them.
You just don't understand the system. It would be a security vulnerability if Apple allowed non-Apple wheels to be installed, or even genuine-but-used parts to be installed as repair parts. The transponder in each wheel might seem like pointless complexity, but it's really for your own security that the system refuses to boot if the wheels are removed or different wheels are substituted.
This is quite an old concept.
Almost 30 years ago I worked on a compiler for dbC (https://ieeexplore.ieee.org/document/279474). A key feature of this language variation was arbitrary length variables. The initial motivation was SIMD, then quickly FPGAs.
Since the hardware structure has no inherent word size, a language that allowed exactly the desired precision resulted in code that was smaller and faster. And for a certain class of problem that used modulus, it resulted in significantly clearly code.
The problem was that for all other types of problems, allowing arbitrary precision was a huge distraction. Programmers micro-optimized the range, and then were bitten by bugs or unexpected behavior. A 32 bit variable is a huge waste when you are typically iterating to 100, and still a huge waste when you change that to 500, but you don't have to worry about a u_int8 (or the equivalent of a u_int7) biting you in the ass when the change is made.
Back then it wasn't a stupid idea. It was a research project that happened to product a negative confirmation. Today...
Biting the hand that feeds IT © 1998–2022