Posts by DS999 6057 publicly visible posts • joined 9 Jun 2020
Microsoft reportedly granted Qualcomm exclusive rights for official support of their ARM CPUs in exchange for Qualcomm supplying funding and support for the Windows port to ARM.
Not sure when this deal expires, but that's the reason Microsoft doesn't sell licenses for Windows/ARM and thus you can't legally run it in a Mac VM. The only legal way to run it is to buy hardware using a Qualcomm ARM CPU.
To be more comparable to the density of TSMC's. Since terms like "7nm" or "3nm" no longerhave any meaning as far as actual physical dimensions of the transistors, it is probably fair to do so.
Thus Intel's latest "10nm" is called Intel 7 and is comparable to TSMC's N7, and their upcoming process is called Intel 4 and is comparable to TSMC's N4 process which has recently entered mass production. They claim (but it remains to be seen) their Intel 3 process will enter mass production before the end of next year, which would be comparable to TSMC's N3 which enters mass production around the end of this year.
So while still behind, they aren't quite as far behind as the article makes it sound.
The power cable probably runs alongside (or with enough armor cladding, is) the anchor cable.
Far offshore oil rigs manage to stay anchored in rough seas (some use station keeping as well but not all) and they are MUCH larger/heavier, so it sounds like that part of it is a solved problem.
going that far back makes it difficult (impossible?) to get the 100X performance improvement they want
Depends on from how far back they're coming from. I remember not terribly long ago hearing that NASA was launching stuff using rad hard 8 bit CPUs, anything from 2001 would go way beyond 100x improvement over that.
Edit: a little googling shows the Mars Rover uses a PowerPC CPU similar to what you might find in a late 90s Mac (though clocked slower, probably because of being made differently for the radiation hardening)
It looks like the HSPC project mentioned in the article is going to base the next one on ARM Cortex A53 with up to 8 cores - so there's 8x of their performance increase.
People who think Russia is coping well with sanctions don't understand the impact they were intended to have. The idea sanctions can be placed and they will have immediate impact is silly. It prevents future trade, it doesn't magically undo past trade.
Russia cannot get parts they need to make new products (both military and civilian) or repair existing products. They had stocks in country so that isn't an impact that's felt right away, but those stockpiles don't last forever. Even if you can find workarounds for some by using/modifying Chinese made parts, or do a complete redesign that uses only Chinese made parts MANY of the kind of smart people who figure out how to do that have already left Russia.
There is plenty of evidence they are having increasing difficulty because of the sanctions. Heck, just the fact they are forced to use Chinese consumer drones rather than making more of their own tells you all you need to know. The US and NATO can resupply Ukraine's military a lot faster than Russia's military can figure out how to redesign equipment that uses western parts or repurpose Chinese civilian hardware for military purposes.
Not true, the Celeron was a low end chip - there were Pentium IIs contemporary with it clocked at over 450 MHz which is why it worked so well.
Likewise today the high end chips clock at over 5 GHz, so taking a low end version of one that tops out at say 3.5 GHz should produce the same result. If it wasn't for Intel (not sure about AMD) preventing that simple type of overclocking, not to mention the multicore stuff further complicating matters.
Overclocking is less useful today because most chips can overclock themselves to some extent. If the 300A had defaulted to 300 MHz but turboed to 400 MHz there would have been less point in overclocking it. The value proposition is reduced the less benefit overclocking provides.
More likely they had certain people targeted, and you can assume it was not for a good reason.
It would make sense to toss in some randos to obfuscate their actual target(s) and any relationship between them there might be, to make it look like a test or script kiddie games.
Some of those people targeted will have cause to lose sleep, or may end up in a permanent sleep, depending on the reason they have been targeted and by whom.
Best case it is trolls or griefers wanting to harass people they feel have "wronged" them online. But it could be a government looking to crack down on dissidents, or criminals trying to track down and kill someone who robbed them.
Many modern cameras (and certainly all smartphones) have a GPS to provide both the location and date/time the photo was taken. That could be signed as well, then you have proof that not only is the photo untouched but it was taken at a particular time and place.
If you try to manipulate a photo of a particular event no matter how good your "photo of a photo" could be made (and I'm skeptical that could fool more than the casual observer) you wouldn't be taking it at the time and place that event. So you couldn't modify a photo taken during for example a protest to add a gun into someone's hand, for example - because even if you used the method you suggest the metadata will be wrong.
This isn't to authenticate that a picture of you is you, it is to authenticate that a particular digital file is the exact same digital file that was saved by your camera. If you cropped a photo to publish it and then someone questioned "that close up of a UFO is a fake" you would have the original file with the cryptographic checksum that authenticates it to prove that photo is not a fake "look see, this is the same close up UFO photo just not cropped!"
It doesn't matter if the cryptographic checksum is removable if the intent is to use it to prove that a photo is original and untouched. It could be part of the MakerNote section of EXIF data.
It would require generating a checksum somewhere outside the control of the OS, presumably in the ISP and thus could not be spoofed even in a jailbroken device. You could have one checksum "signed by Apple" and another "signed by this particular phone" which would help in establishing copyright or chain of custody (but should be disabled by default as linking your phone to a particular photo/video could have some downsides, especially in an authoritarian government)
Now how well this would work I don't know. There would obviously be incentive for some to find a way around this, and I don't think anyone would argue this is something that a nation state level actor couldn't overcome if they decided the effort was worth it. But if it worked well enough otherwise it could be used to provide a pretty high level of assurance that a photo hadn't been retouched or a video wasn't deepfaked. Apple would update its key for each iPhone generation just in case someone cracks the hardware and exposes the key in an older phone.
Theoretically Google could do it too, but only for the Pixel since they only control the hardware there. But Samsung and other Android OEMs could use their API to do the same, though security would depend on the OEM's implementation and how resistant their hardware was to attempts to compromise it.
Worse case, they try this, people find a weakness in the hardware that lets them extract the private key so they can forge signatures, and you're no worse off than you are today.
Could be kind of a problem, since the latest models allow them to "self drive" without anyone even in the cab. If someone hacked a tractor driving by itself via LTE and steered it off its proper plot of land and made it drive down the middle of the highway or into the nearest town they could cause a lot of havoc.
Intel has cross licensing deals with a lot of companies, that protect against those companies suing them but prevent targeting those companies with lawsuits of their own. By transferring some of their patents to a third party, then those patents are not covered by cross licensing agreements and the third party can troll on their behalf and split the proceeds.
The risk is obvious. What if AMD, TSMC, Apple, Nvidia, etc. that Intel has cross licensing deals with do the same? Suddenly Intel will find itself on the sharp end of a lot of troll lawsuits, when they thought they'd be the one doing the trolling.
Everyone stands to lose here, except the trolls for hire and their lawyers.
That's probably been a bigger hurdle to adoption of nuclear power in the last few decades than anything else, at least in the US. Today every reactor is custom built, so parts are custom made and approval is on a case by case basis, which can stretch things out for years even after all the other anti-nuke arguments have been made and pushed aside.
Designs need to be standardized, so every reactor uses the same design, parts, maintenance schedule, and training, even down to the same layout in the control room. Imagine if every passenger jet was custom made and pilots could only fly the one plane they'd be trained on! If you need more power at a particular site you build additional reactor units. A lot of the overhead like security and waste storage won't go up by much as additional reactors are built on the same site.
If you can spit out cookie cutter designs you will reduce the cost to build, operate, maintain, and retire plants, thereby reducing the cost of the electricity they produce. One of the big reasons nuclear plants aren't getting built in the US is because the power they produce ends up being more expensive than our very cheap natural gas. So utilities will fund new gas plants instead because they don't have to muck about for decades getting approvals and endure massive cost overruns.
So long as they have been able to point to decommissioning of coal plants they are able to claim that building new gas plants is reducing CO2 - but it will be hard for them to make that argument when their last coal plants have gone offline and gas plants are just adding generating capacity. They would dearly love to have a zero emissions alternative for baseload while we wait for the solar / wind / grid energy storage triad to be viable in that role, but not if starting the process in 2022 means maybe they generate their first watt in 2045 after budgeted cost has tripled.
One always goes with the other, similar to how communism and authoritarianism always go with one another. And for the same reasons - because both systems have their flaws, and those flaws lead to negative consequences that come along for the ride.
In the case of communism, because once you get beyond a community the size of "everyone knows everyone else" or at most one Bacon number distant ("everyone knows someone who knows everyone else") it just doesn't work and needs to be forced on a population against their will.
In the case of capitalism, when your whole system is based on private ownership of means of production and the profits derived from them, then those who hold public office are defacto private owners of the means of producing political outcomes, so they profit from "political favors" exchanged for money or other goods/favors like in any other market.
"If you don't give me royalties for successful product recommendations, I'm quitting Amazon and going to work for eBay"
Fine, if you can figure out how to rewrite your own software and physically move yourself to eBay's datacenter and get connected, go for it!
As you say, taken to its logical extreme Apple would own a ton of photo copyrights for those "shot on iPhone" pictures.
Just because there is a so-called "AI" doing post processing doesn't mean the photographer doesn't deserve the credit. Otherwise the crap I call "pictures" I take with my iPhone would be as good as the winning entries. They won because they chose a particular subject, composed the shot properly with appropriate lighting, and so on. Because they are real photographers, and the only thing I know about photography is how to point a camera at something I want a photo of and push a button.
That mostly AI generated picture was Cosmo cover worthy because the creator knew (probably through a LOT of trial and error) how to exactly specify the terms to get what she was looking for. She didn't come up with that quite wordy and detailed query the first time, that's for sure. Plus, and most importantly, she had to come up with the idea for she was looking for.
If she said "give me a Cosmo cover worthy picture" and it came up with that on its own, then I'd change my mind and say the software deserves the credit rather than the human operating it. I feel pretty confident few if any of those reading this will live long enough to see that day.
You're saying "build it and maybe someone will figure out a use for it".
So why not run browers in kernel mode so they can have direct access to anything and everything in the client, without needing any special APIs? While someone "might" find a use for an ambient light sensor in a browser, it is almost guaranteed they will find a use for running the browser in kernel mode so it can access everything.
Maybe though, just maybe, the browser would want to prevent its unlimited access from possibly being abused by the code running on it. To address that issue, perhaps the browser could implement some sort of "operating system", with an API to allow applications limited access to certain functions?
Is there for a browser to have access to an ambient light sensor? This is the kind of crap Google (along with Bluetooth, USB and other stuff) that Google wants browsers to have access to.
I'm glad I don't use Google or Chromium derived browsers anywhere, so I don't have to worry about Google building in obvious security issues because of their insatiable lust for data. They have this strange belief that browsers running client side Javascript or WebAssembly should be able to run anything the client's native OS can no matter how low level the hardware access it requires.
No one should support Google in this goal, having a common API that every client in the world can execute is the fastest way to malware that can simultaneously attack every client in the world. Having clients with different APIs to avoid a monoculture is a GOOD thing, even if it is inconvenient for those who want to rule the world via software.
I always use the "open in Safari" option. Not because of privacy (though that's another reason now I guess) but because Facebook's in-app browser does not use the ad blocker configured for Safari, so the pages run slow because of all the ads they're trying to show.
Ransomware might turn out to be a good thing, by forcing companies to finally treat security seriously and implement policies known to really make a difference like 2FA and zero trust, rather than sending yet another company wide email to be wary of opening unknown attachments and phishing, and making them undergo yet another Powerpoint based multiple choice corporate security "training".
even iMessage is a distant also-ran in most geographies for iOS users
I would expect that's the case in places where iPhone has only a few percent of the market, as places like that tend to be dominated by some messenger most people in the country have tacitly agreed upon, like WhatsApp or WeChat.
In most western countries iPhone's share is far higher than that and I would be surprised if most iPhone owners are using something else. You have any data to back up your claims?
As for me, I use iMessage to communicate with everyone, and since it seamlessly falls back to SMS for those who don't have an iPhone I can communicate with everyone. I have yet to have a single one tell me they want me to contact them via WhatsApp or some other application. I don't have any other messaging apps installed anyway, and likely wouldn't bother for just one person.
MMS is a completely different bucket of hake. MMS is not free, and in fact is stupidly expensive. It costs hugely more to send media by MMS than by IP-based protocols
You get charged for sending MMS? That hasn't been true in the US for ages, where pretty much all plans offer "unlimited texts" and MMS has always been included as a "text".
Pretty sure carriers here count MMS against your data allocation (since that is how it is delivered) so not sure what you mean by "costs hugely more than by IP based protocols". You pay data charges to send multimedia via iMessage or WhatsApp or RCS too, so it is the same cost either way. IP based protocols are how MMS is delivered, with some extra steps that involve an SMS message with the URL for your phone to visit to pick it up after it has traversed one or more carrier servers.
Not only that it requires much more complex carrier involvement than is necessary for SMS/MMS, and as a result is far from foolproof at this time as it depends not only on Google servers but also the carrier has to support it. So even Android users will fall back to SMS if they roam, etc.
More to the point, consider that Google has changed their messaging strategy more often than some Reg readers probably change their underwear. What reason do we have to believe that this 8th time or whatever is finally the one Google will be sticking to?
If RCS is so great why did Google try so many proprietary messaging schemes before taking a shot on this well over decade old technology that's languished unnoticed until recently? If they had a coherent messaging strategy they wouldn't be supporting RCS today either.
Just as if the pro life view was enforced from the federal level imposing it on all states
Except plenty of republicans have already said that's EXACTLY what they plan to do if they take back control of the government. And I don't think anyone on either side doubts judges appointed with the specific qualification that they would overturn Roe would block that law from taking effect nationwide.
An amendment guaranteeing equal rights for women has been waiting for enough states to approve since about the time of the Roe v Wade decision. If you can't guarantee basic equality for women, what do you think the chances were for getting them to guarantee abortion rights? It takes 3/4 of states, or 38 of them. There was never a chance of this. One of the arguments against the equal rights amendment, by the way, is "they already have equal rights so this is unnecessary". They would have said the same thing about a constitutional amendment to guarantee abortion rights.
They could have passed a law that preempted state laws, though this activist court would probably have declared that unconstitutional at the same time they reversed Roe.
Supposedly there are protections in place against that happening this time around. Hopefully the subsidies aren't paid until AFTER the connectivity is provided as promised, or at least they can be clawed back (and companies can't hide behind using a subsidiary to claim them and then bankrupting that subsidiary)
The problem is the people writing these laws and regulations are not as sophisticated as the armies of lawyers who examine them for every loophole. If they get just one thing wrong the cable/telco crowd will seize on that as their way around actually doing what is supposed to be required of them.
That doesn't mean there isn't ALSO corruption - i.e. going out of their way to insure there are loopholes for their industry buddies.
It isn't the fact it is "improving", it is the fact it isn't good enough to qualify today. Early adopters with Starlink who were getting 100 Mbps speeds are reporting (and showing speedtest results to confirm) speeds less than 1 Mbps now at certain times of the day. Starlink has obviously been selling their service faster than they should despite the waiting list, as their satellite launches are clearly unable to keep up.
Given Musk's poor track record keeping promises, they certainly shouldn't believe his claims that it will better once they launch more satellites. I could easily see poor speeds persisting even after they have their full constellation in orbit, and him coming up with excuses like "it has proved too popular and we have more subscribers than anticipated, but our next generation satellites will fix that problem!" He's very good at kicking the can down the road by ignoring the failure of his previous promises when making his next promise.
That's something only broadband providers without competition can do. Everywhere else broadband prices go down and/or speeds improve, because technology is improving at a much faster pace than even our current bout of inflation can put pressure on them to rise.
Starlink would not be able to do that either if they weren't essentially the only game in town for rural customers (there is Hughes in the US, but as a GSO satellite provider the latency and speed is worse, but needing far fewer satellites at least the price is lower also)
It is kind of a damned if you do, damned if you don't thing. Yes, the problems from publishing the names are obvious as you state.
But consider if the names were withheld, and charges are later dropped, or they are found innocent in a closed court proceeding (it would have to be closed to prevent their names from getting out) How do you know justice has been done, rather than been perverted by who the defendants were or the connections they had?
Already the US has issues with equal justice under the law, as rich people are able to afford a much better defense than poor people who have to accept a state's attorney appointed for them. If you hide the identities, it is virtually certain that gap would become far worse.
If you're a rich person indicted for a crime today, maybe you can make it go away if you have the right connections and the means to make the right donations or promise a cushy job to the right people. But when the fact you were indicted is publicly known, the status of that indictment can be tracked and if the charges are dropped it can be looked into - and if there are campaign contributions made around that time or if the person who made the call to drop the charges later gets a $1 million salary working for you it looks mighty suspicious. If that all happens in secret, none of that looks suspicious, so the odds of any rich well connected person ever going to jail falls to near zero.
Edit: another reason pointed out below - in a conspiracy you may not get everyone. Making the indictments public causes those involved who haven't been caught yet to shit bricks they're going to be named. Maybe they choose to surrender voluntarily, admit their role and testify against the rest to reduce their sentence before it is too late!
The US is energy self sufficient and has been for most of the past decade. Sure we do import some oil but we export more oil and natural gas than we import.
Energy is fungible, so unless you ban export you are still part of the world market and still respond to shortages. For many years the US had a ban in place on exporting oil (but not refined products like gasoline) but that was lifted a few years ago because we were producing more oil than we were using and the oil producers said they would have to end investment in new projects. They've (mostly) ended investment in new projects since the pandemic anyway, so I think we should return the favor by banning exports once again. At least our oil price would be a bit more independent of the world price for our trouble.
That doesn't seem right. Why aren't there abandoned oil pipelines filled with solidified oil all over the world if that's the case?
Ukrainian operatives sneaking into Russia to damage pumps on Russia's oil pipelines would be a winning strategy, by denying them money to fund their "special operation" as it would take years to build a new pipeline alongside the now-abandoned one...
Though after seeing video of what the Canadian "tar sands" oil is like I have wondered how the heck it is possible to get that through a pipeline at all, as it is practically a solid already.
Insider trading requires possessing insider knowledge about the company whose stock is being traded. There was no "insider" knowledge of Tesla, Musk is selling stock based on uncertainty about the outcome of his Twitter deal and whether he'll need to raise additional cash.
If Tesla shareholders try to sue over this it will be thrown out of court without even reaching trial.
Pretty sure it is not illegal sell shares now because you are worried that potentially adverse news about you personally in the future will make others want to sell. I have plenty of criticisms for Musk but I think he's in the clear with this one.
True, but requiring alternate payment systems for in-app payments might eventually force them to operate their app stores for free. Already most apps are "free" with in-app payments required to get useful functionality to work or prevent ads that fill the screen and make you wait before you can clear them.
When app vendors think "well I can charge $10 for my app and keep $7 or I can make it free to download, charge $10 in-app to begin using it, and I'll keep it all $10" why should we expect to see ANY non-free apps?
I guess that would solve the third party app store issue, as no third party app store could survive if they knew they wouldn't ever make a penny in revenue.
A small town near where I live used to have a radar speed sign just before entering town, below a "speed limit 25" sign. They took it down because in the middle of the night someone kept sticking a piece of cardboard on the signpost below the speed indicator listing their "high score".
That was many years ago, before photo radar was a thing.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
