Posts by doublelayer

Europe lagging behind South Korea, Japan, US in 5G rollout

Re: 5G Ohhh Ahh!

You certainly don't need 5G for a 10 Mb/s connection. 4G could have and did do it. They're probably using 5G because there's more unused capacity in it and a fixed receiver is likely running on mains power so the increased power usage isn't an issue. Of course, with that speed cap, people would probably only buy this if there's poor cable service but good mobile service in a location. That speed is probably acceptable for a single user, but scaling to a larger household probably doesn't go well.

Re: 5G Ohhh Ahh!

If the cable connection is bad enough (which their multiplier suggests isn't the case for them) and that speed is consistent, that could be shared between multiple devices on the home network. Many devices using that connection simultaneously could hit that speed limit, and are much more likely to hit the limit of 4G service.

The main problem I've seen with using mobile networks as home internet is that there is usually some limitation to the data that can be used (sure, they say unlimited, but there's often a throttle threshold somewhere and it's often low enough that someone who streams a lot of video is likely to hit it frequently). If your speed gets cut after you've used a hundred gigabytes, then the excellent 5G performance won't be very helpful anymore.

Huge nonprofit hospital network suffers IT meltdown after 'security incident'

Re: Shocked, shocked I am!!!

I think a lot of large U.S. hospitals work that way. The places providing the care may be nonprofits, but that's not the same as free or charged to governments. It just means that the money collected from patients goes to a fund to continue upkeep of the facility, not to shareholders.

Re: Is it time?

It depends. What do you want that to accomplish? If we use "terrorist" as just a type of criminal that could be better or worse than a different kind, then no because the definition tends to include committing violence for a political end and neither of those happened here. If it's now to be used for criminals that cause large amounts of damage and we want to put extra energy into capturing them and terminating their activities, sure, but that's probably a different word.

I think the best approach is to have a new term for the type of criminals this contains, as overloading the word terrorism hasn't produced useful results in the past and can lead to unexpected events.

No, working in IT does not mean you can fix anything with a soldering iron

Re: Phone Cabling

That does seem to be the general expectation. Not that I fight it too hard, but I don't expect other people to do whatever they're good at for free. Somehow technical help (which since I'm a programmer my acquaintances think covers everything from hardware failures to data recovery to web design) is the exception to this. I'm usually happy to do it for close friends and family, but there is a fine line between my willing to be generous because I like them and people expecting that I'll do anything they want at any time.

Rather than take the L, Amazon sues state that dared criticize warehouse safety

Re: Who are their lawyers?

"I wonder if 3 shifts/24 hours of manual labour for a year costs more or less than the $20,000 Musk reckons his humanoid robots will cost?"

More. I found a few estimated average for Amazon's warehouse workers in the US ranging from $15.50 to $17.00. There appear to be some jobs and locations significantly exceeding these, but they didn't provide enough information to filter out higher-level supervisor jobs in the warehouses so I can't confidently use those. I know they've talked about increasing that and may have done so, but let's assume these are still accurate. In fact, let's assume these are overestimating and use a nice round $15.00 per hour.

If they have absolutely no overtime, then wages alone for a single worker-year (full-time 8 hour shifts 5 days a week) would be $31,320. Extending that to a single worker for every hour in a 24/7 setup would be $131,400. Amazon also has health insurance benefits, payroll taxes, unemployment insurance, and other expenses, so these numbers are significantly below what they have to spend.

The only problem is that a humanoid robot may not be able to do what a human can. I don't work in robotics, but I know enough people who have to watch them do a lot of work to get a robot to do something that comes very naturally to a human. If the robots are to be deployed in an environment with unplanned obstacles and for tasks without an easy deterministic answer, the robots may be incapable of reaching the efficiency of a human.

Re: Key question

The amendment concerned does not say "no US citizen can be denied due process of law". It says that no person can be denied due process of law. The distinction is relevant because corporations are considered legal persons but not citizens. The debate over exactly what the personhood definition should mean has come to include a lot of things, but it's pretty clear that one of the most basic ones is that the laws apply to corporations as they would apply to individuals. Being in a corporation shouldn't mean that you can do whatever you want or that the government can do whatever it wants to you.

Whether you implement that with a legal personhood hack, by extending the language, or (as was often done before the legal personhood system) making a link between the people owning or running a corporation and that corporation and using their rights as the corporation's rights, you'll get to the same place. Other aspects of legal personhood are not necessarily included. The last approach breaks a lot, which is why it was replaced.

Former Uber CSO convicted for covering up massive 2016 data theft

Re: A fine is it?

I don't think anyone will argue that there was terrorism involved in this hacking for money event, so I think the realistic maximum is five years. As a first offense and depending on how successful his lawyers are, he's probably not getting anything close to that.

Re: Uber. Again.

In their defense, it's more the same period, again. Uber may not be a great company now, but they were a really bad one in the mid 2010s. The fallout from that period, including this action from 2016, doesn't necessarily reflect on them today. They could theoretically have improved massively and this legal decision would still be required. I have no idea if they have improved or not.

Cyber-snoops broke into US military contractor, stole data, hid for months

Prepositional phrases in English may proceed verbs. It may not be the most typical order, but it is for most uses accepted by grammarians. Sometimes, to comply with their other rule of not putting the preposition last, this pattern ends up being more common in formal writing than informal speaking, where most rules of grammar are discarded in favor of the "it sounds right and I'm not going to complain about it" principle.

Re: Given the mentioning of two python scripts being involved, isn't the solution to disable python?

Given your mention of Cobol, I don't know if you're being serious about blocking Python, but it's an argument I've heard before from people who definitely intend to do it. It doesn't work.

Of course, blocking Python would prevent some infection. And execution of any code outside the Windows directory. And inside that directory. And in the bootloader. I can provide you perfect security in this vein using the fail-safe security tool known as a blowtorch.

If you disable every function of a computer, it becomes a lot worse for doing useful things. Maybe nobody uses Python themselves, but there are still applications written in it which disabling every copy of Python will break. That's not a realistic way to block malware because it's a lot easier for the malware writers to port their script to something else or hide their interpreter than for the average user to get around a block that prevents them from working. They clearly didn't think they needed to hide their tooling this time, and they were right, but if it turned out they needed to, that's a day's effort for one programmer and your efforts to block Python are circumvented.

Block this: Using satellites to plaster ads over our skies could work, say boffins

Re: Come back Soviet Union, all is forgiven.

They had ads, just not for the same thing. Ads that are only for the government don't make them better, they just make the list of people you hate for putting them up shorter. Another reason it seemed like there were fewer of them is that a lot of the places to put ads weren't available yet. There also weren't many ads on the early 1970s internet, and the Soviet communications system didn't get much more advanced than that until modern Russian internet. Admittedly, I haven't watched a lot of Soviet television, but if they operated like many of their allied countries, there are lots of interruptions from the government-approved news for even less factual stuff.

I think that the Soviets would have eagerly accepted the chance to put some propaganda message in the sky at various points in their history, assuming they had enough funding or could get some other benefit from the required research. A lot of their space activities had propaganda goals as well, so it wouldn't be out of character.

China upgrades Great Firewall to defeat censor-beating TLS tools

So, if I'm understanding your claim, you sent out a DNS packet (which doesn't identify your machine) inside China, then moved the computer outside of China, and it was blocked? There doesn't appear to be any mechanism for them to identify that computer if they wanted to block it, as the DNS packet only contains the IP address of the requester, which would have changed if you left China. Unless I'm misunderstanding your testing, I think you may have misinterpreted the results.

You thought you bought software – all you bought was a lie

Re: Modern printers....

"Progress, yes, but a death knell for all the still serviceable older inkjet and laser printers once this becomes the only way to print."

It should be exactly the opposite of that. As long as the format to be printed is known, then you can just write a shim that translates that format into either a bitmap or bitstream read by those printers. That's no harder to do than the current driver, in that if they update the driver, it will do it for you, and if they don't, you have the same emulation options you would have had with the out-of--date driver. Most things can benefit from not having hardware-specific drivers if feasible, because a standards-compliant device can be made to work with almost anything.

Re: @Ian Johnston - Implicit in the article, but not explicitly stated:

If you can achieve similar results for similar effort using Gimp, that's great. If the worker doesn't think they can manage that, it makes sense that they don't want to try doing so for reasons they think are invalid. You can always fire them, but you'll probably have more problems finding people who have experience or the desire to use the tool you want if it's so niche.

Take programming. My employer could ask me to write in a number of languages and I'll accept. If I don't know the language, I can always learn it. If they tell me that our project's to be written in Apple II integer basic, though, I'm not likely to put up with that craziness just because someone issued an edict. After attempting to convince them otherwise, I'll decide whether it's worth leaving not to have to do that. They're likely to find that most developers don't want to use a tool that, while technically capable of the job, is painful and unproductive for the task at hand.

Re: "You own, at most, a serial number"

The URL contained in an NFT isn't hard to copy. The thing that makes it unique is the private key attached to it which is the only thing that can transfer the "ownership" of that token to another person with their own private key. While that private key remains private, the contained element can be attached to a single identifiable owner who has the exclusive right and ability to sell their token. The included token happens to be worthless, but it is a uniquely theirs worthless thing. Now if only that could be extended to do something useful (spoiler, it can't, but there are people who will pretend otherwise if you let them).

Elon Musk tells Twitter: My takeover deal is back on

Re: Talked the price down

So, if I understand your hypothesis correctly, he made the price decrease so he could buy it cheaply, then he could execute a sale at the original higher price, so he gives more of his money to himself. And because he decreased the price so much, he now gets to pay even more of that money in capital gains tax. And he bought stock through proxies to avoid it being obvious to other investors, which by the way is a crime, so he'd have to find another way to get the money he gave to himself via illegal proxies back into his bank account without tipping off either the market regulators or the tax authorities.

I think maybe he had a different set of reasons.

DoJ ‘very disappointed’ with probation sentence for Capital One hacker Paige Thompson

Good questions. I'm going to take them out of order, though.

"What exactly is the judge's role, and who are we to second guess their decision?"

The judge's role is to look at the evidence and the law and assign an appropriate sentence, keeping in mind that the law may state sentencing requirements or recommendations that limit their power. We are not only worthy of second-guessing that decision, but it is meritorious for us to do so in our role as citizens. We don't get to countermand the decision, but if we think that the sentences are consistently unethical in either direction, it's a thing that we, through our democratic processes, can change by altering the aforementioned sentencing requirements in law.

"What is justice, and what, ultimately, is the point of it? How does it relate to courts and the penal system?"

That's the larger question, and I don't have a pithy answer to it. Part of it is ensuring that new crimes are not committed, by this defendant or by others. Inadequate penalties can produce bad results, but massive deterrents aren't perfect either. Some degree of equality in justice is important as well.

Re: Whose fault is it?

I strongly advise you not to test that idea you've had. I'd like you to learn that you're wrong, but you can learn that by reading it online rather than by spending time in jail for theft.

USB-C iPhone, anyone? EU finalizes charging standard rule

Re: By the time it is standard

They've got magnetic USB-C cables. You put a small metal part into the port, connect the cable with magnets, and the two can come apart more easily. It also puts less strain on the port. You don't need USB-M for that, and it allows you to choose whether you want magnetic or more firmly connected for each device. If you don't happen to have your magnetic cable, you can still remove the connector and use a standard cable, so you have the best of both worlds. The only downside thus far is that I'm unaware of a single standard for those magnetic connectors, though some of them do orient the pins the same way. If they want, the USB consortium could design that standard, and that would help.

Re: Lint Magnet

"Is it possible for the USB-C socket to incorporate some sort of silicone cover, or dummy plug, without breaking USB-C standards?"

Yes, very easily. Neither of these would have any problem with the standard. Nor is there a problem with the magnetic connectors that remain in the port and connect to another cable, though I'd like to see a standard for those connectors as well because they provide a MagSafe-like connector, which I value in laptops.

Laptops aren't on the list yet, they specified a threshold of 100 W above which devices can use a different connector, and changes to the USB-C specification now allow it to carry 240 W, which is usually enough even for those laptops. For three independent reasons, this is not a problem for such machines.

FBI: We tracked who was printing secret documents to unmask ex-NSA suspect

Re: "This genius is a doctoral candidate???"

It's true that they're not the same, though as this episode indicates he had neither, I don't think we can say it proves that.

On at least a few occasions, when someone contacts an embassy offering things and the judgement is that they're not worth bothering with, the embassy turns them over to law enforcement on their own. I don't know how often that happens, but more than zero. Doing that builds a relationship with the host country at least a little. It's possible that happened this time.

Re: Very strange

If you're doing espionage right, you don't tell your spymasters who you are. After all, if this guy had extracted files in a way that didn't involve printing them on NSA printers and didn't deposit the money directly into his bank account, the FBI wouldn't have known who he was. That's ruined if you give anyone enough information to verify the amount of debt you claim to have.

It feels weird to give advice for how to spy properly, but I'm going to do it anyway. If you're going to do it, you want to be as anonymous as you can be. If you end up talking to law enforcement instead of who you think you are, you don't want to be identified. If the country you're spying for decides that it wants to negotiate with the one you're in, you don't want your identity to be on their list of bargaining chips. If you end up regretting your decision to spy, you don't want the country you were spying for to have blackmail material on you (for example that you were spying). If they can verify the information you give them about how much debt you have, you've failed at this important step.

Re: Very strange

Even if you decided that "I have debts" is the right answer to that question, you don't have to tell them an accurate number. Having never recruited a spy, I don't know what answer they'd be most comfortable with. Pretending to like their country over yours is probably the best thing if true, but probably also the most frequent lie they get. A financially-motivated spy might be more likely to be caught by incorrectly managing the payments, which could be dangerous as well.

Tetchy trainee turned the lights down low to teach turgid lecturer a lesson

Really? Because one person turning down a brightness knob when everyone else was gone is too implausible? Maybe I'd be inclined to believe you if the recovery included a long story of how flummoxed the victim was, but it seems they figured it out pretty fast.

I don't think they went looking for someone, as what would they do if they could identify them? Anyone could have done that with five seconds, and could have done it as a prank or out of irritation. Nothing to be done but fix it and move on.

Re: Hand written notes.

There should be some kind of rule prohibiting a professor from writing their own textbook for a class. If the professor's bad and the book is too, things are not good. I had a textbook written by a professor at my university. He was very proud of it, but I didn't find anyone who thought it was good. The only reason I learned anything was that a different professor was teaching the class that term. The responsible professor had already done his damnedest on a different course, so I know he couldn't write course materials that any student could understand or learn from.

This became awkward (just for me, but I certainly felt awkward) when I met him as a possible research advisor. He was nice, seemed happy for me to join his team, and was researching something I found interesting, but I ended up going to a different team just because I'd spent so much of a previous term hating his textbook.

Re: Hand written notes.

I don't know about the form of writing, but I agree that taking notes helps with learning. That doesn't prevent handouts being useful. The notes the student writes should be their judgement on important content from what's being said, not a copy of information that they won't get to see again. If just having the lecture notes isn't enough, which it usually isn't, they need to note extra things in whatever structure they think best helps them to study. If just having the lecture notes is the same as the lecture, then the lecturer isn't useful.

Re: Old School

"plus, it forced you to pay attention, rather then to think you'd have the printouts of the ppt so you can sleep through the lecture ..."

Usually, the slides or notes are aids for structuring the lecture and providing a summary, not the totality of what you need to learn. The stuff the lecturer says, questions they ask, and occasions for student participation can often include a lot that's not on the prepared documents.

That is if the lecturer bothers to do those things. If they come to class with the idea that you can just read their PowerPoint and know everything, then they're probably not bothering to teach otherwise. If they're doing that, you might as well get some copies of their slides and fire them, as the students won't be disadvantaged. Forcing students to manually copy down text from a screen not only doesn't teach them, but it does waste their energy if there is anything useful going on in the lecture. We've automated the process of copying text. Students don't need to do that anymore.

Big changes coming in Debian 12: Some parts won't be FOSS

Re: The installer

"Question: How is a user's honour affected when the same firmware is loaded into the co-processor from a ROM baked into the hardware rather than from temporary or removable storage?"

I'm not sure where "honour" comes into it in the first place, but one risk with having blobs loaded at startup instead of loaded from ROM is the distribution rights. If the hardware manufacturer includes a license either restricting or putting conditions on who can distribute the blob, then an OS that includes it could be responsible for following those terms. That could include things like not allowing it to be used in certain industries, which is against the guidelines for free software and which the distro maintainers don't have any logical way to prevent anyway. When the blob is in ROM, then the distro provider doesn't have to follow any license as they're not distributing any code covered by the proprietary license.

Text-to-image models are so last month, text-to-video is here

Re: Willis: not "for any future movies", only for that one ad

It's not a great idea to use anything you leave behind as authentication for critical things. That includes your fingerprints and your DNA. If I can get a copy of those by being near you for an hour or so and monitoring what you've touched, it's not good for proving your identity for a one-time thing. It works a bit better for something where you provide new samples each time. For example, using a fingerprint to unlock a door or device makes more sense than using one to sign a contract, but there's a risk for both so if it's really important, don't even use it for the door.

Samsung’s Smart Monitor tries too hard to be clever

Re: If you want smart, put an operating system on it

There's a reason the suggestion started with "If you want smart", though maybe I should have put some quotation marks around smart. I would also suggest a simple screen that does nothing on its own, but if a manufacturer thinks there's value in having functions that don't depend on any external device, they could still do it better.

To be fair (possibly too fair) to them, there is value in doing it the other way around. I know some people who have iPads and value that they can use them as extra monitors for Macs when they're not being used as tablets. As I understand the feature, it only works with Apple systems (and I'm guessing occasionally breaks there too), so having a tablet that doubles as a monitor without the manufacturer-specific limitations could be useful to someone who already likes tablets. This restricted environment device, however, lacks the advantages of the tablet part and doesn't do anything to justify its increased price.

If you want smart, put an operating system on it

The title summarizes my suggestion. If a company wants to make a monitor that's much more than just a screen, they shouldn't do it this way. Putting their own locked down pseudo-OS on the device means it will require work they're probably not doing to get the included applications functional. There's an alternative: use an existing OS stack, add software for any special features, and let existing applications do things the manufacturer doesn't want to write and support.

If this was a big Android tablet that could be used as a monitor, well I still wouldn't want it, but at least the Office365 integration would probably work pretty well as it could use Microsoft's existing code which runs natively, supports multiple windows, etc. Similarly if it used a full Linux installation (I'm guessing there's a Linux kernel in there anyway, but not the typical layers above it), then you could use a browser that's better supported which would also support multiple windows. Whichever way they went, it could also be customized by the user if they wanted it to do something else.

The open internet repels its most insidious attackers. They’ll return

Re: We're looking at the wrong problem

It's not really about one person, but about the string of one-person decisions that could follow. Take a similar example at a smaller level. This paper has described over the past few years the history of the UK's domain registry, Nominet. They completely changed from an organization acting for the public benefit to one designed to line the pockets of their directors. Similarly, we also had the fight over the future of the .org registry. That one didn't get that far, but it was close until a government started investigating it.

In each case, the number of people required to start the process of corruption in these institutions was small. Pointing to a single villain who single-handedly did the deed isn't possible, but the list of people responsible for keeping it that way was at most in the double digits. A few people saw an opportunity, found some others to do the work for them, and took advantage of the fact that people don't pay that much attention to the inner workings of something that seems small and technical until it's too late. It took thousands of people shouting for months or years to move the needle on either of those things, and that was against two relatively small corporations without government backing.

The ITU couldn't become corrupted with the election of one director, but it probably could with the election of about twenty. A lot of governments don't pay that much attention to what such a minor bureaucratic agency does, but they do accept their regulations. Unlike Nominet, there is no process for calling an emergency meeting and kicking out ITU members when the extent of their goals becomes concerning, nor can a single government threaten them into paying attention to public desires. Right now, this isn't concerning because the ITU doesn't have a lot of power over the structure of the internet, but that's due to custom, not law. Do you trust your politicians to understand why that's a good thing?

Steganography alert: Backdoor spyware stashed in Microsoft logo

Re: Trusted hosts

It's not about personal trust, as nobody was sent to GitHub to retrieve this file. Any person who trusts any file they get from GitHub has a very bad security posture. It's about what sites set off alarms, get blocked, or even get flagged as unusual on automatic filters. Most sites don't have GitHub in their filters of suspicious domains. There are probably other sites where uploading an image is possible and won't be blocked automatically by the traditional filter lists.

Digital Ocean won't let new customers create resources in four DCs, won't say why

Re: Energy prices?

In all cases, they have another datacenter in the affected city where you can start up new things. It's possible that, for example, Amsterdam DC number 1 is efficient and number 2 isn't, but it's not about larger geographic locations. In addition, they've only frozen one European DC and three in North America, but most energy prices have increased more in Europe.

I'm not really sure why one chooses AMS1 or AMS2, so maybe they had too many people selecting one and need to get customers to select the other one. Or they just don't want to take the risk of scaling because they're cautious about the future demand but they're already close to limits.

Fixing an upside-down USB plug: A case of supporting the insupportable

I had a device which decided to only use friction to keep the card in the slot. That is not a good design if you expect the device to fall to the floor. Micro SD cards are wonderfully designed to bounce into places that seem unreasonable (when you find it a week later because you gave up after eight hours of looking where it was logical it'd be).

Putting contacts on both sides would not only require the card to get thicker and expose it to twice the surface area for damage, but it would also need the designers to either make two orientation corner cuts or remove the one they have now. On the surface, this doesn't seem like an issue, as that cut is there to prevent the card being inserted the wrong way up, which is no longer a thing. However, they're also useful in preventing the card from being inserted backward. Unless, that is, we should have four sets of contacts so that it can be inserted on any side and in both directions that fit the dimensions.

Maybe we should only have circular devices that have identical systems on each side and must be inserted like CDs in a tray or spindle. At least then, no user has to look at which way up they're holding it. Knowing users, though, they'll find a way to break something anyway.

Reverse DNS queries may reveal too much, computer scientists argue

Re: "For devices on, say, university LANs that are assigned public IP addresses"

That is possible, but the current theories are:

1. Brian bought an electronic device after a holiday where buying things is common and stores frequently run sales.

2. A new person was hired to start immediately after a holiday and not in line with the schedule normally used.

Both are possible, but one seems more likely to me. Of course, there are other possibilities, such as Brian had a celebration at which he was given a phone as a gift, Brian received a loner device after breaking his previous one, and someone overheard the researchers and set up a plot to confuse them.

Re: RTFM

That's not failing to follow the spec. It's expected and shouldn't surprise anyone. Hosts lose connection frequently without having sufficient notice to send that message. If I disconnect my computer's cable, do I go through the network configuration to release my address? When I walk out of WiFi range with my phone with me, does it know that it's going to a new place, rather than back into range, so it can use the waning signal to release its address? If I have a power failure, have I included a bigger backup battery so that, while it's syncing disks, it can also clean up its network assuming the network device wasn't also in the power failure? In personal usage, the occasions where it's feasible to release the address are dwarfed by those where it's not.

Arm founder says the UK has no chance of tech sovereignty

Re: The UK these days, and for decades

No. The value would be between the two, and if they manufactured their own chips but kept all of ARM's existing business around, it would probably still be mostly IP. ARM's cores are manufactured by so many people that just having some fabs wouldn't make that their main business. If they forbade anyone else from making them, then it would still be a lot of IP, because their chips' value would be in the architecture they created. Places like TSMC are mostly fabs without any chip design IP. Places like Samsung have some of each, but Samsung has more fabrication capacity than designs for CPUs (other chips may vary).

If we insist on using a simplistic "agriculture, manufacturing, services" trichotomy to categorizing all economic activity, ARM's work is services. It's not a great categorization system, if you ask me. We could try a new one based on creation vs manipulation, but there would be a lot of definitional questions, so maybe it's easier not to.

Re: Digital, Culture, Media and Sport

Forty years ago, it was obvious to you (I would say us but I wasn't around then) that computers would be critical because we liked them and knew a lot about them. We're not talking about those people. We're not even talking about the general public. We're talking about politicians, who display with some frequency that they don't know how technological things work at multiple levels. They were one of the last groups that had anything to do with it to realize what technology would come to mean after the "expensive metal stuff the defense and intelligence people ask us to fund" phase. It's pretty clear that when DCMS was organized, they weren't expecting that department to spend a lot of its time with details like what architecture and manufacturing systems were used for processors.

Uncle Sam to unmask anonymous writers using AI

Re: If you want to analyse my messaging, there's a tiny problem for you to overcome first.....

Yes, there is, at least for those people you want to see anything. You realize that this surveillance only is relevant if they have a copy of your messaging. If you encrypted it for private consumption, you're already fine. In fact if you didn't encrypt it but sent it for private consumption, you're still fine (either they didn't intercept it and don't have a copy or they did and know where it came from). This software would be used on something you released publicly, and encrypting something you publish just makes it difficult for anyone to read it.

Summary: encryption and language-anonymization are for different things. Having one when you need the other is not helpful.

Re: Defeating Computerized Author Attribution

Scroll to the next paragraph for what I really wrote.

The method you describe is great for hiding source text in programs created here, but doing so requires a lot of negotiation with other needs. The interpretation of these ideas does not seem to be easy to understand, and the people who propose them may not have a clear understanding of what they mean. Readers who don't know writers who do this are easily overlooked because they don't know how to write a series or how to read the news. As for author privacy, it only works if the author does not use offline translation software. Anyone with access to Google Docs has a unique fingerprint because they can identify someone who uses Google to translate. Unless Google decides not to release this data to the government (good luck), it's a more confusing combination than natural language.

The method you describe would work very well at disguising the source of text from the program being built here, but in accomplishing that goal, it would end up compromising too much on other requirements. I'm guessing the translated version of this comment wasn't easy to understand, and someone who produces this result is likely not to be clearly understood. Readers unaware that the writer has done this could easily dismiss them as being unable to write coherently, which might cause the information to go unread. For the privacy of the writer, this also fails, at least unless the writer only uses offline translation software (having tried some of the options there, I can say with confidence that this would be even harder to read if they do). Anyone with access to Google's logs can identify the connection that performed these translations and it's Google, so they have a bunch of fingerprinting techniques. Unless Google decides not to give that information to governments (good luck), it could end up being a stronger piece of evidence than the original fuzzy match from language alone.

Intel's 13th-gen CPUs are hot, hungry, loaded with cores

Re: On demand options

At that point, you don't have many options short of getting some lower-power hardware between the socket and the power supply so that it too can be toggled remotely. I know there are at least some devices like that that can be purchased, and it's a relatively easy thing to DIY if you like those projects, but at that point it's probably worth asking if it's worth the effort to have two remote-wake systems for power supply and computer itself.

Re: Seems an odd choice for 2022

My theory is that they're listening too much to people who like treating the processor market as a religion. For a while, AMD's been touted as having far superior parts. In some ways, they did; AMD was able to put a ton more cores on the chips and clock them high enough to get better benchmark ratings. AMD's chips were also logical choices for cases where high-performance machines were needed. For the average user machine though, AMD versus Intel wasn't that important.

I think at this point, Intel decided that, instead of sticking to producing chips that are functional but unimpressive, they had to take that benchmark trophy off of AMD. Their manufacturing methods aren't in line with the TSMC ones that AMD is using, so while they're trying to fix that, they just pumped up the power usage of their chips and added cores to get their numbers up. AMD, seeing this, decided that they also needed the benchmarks up, so they've done it too. In the end, anyone who puts the flagship chips from either company into a machine doing an average workload is likely to find they've wasted power and money.

Like computers or mobile phones, companies can't resist the idea that, even though their mid-range products from five years ago are fine for most people, they can get a lot of interest if they just turn some number up high enough. That might even work if they paid attention to the things that come along with the advancement. It hasn't worked for the phone manufacturers who have been playing the game of how many megapixels can you get from a camera sensor so it can throw most of them away, at least partially because they keep raising the prices. I doubt it will work for CPUs either.

Re: On demand options

You can disable some CPU cores, but in most cases, it's not saving much if any power. Modern processors already put cores into an idle mode when they're not in use. Many don't have a facility for turning off the power to a core directly, so if you disable one from a cluster but continue using the others, it will just be permanently in that sleep mode. Many of the ways to keep a CPU's power usage down involve limiting how it responds when coming out of idle, for example not allowing it to clock up as fast as it can to limit peak power consumption. Having tried and mostly failed to find good statistics on CPUs' idle power draw, I understand that this answer isn't necessarily useful.

Depending on the idle draw of a more powerful machine, it could be acceptable to leave it operational with little or no load. If that's not good enough, you might consider having a remote wake function (depending on your hardware, there are a few ways to do that). That way, you can start your powerful machine whenever you need its power and shut it down when you're able to use your laptop alone, with the cost that you'd have to wait for it to boot. You can speed up that process by using a sleep or hibernate mode.

The web's cruising at 13 million new and nefarious domain names a month

Re: Day Old Bread

It's probably not related to what you meant, but I would suggest treating any domain as probably malicious until it's existed in DNS for a week (that data is publicly available in whois). I see so many phishing or malware setups use fresh domain names that they intend to run for a few days and cancel with their registrar. There are registrars that allow for refunds if domains are canceled in a short period, so they get their endpoints for free. I usually argue against blanket-bans of stuff, but this one is an exception as almost all legitimate sites are set up with enough forethought that they'll have a domain a week before it goes live for the public.

Soaring costs, inflation nurturing generation of 'quiet quitters' among under-30s

Re: Reality Check

The statement about employers not being able to pay in line with inflation isn't always wrong. Yes, there are many large companies who have a lot of profit and could afford a lot of things. It's easy to consider those alone because they're run by wealthy people and could easily sustain a drop in profitability without too many consequences, with the bonus that they're often unsympathetic. The world of employment is not only made of those things. There are a lot of small employers who don't make a massive profit already, whose owners are not very wealthy, and whose costs have been rising already. They can only raise the wages they pay so far before they'll have to raise prices as well, which will in turn increase the cost of living, requiring another pay rise. There are also large companies like that, so depending on the industry, just avoiding working for a small business is no guarantee of having someone who could theoretically afford a full raise.

When wealthy people pretend they couldn't possibly give people a raise to help with rising prices, it's troubling and obviously false. There are similar risks in assuming that, whenever there is economic suffering, there's someone available who could provide that and is only not doing so out of greed. When we have both of these arguments, we turn a situation where there is a potential for a lot of harm into a fight where managers complain about the "greedy, lazy workers" who complain about the "greedy managers", but nothing else happens.