Posts by doublelayer

A right Royal pain in the Dallas: City IT systems crippled by ransomware

Re: One Network...

The only explanation I can think of is that all of these things involve financial actions. The water department systems that aren't working are related to bills and the court system needs to collect fines and pay jurors, and many other parts of the government will have to handle some financial matter. Maybe all the financial stuff is run by one part of the city bureaucracy which has a common network. That's not necessarily a good idea, but it's also not that unusual for systems run by one organization to have less internal separation than would be desirable.

New York AG offers law to crack down on backfire-happy cryptocurrencies

Re: Ban it all

"You would delete the entire ransomware industry practically overnight."

I don't think you would. Even making paying the ransom illegal tends not to completely destroy the ransom business model, but at least that one would directly act against it.

I know that ransomware today nearly exclusively uses cryptocurrency to transfer the payments, and if cryptocurrency had been banned a decade ago, it might have had an effect on the small upstart criminals who started it. We're not in that world anymore. Modern ransomware organizations are quite large and operate like companies, with such things as a helpdesk for victims who have paid, translation teams for global effect, and for some relatively unwise groups, a PR wing. They can manage that because they sometimes get really large payouts. This also means that they have a lot more reason to want to keep up a successful business model and that they have the resources necessary to find some other way of getting money from victims to them which is not banned. It would be great if solving ransomware was that easy, but big problems rarely are.

Ex-Uber CSO gets probation for covering up theft of data on millions of people

Re: I think we've spotted the problem

I agree that a harsher penalty is desirable, but I don't think destruction of the entire company should be the goal here. The people who committed the crime should suffer more than the rest of the company, and we should try to have a penalty that acts as a deterrent to committing the crime, not a deterrent to anything happening at all. That means that the people at fault get most or all of the penalty, and those who had nothing to do with it are spared.

You may not care about this company, and I'm not that bothered about them either, but would you feel the same if some executive of a company you appreciate did something similar? It's not unusual for people with power to do something unethical that results in legal charges, but do you want that to mean that the rest of the work done by the company gets hammered down, even when that work was likely unrelated to the crime? My guess is that you're trying to prevent the cases where an executive commits a crime and the company gets a tiny fine which doesn't change their practices, and I think we can more easily deal with that by punishing the people who broke the law personally rather than by punishing the company, whose bills the executives won't be paying, and letting the responsible parties off easy.

How to tell an AI bot wrote that scammy-looking tax email: No spelling mistakes

Re: Another advantage of not being in the USA!

This is incorrect. Taxes are also deducted before people are paid, as is quite common because countries like having the money now, not asking for it later. The software companies lobby for making the tax preparation process more difficult so their software is needed, but the money from wages has already been withheld.

As far as I can tell, neither the method of paying taxes from wages nor the complicated system for preparing documentation has anything to do with how the tax authorities communicate with taxpayers. However your country does it, the communication method is subject to fakes. Whether you're used to getting letters on paper, calls, emails, or anything else, someone can try to imitate the method they used, or as many scammers do, just send something that some people will believe anyway even though it's not how it would normally be done.

Re: How is this a new threat?

I've heard a theory that at least some scams use the poor grammar as a method of filtering out people who are likely to smell a rat later in the process. If they have to expend manual effort on people who believed the email but don't believe the phone call, they might end up spending more on executing unproductive scams than the people they lose. I doubt that is the reasoning for every scam, and it may not be that common, but it is a plausible hypothesis.

Of course, people who have specific targets or a more automated attack plan don't bother with that, so there's no guarantee of such easy tells.

OpenAI's ChatGPT may face a copyright quagmire after 'memorizing' these books

Re: Odd how the copyright problem gets swerved.

Yes. Usually it started first with music, with the music publishers being certain that every advancement in technology would mean a complete loss in music revenues. Most of the time, they were mostly wrong, but they've had several occasions where they had a point, such as the first music sharing services on the web. They sued those services, and they won, but it didn't happen in a day. It didn't happen that quickly even when it was a clear case, such as Napster and similar services which were effectively Piracy Inc, with an easily found corporate entity running it. OpenAI has a lot more arguments for why their thing is acceptable which Napster never did, and although I don't find those arguments convincing, they'll likely need to be tested by lawyers for a while.

Re: Odd how the copyright problem gets swerved.

"Would a publishing company sue a person for being well read, and being able to quote literature?"

No, but a publishing company would sue a person who was well read, able to quote literature, and used those skills to write, sell, or give away copies of their books. The thing that would make their actions illegal is the distribution, not the knowledge.

"Would a record company sue an artist, who was deeply immersed in their field, and could play any piece of music from memory?"

No, if they played from memory to a group of friends. If they performed those songs in public, or if they recorded themselves playing them and sold the recordings, a lawsuit is more likely.

Re: Is copying large amounts of text or images for training the model fair use?

The answer is both of them. It was copyright violation for it to end up on the internet, and it is violation when OpenAI makes a product that continues to redistribute the copyrighted work. It would not be a violation if you stumbled on the file by accident, but it is if you keep a copy for your own use, let alone sending that copy to others which is what someone can argue that GPT is doing.

In this case, OpenAI should have, and certainly did, know that there was going to be copyrighted content in their training data somewhere. I'm not sure they put much effort into looking for it or doing anything about it, which is probably not going to please the holders of copyright. The same thing has happened with Microsoft's GitHub Copilot bot which likewise copied a lot of code that has copyright and license conditions.

If it reads it to you in chunks, you can just write a script to keep asking for chunks and attaching them together. That would take about two minutes. Then again, finding a pirated copy somewhere probably takes four minutes, so it's not that big a risk. I doubt publishers will feel the same way.

Re: Stop anthropomorphizing computers. They hate that.

And when a computer "stores" some data, isn't that anthropomorphizing it? After all, "store" is used for many operations, including writing some bits to volatile memory from the registers, which certainly doesn't count as putting something in storage. How about reading? A computer doesn't read in a human sense; it copies data from one form of storing it into a different form, or it takes a series of input events and stores them. Or how about "file". A computer doesn't use files in the sense that the term used to be used; it's a chunk of bytes that can be read or written in series. Except, of course, that every definition I've provided here relies on another of the words that represents a different human activity.

It's natural for us to use verbs for an action that a computer does that is analogous to it, and we do so all the time. For example, we frequently refer to RAM or sometimes storage as "memory" since it records data which can be referred to later, and the process of "memorizing" a book involves conveying that book into "memory" from which it can later be retrieved. Why shouldn't memorize describe that action?

Insurers can't use 'act of war' excuse to avoid Merck's $1.4B NotPetya payout

That's all true, and it's certainly no panacea. However, I hope that, if insurance companies decide they're still going to insure cyber risk, they'll eventually put some effort into figuring out which certifications of security policies are good at proving real benefit and which are not. That will only happen if there's a cost to them of not checking, and that will only happen if they have to pay a lot of claims from companies they've insufficiently checked so they fix that process. Maybe some insurance companies will decide that it's still a lucrative market and they'll hire some people to make a better method of auditing and certifying security practices. Or they'll decide it's not and they'll stop selling the insurance, which is also fine with me because then it goes back to being the companies' responsibility to pay for the results of their mistakes.

Which is not as pleasant, but at least they'll have to make things clear when you sign the contract rather than trying to slip it in. If a contract makes it clear that I'm not buying something, at least that means I can compare it to other clear contracts. The worst situation is when it looks like I'm buying something but I'm actually not.

That's why insurance should be more expensive if you don't do things correctly. That way, a place can still get some degree of protection from an attack, but they have an incentive to secure themselves properly to minimize the risk. Insurance companies should, if they intend to cover people for computer security events, audit their clients' security and use the results of that audit to set premiums. I think this process will work much better if the insurance companies have to do that for their own profit motive, and that is less common if they always find some excuse why they never have to pay a claim. So, in a complex chain, I think making the act of war provision more difficult for insurance companies to use will probably improve security of companies that choose to buy insurance.

Orqa drone goggles bricked: Time-bomb ransomware or unpaid firmware license?

Re: Offset Date. Device owner solution

You'd have to check whether that works anymore. If this is in firmware, it likely doesn't boot up far enough to connect to an NTP server. If the user had set it to an earlier date before the lock engaged, maybe that would have worked, but probably not now. Meanwhile, the temporary license that works until July might be set to not work before May, and if the person writing the code is any good at it, it might detect tampering with the clock and lock again, the way the 30-day trial modes on many shareware programs detected and probably still do if you tried to change the clock to extend the period. It's worth trying, but there are reasons it might not work.

Unlike your iPhone, Apple's batterygate controversy refuses to die

Re: No good deed

It looks like we're saying the same thing. Your summary is exactly what I think happened: "Most of the time it's a case of unintended consequences. What's more often the case is that after something happens the company will try to cover it up,"

I don't think Apple deliberately specified the wrong batteries. I called it a "design error" and that's what I think it was. When they discovered that they had built millions of devices whose batteries wouldn't work for very long, they went into panic mode and released software to automatically and silently reduce the hardware's capabilities so the batteries would last longer. Their changes were unable to make the batteries always work, but they prevented the crashes from happening even more quickly after purchase and frequently. Probably most of the desire was to avoid the bad press of three generations of iPhones keeling over after only a little use, but another factor would be the cost of repairing all those faulty devices under warranty. By pushing it out, they managed both to have the failures noticed after the warranty period and several months longer without people noticing.

This would have worked out quite well for them except some people noticed that they were slowing down the processors and assumed that was the plan. As I said, I don't think that was the plan at all, since their new processors were going to be faster anyway. The fact remains, however, that other devices don't have the kind of battery problems that these phone models did. Laptops don't tend to fail abruptly when their battery's a year old. Even the ancient ones I have whose batteries last twenty minutes just show the battery level dropping frequently, not claiming to have 90% and dying anyway, only to show that the battery really does have 90% when connected to the mains. Other iPhone models don't do it either, neither those before or after the faulty models.

Re: No good deed

"here was an example of a company trying to keep older devices working properly for a longer period of time."

That wasn't it, and it wasn't about slowing down the devices. They had specified batteries that couldn't run their devices, and they were trying to hide their design error. Their batteries were not sufficient to run the electronics they attached to them, and they figured this out, so they artificially hampered the electronics to avoid crashing their users' devices so frequently. They still ended up crashing, but it took a bit longer to get there, meaning that warranties and consumer protection timelines ran out before the problems became as evident.

The problem comes because the feature was turned on without people knowing and because the problems when the feature is not turned on is so bad. When a normal computer's battery is old and the maximum performance setting is set, the battery doesn't last very long because it discharges to zero quickly. When the iPhone's switch is turned off, or even sometimes when it's turned on, the phone spontaneously reboots even though the battery says it has, and actually has, plenty of charge left. I had one of those phones. I never disabled the throttling. Even with the throttling on, it got to the point where doing something as simple as answering an incoming call would sometimes require more voltage than the battery would give and cause the phone to shut down. I would have to plug it in to reawaken it, the phone would come up saying that the battery had 45% or so of battery remaining, and I could call the person back and apologize for the downtime.

Other devices don't do that.

Apple, Google propose anti-stalking spec for Bluetooth tracker tags

Re: Ah the days when you could take the battery out

"'no radios when flying' law."

The regulation is not that restrictive. WiFi and Bluetooth may still be used, as demonstrated by all the planes that have installed WiFi. What can't be used, and therefore what is turned off by the airplane mode switch, is cellular connectivity. There are separate switches for WiFi and Bluetooth which can be used to disable those. While it's possible they will be changed later to not reflect your setting, they do disable the features at the moment, which you can check with a cheap radio analysis tool.

"Nobody needs these tags, we all managed just fine before they appeared."

Nobody needs computers, we all managed just fine before they appeared. Computers can help someone who wants to kill someone else. Smash all the computers today.

You'll need better logic than "at one point we didn't have X, so X should be forbidden".

CERN celebrates 30 years since releasing the web to the public domain

Let's see if DuckDuckGo is affected by the "worship" you suggest must be there:

1. Computer scientists Vinton Cerf and Bob Kahn are credited with inventing the Internet communication protocols we use today and the system referred to as the Internet.

2. Bob Kahn, now at DARPA, and Vint Cerf, at Stanford University, published research in 1974 that evolved into the Transmission Control Protocol (TCP) and Internet Protocol (IP), two protocols of the Internet protocol suite.

3. The birth date of the internet is widely considered to be Jan. 1, 1983, but the road to creating it started long before the technology required for the internet even existed. [...] According to History.com, “the first practical schematics for the internet would not arrive until the early 1960s,” when Joseph Carl Robnett Licklider came up with a vision of using computer networks to communicate. [...] In the late 1960s, with the creation of the Advanced Research Projects Agency Network, which was funded by the U.S. Department of Defense, the “first workable prototype of the Internet” was born.

Does your theory expect that eventually the worship will insert TBL into the search results so he can push out all the people who are actually in the search results? While some of those articles do mention him, they mention him later than these other guys and they explain that what he built was a part of the system we use today, not the whole thing.

Re: The only reason that WWW ...

"It's not simply disagreeing with an opinion. It is decades of past experience demonstrating that a one-word answer to a multi-level problem likely indicates a lack of familiarity on the part of the party uttering that one word."

The post you replied to has more than one word, and was not entitled "A comprehensive review of the Gopher protocol's advantages and disadvantages". The post was not trying to encompass every detail of the protocol, whether it was good or not both objectively and in comparison, and how it was and is used. It was expressing an opinion, and the one word you chose to quote simply indicates that the opinion is negative.

Not only do you appear to have completely ignored that reality, but you assume that the person never used the system. Their negative opinion, with which you disagree, can't be because they came to different conclusions than you did, nor can it be because they were unfamiliar with something you used which improves the experience. No, you allege that they didn't use the system at all, and presumably just made up an opinion out of whole cloth. You dismissed an opinion based on no evidence at all, and in a way that doesn't make logical sense. I never used Gopher, and I have no opinion about it. I could try using the remnants that are still around and come to an opinion, but unless I do, my view on it is neutral. Their view is not, which is a lot more likely to have come from using it and not liking it than deciding to make up an opinion for no reason.

You've expressed negative views on systemd in other posts. Usually, you don't write an essay explaining each complaint you have about systemd's design or use, you simply say it's bad or you call it a "cancer". That's "a one-word answer to a multi-level problem". How would you react to someone who said "How to tell me you've never used systemd"? It would be incorrect and I'm sure you know that.

Their characterization, that he only wrote a browser, is wrong. Your characterization is also wrong. Of course it wasn't unheard of, but it also didn't exist in a format like that and the format that was created has remained useful in a form quite similar to its original form today, rather than being superseded.

This is similar to the development of the internet. You could make a similarly derisive, simplistic statement about how the internet wasn't revolutionary: "they took computers that used digital data and had them send that data, still in digital form, over a phone line. Big deal." That is inaccurate about how the internet works and why it is actually new, and it ignores that the internet is still in use today, without having to be reinvented. Nor do I worship TBL's contributions, just recognize that they are important to a piece of technology that is very common in the world today, and thus his work has proved important.

Allow me to introduce you to humanity.

Every major invention has been idealized by somebody, often the initial inventors. Often some of those ideals truly come to pass. Never are those absolute ideas met absolutely, because humans want to do bad things and they will use any tools available to do so.

It did make knowledge available to all, not every bit of knowledge, and not to all people equally, but there's a lot more availability than there used to be and it covers a lot more people than any previous system did. However, it enables people to post any information, correct or not, that they want. It's obvious that that was going to happen from the start, even if people hoped it wouldn't happen.

That's not unique to the web, the internet, the computer, electrical technology, or anything in particular. Any advancement produces some benefits, often the ones the idealists hope for but in a limited fashion, and some downsides. It's not even as if the downsides weren't predicted. Science fiction writers are good at coming up with ways that a technology completely breaks things, and although many of their dreams end up being unrealistic, others prove prescient.

"I just find it disingenuous they attribute it to one man who [...]"

If you want to argue about disingenuous, you might want to learn what that one man actually did. It wasn't an invention of every internet-related technology used today, and nobody, especially him, ever said it was. Neither was it as limited as you claim.

"created a GUI method to view web pages."

Yes, he did that. And a GUI to edit them. Of course, what is a web page? If you asked a contemporary, they would have to guess at the details, because he also invented those as well. HTML was his invention, without which the web page concept isn't formalized. And that's just the page part. What makes it a web page? Part of it was the transfer protocol which he also invented, although it's a pretty simple protocol so it's not a really important or difficult invention. Except that doesn't do a web either, because that's still a one-to-one connection with a server. The part that makes a web is the URI system to identify resources from other ones, thus enabling hypertext (which already existed) to refer to other locations without having to agree on interoperability first. That's a short list of stuff that he invented.

"Didn't even create a search engine": No, he didn't. Why did he have to invent a search engine when the way he enabled global hypertext is the reason that search engines tend to work these days? He doesn't get the credit for inventing the search engine any more than the developers of TCP would get credit for his work, but each step allowed the development and proliferation of the next to happen the way it did.

"or DNS. No DNS no http.": Wrong. No DNS a much more annoying experience using HTTP, but it can and has been done. Those protocols are independent and you are free to use one and not use the other.

"The internet is the WWW. Who separates the two?"

People who want to be specific and refer to them separately, although your next statement is a bad example of this.

"How many times have you heard someone say I'm going to check the internet when they really mean the WWW?"

Not a distinction that matters. What they mean when they say "check the internet" is that they'll connect to something that will tell them information. I say it, and often I do mean the web, but sometimes I mean that I'll use a different communication system such as connecting to a remote terminal, accessing a database, email, or an FTP server. All of those are on the internet, and insisting that people use a specific service name instead of the generic and well-understood term is useless.

"Beside "World Wide Web" is the technology behind HTTP. HTTP is just something that sits on top of it and translates text to pages after you use DNS. A GUI."

All of that is wrong. Starting from the end. HTTP is not a GUI. It's a protocol for transferring things. HTML isn't a GUI either; it's a language for specifying content that can be rendered graphically or not. The GUI is in a browser engine such as Gecko, WebKit, or Blink. The WWW is a collection of inventions including HTTP and HTML but not limited to those. In fact one of the most important parts might be the URI system that is used by HTTP resources but also many other systems, both ones that existed pre-HTTP and ones created later. DNS doesn't necessarily come into it, as that is an internet system that is often, but not universally used in HTTP or other WWW technologies. A URI that has an address instead of a domain is still a valid one. Systems that have their own naming systems independent of DNS and use HTTP on top are also valid ones.

They are related technologies. The WWW could continue if the internet shut down by using a different communication system, but since we don't have another world-wide communication system available to the general public, it would be a much smaller WWW and there would probably be multiple separate ones. For the same reason, I could complain about anyone who gets credit for the internet because you're just piggybacking on the existing inventions made for computers and communication systems, which would be similarly simplistic. The internet was an important invention, and HTTP and related technologies another one. Both are extremely important to the way we live life today, and fighting for who should get credit, or rather in the case of this argument who should be denied credit, is pointless and petty.

Re: The only reason that WWW ...

How to tell me that you never used Gopher without having to say "I never used Gopher"

I really like how simply disagreeing with your opinion means they can't possibly have any experience, and that they're lying to boot. Can you tell me the secret to making my opinion the right one, disagreements with which are universally wrong without having to know any facts about the person who disagrees? So far, all the statement says is that they don't like Gopher, not any reason why or any allegations of what Gopher did or didn't do.

Apple pushes first-ever 'rapid' patch – and rapidly screws up

Then turn it off. If automatic updates are off, then you won't install it unless you push the button. The "rapid" part of the name is just because it no longer needs to send a full OS image, so it's quicker to develop, test, and download. It doesn't make it mandatory if you have disabled that switch.

Did the patch need to be installed again after that? If not, then it sounds like the normal installation process, since the phone has to be rebooted while the patch is applied. You could have just seen it during that process.

Intel to rebrand client chips once Meteor Lake splashes down

Re: Too many SKUs

You're including a lot of generations in your count. They don't release anywhere near that number every year, so while you can still choose last year's chip, you don't have to go through that choice if you're planning to build with the most modern parts available. They also need to have chips for a wide range of devices with CPUs, from handheld units to servers more powerful than the average server, so they need to have a lot of options just for power and performance. When you add in that people often don't need the most performance for their power situation that can be achieved, it makes sense that they also add in versions of most ranges that are cheaper and less performant.

These factors mean that you don't really have to look at all those models when choosing something. If you're building a desktop right now, you can eliminate the chips that are too expensive for your use, likely most of the Xeons, and you can eliminate those intended for laptops or low-end devices like the Processor range*. Furthermore, you can eliminate many of the older generations unless you're buying used or recycling because, even though the 9th generation is still fully supported and is perfectly good, it's not sold very often by retail providers of CPUs. There will still be many choices, but not hundreds of them.

* The Intel Processor range: it still sounds stupid when I say this. I would have preferred them to stick to calling them Celerons or Pentiums. They could eliminate one of the names and consolidated, but eliminating both of them just makes a mess.

Re: Need to get away from these names

I explained so you don't have to be, at least for now. The algorithm is as follows:

Check first digit. If it's not a 1, that is the generation number. If it's a 1, append the next digit, and that is the generation number.

Parsing the rest of the digits is much worse, but I never said I could understand those digits. Basically, the best answer I can give you is that bigger number beats smaller number, something I can't say of AMD where the most important digit is the third because it represents the type of cores you'll get, but not the number of cores, so you pretty much have to go to benchmarks to figure out where you stand. Intel could certainly improve this, but I'm expecting they'll just create a new arbitrary system and the small amount I've managed to learn about their system will become obsolete.

Re: Marketing: why do we need it again

It's not so much that rebranding is usually negative, but that in most cases it doesn't do anything. Mozilla rebranding to Firefox was a legal necessity, but they could have chosen a lot of other names and gotten the same situation. Mac OS to OS X and back again did basically nothing. I'm sure they thought a new name would be good at indicating that they were now using NeXTStep, but if they had just called it Mac OS 10, people would have figured it out. I don't object to name changes, but I question whether the people who spend a lot of time and effort thinking of names before coming up with something rather obvious are very useful to the company or to the customers.

Re: Need to get away from these names

Intel runs a database that lists all the devices and gives you information on when it was released and what features are supported, so if you want to find those things out, you can always go there. I think the generation numbers have been generally easy to parse from the model number. It's the first digit or two digits, and they didn't have a 1 generation of these chips because they skipped from 0 to 2, so if it has a 1 as the first digit, read the first two digits. It could be better, but this is marketing, so it probably won't be. Of course, now they have a chance to start over and confuse things again.

What would that accomplish? Unless the actual performance is radically different, it will still mean the same thing to the user who's going to buy one of them. The I3/I5/I7 range has always translated to cheap/medium/faster, with the I9 representing basically just desktops and quite fast. The basic information can be quickly read in that two-character summary of where this is in the range, the generation number, and the suffix denoting how much power it's using. Why should changing the structure of the chip change this method of informing the buyer how it compares to the others? Most buyers don't care in the slightest how the thing is manufactured; either they want a CPU that's good enough, or they look at benchmark numbers and power usage information. Either way, the structure only matters as it changes those values.

Marketing: why do we need it again

This is the same thing they did a few months ago when they rebranded their low-end parts, formerly Celeron and Pentium, to just "Processor". That was probably a waste of time. The only time where a brand name change made sense was when they stopped releasing things with the Atom branding, having cemented the impression in anyone who cared that it translated as "so underpowered it isn't useful for anything". Calling everything Ultra won't convince people that it is in fact ultra, especially if it's 90% of the Intel parts available for typical computer purchases.

I get the point of some marketing people to toss around names for new things, although I'm not sure whether you really need any of the credentials that those people have since basically anyone can think up possible names, but I don't know if I've ever seen a renaming that had any benefits to the seller or to the customer.

BOFH takes a visit to retro computing land

That doesn't necessarily indicate technical knowledge, though. I wasn't around for those, but I did use computers not long thereafter, and I learned to get the more complex setup working. However, I managed that by reading and memorizing instructions and guides, not by having a lot of knowledge of why the settings had to be configured the way they were. The better understanding of what the components were doing when I issued commands only came later when I started focusing on understanding what was happening and why, not what I had to do to make it work.

Throughout my childhood, I assumed that someone who was good at using a computer must be really knowledgeable about how it worked. I knew a couple people, for example, who were quite at home with DOS commands and thus the Windows command line, and I assumed they must be technical people. Eventually, though, I learned that they were just people who had done work on DOS machines and that, although I hadn't, I had a better understanding of what was happening when they ran commands which is why they were asking me to fix things. For context, one of these people was convinced that a Bash session running on Linux must be DOS, no matter how I tried to clarify that it wasn't and that it was 2010 so they should know that a modern laptop running DOS was unlikely.

"If this manager gets all excited about a Soundblaster and CDROM then he has (had) much more clue than the typical manager."

Why? What does this prove other than that the manager knows what good 1990s computer parts entail? It doesn't necessarily indicate that the manager is good at using them, even if we assume that the manager has an intended use for them because they're not so unique nowadays.

I am younger than most of the people telling stories in these threads, which means as a side-effect that I have much less interesting hardware in my storage. However, it doesn't sound like the stuff we're nostalgic for, no matter our age, was necessarily the best equipment but rather the stuff we were familiar with. If I had a ZX81 around, it sounds like people would want it because it's a computer well-beloved by those who started with it, but it's not exactly the apex of hardware from the era or any real use today.

No more feature updates for Windows 10 – current version is final

"I will suggest to people to purchase a Mac if they don't need windows, or are unable to install a new hard disk and Linux etc."

Before you do, consider Apple's support lifetimes for operating systems as well. They're doing the same thing that Microsoft is doing. I have a 2014 Mac which is stuck on Mac OS 11. It can run Mac OS 13 if I hack it, just as a similarly old computer can run Windows 11 if the checks are bypassed, but Apple will not allow me to install new versions normally. When they have made that change, they immediately stopped releasing new application versions for Mac OS 11, in my case the latest XCode version, but it applies to any applications they've made. They also have been known to lag security updates for the old versions if they get to them at all, including critical priority ones.

Microsoft's generation of ewaste is annoying to me, but it's worth keeping in mind that a computer of the same age would continue to run Windows 10, with full security updates, for two more years while Apple dropped the machine 18 months back.

"My prediction is WSL2's internal distribution (CBL-mariner) will "seemlessly" become the main kernel for Windows and the efforts to maintain drivers will land on the open-source Linux community."

You wouldn't be the first to have made that prediction, and it doesn't make any more sense to me than it did before. Windows doesn't just slot into the Linux kernel. Its services are tied into the NT kernel, so stripping that out would take a long time. As for drivers, Microsoft does not benefit from making drivers part of Linux. If anything goes wrong, it is Microsoft that will suffer, not the Linux community, so they don't have any reason to do Microsoft's work for them.

Moreover, if that happened, it would probably be pretty useful for driver availability in Linux, at least at the computer level (ARM will probably still be relentlessly nonstandard, so no benefit there). If you are in the belief that Microsoft is evil, and if you're not there are plenty of others who are, then there's another great reason for Microsoft not to do this: if they did, it makes shifting users and applications to desktop Linux much easier, making it much more convenient to stop using Windows.

Re: Err

If you go through a doorway, you end up on the other side, but not necessarily anywhere else since you can stop immediately on passing the doorway. If we're using the analogy, that means they should be supporting it at 23:59:59 on that date, thus having completely supported it throughout the date they set, and then they can drop support from a grammatical perspective. Since they weren't clear, let's say that's 23:59:59 at UTC-12:00, so that it's universally considered to be past that date everywhere on the planet when support is dropped.

I've been using Windows 11 on my work computer and have installed it on my Windows image for my personal computer. In both cases, it's very similar to Windows 10. A few things have changed, but they don't affect me very much and nearly everything looks the same. There are some problems, such as Windows 11 having more Microsoft account stuff, but since I'm not running the home version, I could bypass that without hacks.

I don't administer it across others' devices, but just as a user, it's not causing me any problems I didn't see in Windows 10.

Dropbox drops 16% of staff, points finger at hard-up customers and AI

Re: OneDrive

"Eventually found the password and gained access to the files, but to change the email (or any settings) it had to send an email to the existing (and wrong) address."

To be fair to them, that's a pretty standard security feature, to prevent someone who has your password from locking you out of everything in perpetuity. Unless you have a second factor that is trusted to also change the email, they would have no way of verifying that you are who you say you are instead of someone who guessed or stole a string.

As for support, I agree with the sentiment, but companies rarely treat the people on the free trial program the same as ones who spend a lot of money. It's not that crazy when you consider the number of free Dropbox accounts they must have from years of offering the service; I have at least two of those, neither of which is used anymore. I don't think someone running a contracting business would treat someone they helped out for free a few months ago the same as their primary client for whom they've been working full-time for years, and Dropbox has a similar situation just scaled up to millions of accounts.

Cloud slowdown hits Amazon as orgs look to rein in cost

Re: F.U.D.

Anyone's itemized list will definitely lead to arguments. Anything that shows high costs of on prem will be derided by anti-cloud people as incompetent, whereas any pro-cloud person will argue that an expensive cloud bill is due to inefficient choices of which services are used to accomplish the goal. There's no winning, because in most cases I've seen they'll both be right. You can do both of them incredibly badly, and there are places who do.

In general, whenever I see someone saying that cloud is always better or always worse, I expect the person to be biased and unfamiliar with one, if not both, of the choices. There are some cases where there is a clear distinction, and many others where the difference will be determined by particular choices of how to implement the desired setup. I'm afraid I've seen too many uninformed people* with very strong opinions that rely on an incorrect or inexact assumption.

* Not that I'm particularly well-informed either, but I at least hope to be more careful about stating things with certainty if I haven't got the experience or rough calculations that defend it.

iPhones hook up with Windows as Microsoft’s Phone Link dials up Apple's iOS

Re: The push to Windows 11 begins...

Since they've already decided to release it, I'm not surprised when new features are limited to that version. Yes, they lied about not releasing a new version, but now that they've returned to doing it, it's predictable that they're using a pretty typical method of encouraging people to update, and one to which I don't object that much; selling a new piece of software with "it has new and useful features" seems fair. I object more to the generation of ewaste because of their unrealistic hardware requirements they're not using; if they allowed 11 to run on all the equipment where it could run, I'd complain much less.

Re: Does anyone use this stuff?

I have used it, but only a small subset of the features. The only part I found very useful was the ability to see and send messages from my computer. I'm faster when using a full keyboard to type messages, and there are some people who prefer to communicate by text rather than voice. That's not exactly new, but since SMS, iMessage, and Signal (no WhatsApp for me) have replaced IRC or other text-based communication methods that such friends used to use, it's handy to be able to use a computer to send those messages. It wasn't big enough to keep me using a Mac, but when I was already using one, I found the feature a bit helpful.

If you don't get open source's trademark culture, expect bad language

Re: Well, I'm not gonna use rust again for the forseeable future.

"Did you need any permission from the Linux Foundation to write that comment? Because you would need permission to write a similar comment about the corroded iron language."

That's not how this works at all. It's completely wrong in every detail. Trademarks don't prevent you from commenting on the trademarked item. Windows is a trademark, and Microsoft defends it. That does not now nor did it ever prevent me from saying "I like Windows" or "I hate Windows", or literally anything I want to say about Windows. What it restricts is me trying to sell some software and call it Windows. The same thing is true of the Rust and Linux trademarks, and Rust could have and probably will keep the trademark as they do now but should have described the usage differently.

The Linux trademark, for example, doesn't prevent me from using the word in a comment, or even in a product. It would, however, prevent me from making my own operating system, deciding that I'll call my OS Linux even though it doesn't work with real Linux software, and making money from that. It also prevents me from setting up the "I really make Linux company" and trying to get people to donate to or pay for whatever I claim to be building. I could still lie about being instrumental to Linux, but I have to use an unrelated name because otherwise the Linux foundation can defend their trademark. Rust is likely trying to do the same thing, but in an attempt to make things easier on their lawyers, they advocated a more restrictive usage policy for things related to the language. That was a misstep, but not because trademarks aren't compatible with open source. It was a misstep because they didn't work with the community and they had a lawyer, rather than a technical person, express the policy. Lawyers work in an area of "Don't do that or otherwise bad things will happen", whereas technical people tend to be a lot more open about what bad things and how close to that you can get before they start to happen.

Florida folks dragged out of bed by false emergency texts

Re: Wot no alert?

I'm not sure where they got those, but I wouldn't count on that being true. Living in a country that has used this system for longer than the UK has, it's been supported for years. I had an Android 4.4 device that was particularly annoying about it, because for some reason it cached an emergency alert and played it every time I turned on the device for the next month, but I blame the very patchy firmware for that. I don't think they invented a new protocol that would prevent all the old devices from sounding the alert.

Re: Running SIMless

Smartphones all let you do that, since they have plenty of other features that you don't need mobile service to use. Android and IOS handle not having a SIM pretty well. I'm not sure what level of non-smart phone you had, but if it literally did nothing other than calls and SMS, maybe the people writing the software figured bypassing that screen wasn't very important, or maybe it was possible to do so but just not intuitive as many of the cheap devices have had exactly thirty seconds put into making the UI understandable.

Re: Big Brother has another way to cock things up.

If it makes you feel any better, it's part of a standard which has included emergency messages for many years, and many countries have already deployed a system that produces them. The UK isn't writing new software to run on your devices; they're just sending messages which will be interpreted by the same software that already interprets them for millions of others who did the testing before it got to you. That leaves less room for the UK government to mess things up, and all possibilities are on the sending rather than the receiving end.

Re: Why would anyone leave these turned on?

This depends on what kind of disasters can happen near you. The one that I've found most convincing is tornadoes, since they can appear with relatively little warning, but you still get minutes from a detection to gain shelter. If you have tornadoes where you are, that might make emergency alerts useful. Similar things can apply to flash floods or anything that you wouldn't know about a day in advance but would know about at shorter notice. I've decided to leave mine enabled, and if they end up being a problem for me, I'll disable them at that point. So far, there has not been a problem.