Posts by doublelayer

Apartment living to get worse in 5 years as 6 GHz Wi-Fi nears ‘exhaustion’

Re: And Gamers...

"What's the experience of others? If you live in such a building, have you noticed lots of wifi congestion that's affected what you're trying to do?"

I have, but that was mostly on the 2.4 GHz band. Terrible congestion on all the channels. Fortunately, after disabling that, 5 GHz has been fine and I've had no need to go to 6 GHz, which only some of my devices would support anyway. There are probably people who are packed more tightly than I have been, and 5 GHz could be more congested for them.

One possible explanation is that I do a proper check of the available bands before selecting one. If I'm able to find a clear one because people are trusting to automatic selectors that don't work, then they may be getting more congestion than I see on the popular defaults. Compared to the way the study did it, I also don't set a massive band, going with the smallest option which has never been a problem. Mostly, that is the default, meaning they may have estimated more congestion than really exists by assuming that many more people will opt to use several channels at once when that is a relatively rare choice.

Sergey Brin promises next generation of Glassholes will be much less conspicuous

Re: I'm probably not the target market for these...

For some, the point of new tech is showing off that you have it. For others, the point is using the new tech. Most of the time, the people building the tech are thinking about the people who want to use it, hence why the original Google Glass team focused on ways you could use their tiny screen and ignored how it looked ridiculous to everybody. With this modern version, they'll pay a bit more attention to making it not as ridiculous because they already cut themselves on that once, but their focus will still be on the tech.

That doesn't mean the tech will work well. Often, people who focus too much on the tech build in something they think makes for a cool demo but ignore "boring" features that someone might want to use, or they decide that some moderately useful feature takes precedence over hardware deficiencies like insufficient battery life. Not every tech designer wants to or thinks they can make their product a luxury good that every user will be showing off.

Re: The problem isn't that they are conspicuous

Wrong comparisons. The thing to compare this to to see how the public response will go is Facebook's smart glasses. These may look different as it sounds like they're still going to have a screen, whereas Facebook's (fine, Meta's, but I still don't like letting them change the name) decided to skip that part. The important part for those worried about privacy is still there: Meta's got a camera in those. I don't notice people wearing those in public, although I'm not looking closely for them. That either means that, although seemingly popular by reported numbers, there aren't many users near me or, alternatively, they aren't as obvious as Google Glass was.

AI can't replace devs until it understands office politics

Re: The Great Replacement

It will listen all day long, but that doesn't mean the summary it provides us will be useful. For example, one of the ways to make sure the thing shows up in an LLM summary is to make sure it is mentioned many times. The most important stuff in a spec is likely to be the parts that don't need to be repeated multiple times when talking to users. So if the users talk to the LLM the same way they'd talk to us, that kind of important part will probably be the most likely part where errors show up. The talking to users part is important. That's why we do it. If all I needed to do was coddle some idiots, I wouldn't bother talking to them. I need to learn important things that they know and I don't, and if I don't, then I'm the idiot who could not build something because I was incapable of doing my job, of which determining requirements is a part.

Re: Coding isn't dead...

This is mostly wrong. The analogy is not to manufacturing things, but to designing them. We build machines to build cars now, mostly, but the cars the machines build are still designed by humans. Programming is not manufacturing work, since it's usually pretty easy to make more of the program you just wrote. Programming is design, and that has not been mechanized just yet. We build tools to speed it up, but so far, none of our tools has eliminated the requirement to think, and programming properly requires thinking.

Re: The Great Replacement

If AI can do that, AI can replace my coding job too. AI can't turn vague statements into a spec that people actually want for the same reason that I can't: I don't know what people want until I extract it from them through large amounts of talking. They could try talking that much to the AI and hope that the specs it makes after that process will represent that talking, but the talks I have tend to be much larger than the context window for the larger ones used in most models. Or mine, for that matter, which I work around by splitting up the task into pieces and writing the spec in pieces so I don't have to hold it all in and write it once.

I think the start of your comment is extremely correct, and in fact, more correct than even you think. People seem to think that AI will easily replace whatever jobs they have no respect for, often by misunderstanding how that job is done. A lot of people don't understand how programmers work or what they do that eventually makes the computer do something. The only part of that process they see is those funny documents with punctuation in weird places, so they think that's the most important part of it. You seem to have a similar view of management. Managers mess things up all the time, but that doesn't make their job as simple as you paint, and that means AI will not be as able to do it. As with ours, LLMs will do an excellent job of pretending to do that job if you give it to them. An LLM can certainly run a status meeting. It can't properly interpret that meeting and a long list of other meetings to understand what people are doing, which ones have the strongest skills in which (undefined and flexible) areas, and understand what people like, how to keep the ones you want, which ones you shouldn't have, then execute on all of that. Good managers can and do that, and bad managers will often do some of it.

UK 'extremely dependent' on US for space security

Re: I can believe it

You know you're talking with an ill-informed person if they use a phrase like "BRICS alliance". BRICS is a grouping of countries because they're sort of similar: large countries that aren't in a traditional alliance with similar levels of economic development. That's all it is. India and China are two members of those five, and they're not likely to ally with each other on military matters any time soon, what with the recent military disagreement (at least a hundred deaths). Brazil and Russia have no military alliance, don't generally agree on diplomatic issues, and have little significant economic relationship other than the normal one you'd have between two largish economies that produce different things. But that grouping for categorization and the occasional conferences between them, marred only by the fact that Putin wouldn't go to the one held in South Africa because that member of the alliance was liable to arrest him, can be used to imply that there's some powerful group where there are five mostly unaligned countries, and five sounds much more powerful than your favorite one.

Re: It’s special alright!

"And they didn’t do it because it was “the right thing to do”. They did it because their supply ships to Europe were getting torpedoed by the Germans and it was …yep …. costing them money."

That's...an interesting interpretation. Maybe they did it because they had declared war against Germany, which happened because Germany declared war against them, which happened because they declared war against Japan, which happened because Japan bombed them about which they were quite angry. Are you really suggesting that they only did this because ships were being attacked. How, in your interpretation, do you explain the decision, much earlier, to give the UK and its allies weapons at US expense?

In reality, there were a lot of people in the US who wanted to stay out of the war because it was a war and those things are not much fun to be in, without putting much more thought into it. That did not apply to those running the military at the time, who, when they got the public support they needed, were quite enthusiastic about joining with the UK to fight, and they did. This is not the only time the US has had almost exactly the same reaction to a possible war, nor is it the only country to have done it. I could, for example, rewind a few years and describe the UK's reaction to German actions that looked very similar. Doing anything they could to avoid war, which of course didn't work, and when war was declared, not being very effective for several months. To explain why would take a lot of paragraphs, and almost any single-clause explanation would be virtually guaranteed to be wrong.

Actors' union complains about Epic Games cloning Darth Vader

Re: A sign of times to come

And, once again, you have the opportunity to attack someone you disagree with and be convincing, because the union representing actors has no reason to control what can be done with recordings the users had permission to use from both the legal owner of those recordings and from the person (or family of deceased person in this case). Armed with this great opportunity, you completely ignore it and go on to bang your "copyright is bad" drum again.

A lot of artists do complain a lot about the threat of competition, and sometimes that's all they're afraid of. In my opinion, too bad for them. But some of them are complaining about their work being used without permission or legal right to do so, and I support them in this. I don't care if the studios use their computers to make fake actors. If they can manage it and it means human actors have a harder time finding work, that's normal change. If, however, studios try to clone the appearance or voice or gait of a human without permission, I have a problem with that. If artists lose to competition from an AI model that has legally licensed all its training data, I won't support the artists. Every model that's competing with them right now is trained on illegally obtained training material, and I support those who did not give permission for that use.

To progress as an engineer career-wise, become a great communicator

Re: Not the case elsewhere? US?

The downvotes are because it presents a straw man (you have to be knowledgeable about literature to be hired), when nobody suggested that. The point they actually suggested, that people be tested on their ability to communicate, is a good one, and elsergiovolador was clear that they disapprove of this, using unconvincing arguments to oppose it.

The kind of communication that needs to be tested is not particularly extreme. Can you write a readme for your code that, when read by a person who hasn't read your code, does these things:

1. Tells the reader what the code is for.

2. Tells the reader what else you need to use it.

3. Explains how to run it.

These aren't high bars. This is one of the few types of communication where typos aren't a fatal problem, as long as they're not in commands or change important meanings. You don't have to work too hard to accomplish it.

Some people can't do this or hate it so much that they refuse to. Some of them have convinced themselves that, because they can't, it's okay if they don't. It's not okay if they don't. In some cases, you might want people with even more communication skills than this, which is still not a high bar, but most of the time, the readme and similarly detailed notes and emails to colleagues is fine. A good grade in a literature class would probably demonstrate you had that skill, but you certainly wouldn't need that to prove it and could gain that skill in other ways.

Re: Reality check

Most of the time, we're not talking about being great at presenting a sales pitch or a viral public presentation, but discussing a technical matter with your colleagues. The programmer who can write the code and it works scores lower than the one who can do that and describe it well enough that a new colleague can modify it quickly instead of having to parse through everything from scratch.

And since you still have your fixation on managers being universally bad, increasing in badness as you go up the org chart, one major situation where this is important to the career of that person is whether they can describe what they did, are doing, or want to do to the boss without making the boss go through everything. Those who can summarize clearly, accurately, and with the right details included and excluded, make sense and are more likely to retain their jobs. In the real world, that's because the boss has a lot of things to do and their job is to direct and monitor, trusting the engineers to build what they decided needed to be built. In your world, you can put it down to managers being lazy. Either way, though, it means that the bare minimum policy you recommend includes being able to communicate well enough that management doesn't worry that your poor communication is a smokescreen for not being able to do anything. I have had the experience of someone using poor communication skills to distract from their poor technical skills. I think they actually could communicate better, but that would involve saying that they couldn't do any of the work, so they disguised that with terse and unclear statements. It didn't work.

Re: Not the case elsewhere? US?

That depends what kind of company it is, but most of the problems they're referring to aren't only encountered when programmers are told to do every job. You don't need to be talking to a customer for the need to communicate to be important. They can employ all the customer talkers they want who will polish up your message for the customer's enjoyment, but if your message is so unclear that the polisher can't understand it, there's still a problem.

If you're going to have the situation you describe where it is your company, then chances are that your company will be small and you can't afford to employ enough polishers to deal with unclear output from an uncommunicative tech team. In that case, you will actually need them to be able to do both of those things, and a tech who can write code but can't document will be unsuitable. Those companies who try to ignore this often find that they have a product and no customers, and it often ends badly from there.

Boffins devise technique that lets users prove location without giving it away

Re: How about no...

"As for battery life, are you saying that disabling GPS doesn't switch off the GPS radio?"

No, kind of the opposite. I'm saying that keeping GPS enabled doesn't mean the GPS radio is always on. I leave it enabled, but I know what apps are able to use it. If none of those are running, the radio is off and consuming no power, the same way it would be if I had disabled it. It'll turn on when one of those apps requests it. If I know what apps I've allowed to use it, whether those are using it now, and don't have something weird in the background also using it, then it works that way. If I didn't have all three of those things, I'd be more worried about not knowing them than about my battery life.

Re: How about no...

If you have WiFi enabled but aren't using it, it spends a little power looking for networks to connect to. If you have GPS enabled and aren't using it, it doesn't need to do anything. Unless there are background processes using GPS, the power explanation doesn't explain it.

On Android, you can use a few mechanisms to create shortcuts to enabling or disabling certain settings, and if you have a reason to want to switch location access on and off frequently, that might be the closest thing to what you want. I just suggest that you check whether your reasons are backed up by what actually happens when you do it.

Re: Isn't this something you could do with traceroute?

This is one of the reasons a lot of systems no longer respond to packets that have their TTL expire. True, the privacy aspect is probably not the strongest reason it mostly doesn't work, but it's one of them. There are ways of trying to geolocate by IP timing, but they aren't the most effective. Also, there are ways to fix that.

This is for a very different kind of location tracking, though. It's mostly aimed at mobile devices, which probably fall into one of these categories that make IP tracking even less useful than they otherwise would be:

1. They're moving, and their provider is giving them new addresses when they do. Mostly not what they do anymore but it can sometimes still happen, and it will look the same if they're frequently using local WiFi.

2. They're moving, so the mobile provider keeps changing their internal routing tables to move their address with them. Your tracking data becomes wrong on an unpredictable schedule.

3. Their provider doesn't have an address for them specifically because they're still using CGNAT, so your attempt to track them down fails because the closest point you can get is their shared address router.

The 'End of 10' is nigh, but don't bury your PC just yet

Re: Skill issue

"That is only 2 of the 4 essential freedoms;"

Thank you for that. Of course, I had never heard of them before. It's not as if I wanted to get to the point without copying an entire blog post. But I'm sure your clarification statement was helpful to somebody.

"Everyone deserves freedom even if they didn't have any freedom at all previously and therefore didn't realise it was a thing."

You didn't read closely enough. Read again. It wasn't that they "didn't realize". They don't care. In order to convince them, you need to explain, to them, not to me, because I already care, why they should care. And, when you've tried and failed as many, including me, have before you, you will need to take the next step and understand what they will care about. Of the four freedoms, three are almost entirely useless to someone who cannot read code and don't intend either to learn how themselves or to ask someone else to do it for them. To the average nontechnical person, your argument sounds like this:

You: This car is better than yours. You should switch to it.

Them: I like my car. Why should I use yours?

You: Because mine gives you freedom.

Them: It doesn't have a steering wheel.

You: You steer with pedals on the floor. You'll get used to it.

Them: How do you accelerate?

You: The buttons over near the glove box.

Them: What freedom does it give me again?

You: You're allowed to melt down the door panels.

Them: I don't want to melt down the door panels, and I couldn't manage it properly if I wanted to.

Until you understand their position, you'll keep telling them things about freedom which they don't understand, don't care about, or think you're deluded about. You need to correctly describe the deficiencies in the alternative, deficiencies that actually exist and don't take the form of "it doesn't have freedom", because they won't understand that and it won't make sense even if you explain it. Have you ever heard the people who protest things, but you can't tell, even after discussing with them, what they're protesting against or what they want to replace it with and you get the idea that they don't know either but they like protesting something? To people who don't already know what the four freedoms are and value them independently of your arguments, you sound like those people. We understand you, but nobody will realize what freedoms they don't have if you keep doing this. Break your loop.

Yes. I have used a lot of things. What I haven't tried to do is to switch a hundred people to any one of them with the hope that they won't be cursing my name tomorrow or requiring me to spend weeks helping each one out. If I decided I wanted to do the latter, I'd be looking for something that made their systems as standard and as easy to use as possible, which is often the opposite of what I look for for my personal machines. On those, they're mine, so I want them to work exactly the way I tell them to. If that means I have to do extra work to get around something my choice didn't build in, then I'll do that extra work. Sometimes, I specifically want to avoid something the more complex systems built in and I don't want. I know people who custom-built distros for their family, with everything nicely integrated and remotely administerable, and I can and have done similar things. That took a lot of work, and I can't and don't want to do it at a large scale.

Re: Not obsolete if you monotask singlemedia

In some cases, especially with nontechnical users, that's exactly how it comes to be a mess. Nothing was there, which means that everything that is there now was bolted on by me at some point, and maybe I've forgotten that I did something. That makes it harder to unbolt things that are causing problems because there's no simple uninstall button for a lot of it. For the normal usage of Alpine, embedded and limited-purpose deployments, that's probably fine. For handing it to a user who might want to do all sorts of things, it makes it a courageous option unless, and possibly even if, you're the only person who will ever be asked to administer it for them.

With each layer you remove, you increase the number of things you'll have to fix. Mint or Zorin are huge, but they have tested the configuration they release and made sure a lot of things are present and functional. Ubuntu figured out several of the things that Debian left up to the user. When it's my machine, I don't mind having to do a little custom-building, and I'd rather do a bit of that than have to deal with a suboptimal default that's harder to remove. That is not the case for a place trying to give unfamiliar people new OS installs.

If you're trying to convert people to Linux en masse, your most likely bottleneck will be availability of technical users' time. That means you want as many things to be easy to do without needing any tweaking. That means, for example, that if there's a piece of software that a lot of people want and only works when there's systemd, you use the bloody systemd whether you like it or not. The alternative is that you use the same time to convert three people, because most of it is spent trying to implement compatibility with the older init with that application, that you could have spent converting fifty people who could click on the app's name. Alternatively, you tell a lot of people that their system can't run that and they leave, annoyed. Alternatively, you explain to people why there's a difference and they grow confused about init systems and why it's bad that Leonard works at Microsoft and leave, annoyed. Your idea actually sounds closer to not invented here since you're going to end up redoing some of the things distros exist to do, just at a smaller and probably less maintainable scale.

Re: TODO

I missed the edit window, so to add one more point, I think one of the ways we fail to convince others to use Linux is that we don't understand what they want to do and insist on putting other things in front of it. It happens a lot in these comments. The archetypal example is the constant comparisons between PhotoShop, GIMP, and other photo editors. That's a frustrating one, since I don't use any of them and have no way to know who is right about whether they have the same features or not. In many cases, the points appear to be decided by ideology. I like open source, so the open source program must work, or I find Linux annoying, so the Linux-compatible program must have problems. Neither approach helps, no matter how confidently they say it.

I think a helpful resource could be Louis Rossmann's guide to an open source life. It does two things that are useful. It demonstrates the style that is needed for people who are trying to use the software but don't understand it, and it points out where our tendencies as developers or power users need to change. Some of the things listed in the introduction as reasons that the software isn't more commonly used affect all of us, but because we can use our technical ability to eventually work around them, we forget how big a roadblock that is. People who find it harder to experiment their way through will often stop there. If we want them to use the same tools they are, then we need to change our ways. To some extent, this might be the occasion to decide whether and why we want them to do that so we can take actual steps toward achieving whatever goal we decide on.

Re: TODO

"any distro you'd advise them to install has its GUI interface to the repository which they should be encouraged to use and has Wine in that repository."

True, and the Wine wiki I linked to has a bit of a recommendation about those:

Many Linux distributions come with an included Wine package, but due to Wine's rapid development rate these are usually old and often broken versions. It is best to uninstall your distribution's included package versions and update to the latest Wine version available here.

So when you give that advice, you'd better be confident that, should the user try it, your version is going to work. And you'd better direct them away from pages like that which tell them it won't. And we haven't yet gotten to how they're running Wine, and if they'll use one of the frontends around it or not, and which one, and how they configure it. So far, at best, we've installed it. Sometimes, making everything functional to people who don't want to think about complicated things is more complicated than using it the way we do.

Re: TODO

That is a point I made in a similar comment, and you responded to that one, so I know you already saw it. If you're going to make an age-based stereotype, saying that people of a certain age are familiar with computers because you're familiar with computers and are of that age is a very unconvincing way to do it. The fact that many people of that age on this site are familiar with computers doesn't help much, because that's true of literally all the numbers you might put in that have more than zero active forum contributors. I was pointing out that someone who worked on computers is going to have more experience, no matter how old they are, than the people who did something else with their time.

The stereotypes are mostly fallacious because most people have no or very little experience with desktop Linux, no matter how old they are, but if you insist on making them and saying things like "I'd expect there is more boomers that know about PCs than millennials", you have to consider the general public, not your friends who are about your age and worked in the same field versus the millennials you've met by chance. Similarly, you've described one type of experience with computers, the person who built things on all the generations and knows them all inside out, and those people definitely exist, but there's a lot of variation in both groups. I had a friend, now deceased, who was a very knowledgeable programmer from the 1960s through the 1980s, mostly on DEC computers. He needed my help with some basic tasks on modern systems because he retired in the mid 1980s and, presumably, never bothered to deal with computers after that. He would not be right at home in front of desktop Linux any more than he was in front of his Mac. Someone born in the same year who didn't work on any computers would probably have an even harder time, assuming that they didn't voluntarily learn about them later on. Does that mean that all the old computer people don't understand how to convert and archive files? I don't think so. It means that trying to simplify anything into "old people do X, young people do Y" is doomed to failure. But if you want to try anyway, the least you could do is to not make the really obvious broken comparisons.

Re: TODO

You are correct, but this is still important because, if the person sees suggestions to use the CLI when they want to do something, it may scare them off. That's why a user-friendly introduction is needed if you want them to use the thing. For example, something that's come up several times in these comments is the following exchange:

Person 1: What if there's some piece of Windows software the user wants?

Person 2: There's probably some Linux equivalent.

Person 1: There's not. / What if there isn't?

Person 2: Use Wine.

Great, the user has their plan. They'll use Wine. So they go to Google and search for Wine on Linux and, after clicking around for a while, they find the most official-looking guide to installing Wine which is entirely CLI-focused, which is fine, because it's aimed at people like us who are fine with and may actively prefer to use the CLI. If you want to convince someone who is less comfortable with that to use Linux and you intend Wine to be an option, then you should suggest something else and provide instructions. The alternatives are to avoid using Wine and have some other answer for the Windows-specific thing or to not bother trying to convince them at all.

Re: I am incredibly dissapointed

Yes, you should. You shouldn't expect it to work, because you're great at preaching to the choir (they've probably gotten bored) and this is about finding and convincing people who have never cared about modifying the source and don't care now. However, maybe your experiences trying it will help someone. It would be better than continuing to complain to us who already understand these differences and have decided how much we care about each part of it.

Re: Decision tree?

The previous question would not help you. You'd already have figured out by that stage that they do use a spreadsheet, and specifically Excel, and you're now trying to figure out what kind of Excel user they are. Are they the person who uses it to do some accounts with simple formulas which anything can handle, or are they the kind of person who really should have been taught to program properly in 2000, but since they weren't, they have self-taught many of the concepts in a massive spreadsheet of macros, formulas, little weird Excel things that take the place of necessary functions, and through a tangled bunch of spaghetti formulas run a complex process? Those people will get annoyed when trying to make sure all of that works in LibreOffice, meaning you might want to guide them to something else, like running Excel in some way rather than trusting LibreOffice to do it. The problem is that it's hard to find a way to express that without confusion.

That's just one example. Very few of the questions you'd want to ask in that tree are going to be easily answered by everyone who visits. If it doesn't work unless someone can answer all of them, then it will get a lot of people leaving early. For example, one of the things I'd ask when trying to shift someone to Linux is how much free disk space they have for a dual boot, assuming they're not willing to lose Windows if it doesn't work out, which few would be and many of those who say they are will be unhappy if you actually do it. If the answer is that they have a small disk or one with limited free space, I would ask how easy it is to install another disk. That's not something the average user knows. I have to look inside to see if there's another bay in their laptop, which there usually isn't, and whether I can replace the disk at all, which I usually can but no guarantees. But if you don't ask it, then they get to the part of the installation where you're hoping to talk them through setting up that dual boot and they have just enough free space to get the distro on, but not enough for applications or data.

Re: Skill issue

And, from your perspective where the most important factor is how free you are to modify and distribute the source to a program, you are right. From the perspective of the people targeted by the website mentioned in the article, it's not important at all. A lot of people do not intend to modify the source of anything. They often don't understand why we care about being allowed to, and when I have, at long last, explained it to them, they understand why I care but not why they would want to.

Someone in that position will not care about freedom to modify the source. They will be asking questions like "is there anything I can do with Photoshop which I can't do with GIMP". I don't know the answer; I use neither of those. If the answer includes things they use or plan to, they're not going to accept source freedom as a reason not to care. If the answer is that they have the same features but the GIMP interface is worse, they're not going to accept source freedom as a reason not to care. I don't know what the difference actually is, but I see a lot of people saying there is a massive difference in feature set and interface quality and spirited defenses. Some of those defenses are obviously motivated by support for the source license, which makes their opinions on the feature set less convincing, but not all of them seem as motivated by ideology. You are willing to accept source freedom as a reason not to care about almost everything, but that's not going to convince the people this site exists to convince.

Re: TODO

You write software and are an electronics engineer. How many people your age globally are electronics engineers? Consider people who might have done different things with the first decades of their lives and consider how likely they are to have used as much Linux, or any at all, as you have. Similarly, there are some very young people who know everything about Linux because they had a Raspberry Pi as their first and only computer and they can tell you all sorts of little details about kernel structures and how they interact with hardware and how software interacts with them. Do you want to take a person at random of that age and trust them to do kernel dev right?

Re: TODO

"People of my generation are likely to have run both Windows and Linux, certainly I have since version 0.1 or thereabouts."

Please don't ascribe your experiences to everyone in your generation. When you were testing version 0.1 of Linux, a lot of people of all generations didn't have a home computer yet. No, most of your generation have not run Linux before, just as a lot of people of all generations have never run it on a desktop, have never run servers, and don't interact with the Linux part on any of the embedded devices that have it. When we're on this site, we can assume that most people who were young in the 1980s programmed a ZX81, that people who were students or young adults in the 1960s or 1970s punched cards, that people alive in the 1990s toyed with Linux, BSD, RISC OS, or BeOS. Oh, and often when we assume this of everyone, we're still wrong. But if you make similar assumptions about the world in general, those assumptions are so often wrong that it rounds to always wrong. Most people did not program anything, ever, and they don't want to do it now, and one of the challenges of getting them to run Linux is convincing them that they don't have to program the computer to make it work (they will count CLI commands as programming). These are not people who have already given it a spin.

Re: Decision tree?

There's logic in it, but it's a lot more work and there are potential problems, for example:

"Do you use rare functions of Microsoft Excel which won't work well in LibreOffice Calc?"

Some users select yes because they use really complicated features of Excel like the formulas and that box that makes graphs.

Some users select no because LibreOffice looks like it's got a lot of features, so it will probably work.

Most users select "I have no idea" because they've never tested anything in LibreOffice.

It's hard to ask sufficient questions to give a good answer without running the risk of confusing users, possibly causing them to just give up. In some respects, it might be more reliable to simply give them some version that's likely to work and letting them see if it can do what they want, and if I had to recommend something for people who have no support and insufficient knowledge to figure it out for themselves, I might propose that method*. The problem is that it's not easy to coach them through installing it in such a way that, if it doesn't do what they want, they can go back to Windows safely, because it's very easy to break dual booting if you don't know what you're doing during installation and that will just mean a lot of people angry about how Linux broke their computer.

* I have helped people by giving them equipment which I created out of discarded computers and parts. I didn't have Windows license keys just sitting around, so they came with Linux, usually just the latest Ubuntu LTS, and I told users I'd help install Windows if they bought or already had a license and asked for that**. Quite a few of them were able to find the applications they needed. I'm not sure I ever found someone who didn't react with some uncertainty when faced with it, but since many of them just needed a web browser and word processor, Ubuntu was often fine. At least, if it wasn't fine, they would have a specific thing they had tried and didn't work which I could review which was more reliable than asking them beforehand and hoping they gave enough information to predict it.

** I wasn't trying to promote Linux there. Whatever the user wanted and I could provide, I did. I did try to avoid spending a lot of time personally teaching them how to use the machine, though, because that is a nearly infinite time sink. I really respect those who have the time to do that more consistently.

Apple to add fresh accessibility features for 2025

Re: "Lie-nucks"

"So I don’t quite share the author’s sunny appraisal of Apple’s assistive features, no."

I think you're thinking about the wrong thing. Speech recognition has lots of flaws, and likely always will although I've seen significant improvements. But that's just one tool and assistive use is just one of the things it's for. People without disabilities use that feature all the time. There are a lot of features which are specifically for use by people with disabilities in most operating systems, and, for example, only one of them in IOS is related to speech recognition, namely the feature used to control the device by voice for those who cannot or find it difficult to use the touchscreen to do it, and those users also have switch control as an option, where they use a separate controller which they can more easily interact with to control the device. Features meant for people who have trouble seeing, hearing, or speaking are completely unrelated to this, so the failure of a speech recognition model wouldn't be useful in judging the quality of any of those.

Meanwhile, I'm not sure where they get those models, but it's likely that they, like a lot of operating systems, license it from a company that specializes in making them*, so they can't actually improve that part easily. Who knows, as Apple is big enough that it might have brought that in. Microsoft bought Nuance, one of those specialized companies, so they're presumably using that, so Apple could have done a similar thing.

* It's a little easier now with neural networks and cheap computer power to keep throwing at them, but for a while, they employed a lot of people who were really focused on linguistics and acoustics. Building a model for one language was hard. Doing it for the thirty languages most operating systems want to provide was extremely hard. Specializing was the most logical way to make it feasible, since the work could be spread out among lots of users.

Whodunit? 'Unauthorized' change to Grok made it blather on about 'White genocide'

I'd consider the singer and what their intent, ideally their explicitly described one though they might hide it, was. I would not ascribe that attitude, no matter how much I might detest it, to everyone else in society who was not singing it. Of course, when there are allegations that white South Africans are subject to "genocide", I need more than a lyric to demonstrate that anything of the sort is happening or planned, even if that lyric were unquestionably calling for it. I have, unfortunately, heard calls to mass murder from fellow residents of my country, but thus far, their suggestions were rejected by the rest of us and their statements don't represent what has happened.

Song lyrics are an accurate representation of everyone's views, I'm sure. Shall we look for lyrics advocating other positions and ascribe them to you, without any evidence that you've ever sung them or knew someone who has?

Apple slams door on Fortnite's stateside iOS comeback

And you have taken the completely opposite tack. They don't like Apple and suggested weird, ineffective, and possibly unethical approaches to try to harm it. You have instead decided to completely swap the positions and ignore the fact that Apple has violated laws in a lot of countries, and that is a bit of a problem for which they do need to face consequences. Not individual employees, although the one recently cited for lying to a court might be an exception, but the company must not be allowed to continue to ignore courts. Anticompetition legislation exists for a reason. Apple is covered under it based on its massive market share and single realistic competitor. If you don't like it, campaign for the law to be changed. Otherwise, it can and should be enforced.

Uncle Sam claims H-1B fraud crackdown is working as registrations drop 25%

Re: Green card – which is itself a pathway to citizenship and freedom

In most countries, in order to naturalize, you have to have been living, legally, in the country for a certain number of years, not counting certain types of residence that don't qualify you like student visas. If you don't have one mechanism for doing that, then you need another one. There are often not too many choices, and it's not like you get a long menu because you might not qualify for all of the ones that do exist.

Re: Scam

It depends how you offshore. The problem you mention happens if you subcontract it to a local company in a different country which is contracted to provide a service, but if you set up your own local subsidiary and employ people as employees, then you do know exactly who is doing the work because you selected and approved them manually. The two are very different, and comparing them as if they all work the same will lead to incorrect conclusions.

VPN Secure parent company CEO explains why he had to axe thousands of 'lifetime' deals

Re: How is this legal?

No, that's not how anything works. If I provide you something that I have no obligation to provide you, I don't automatically adopt an obligation to continue providing it. There's no automatic assignment of liabilities. It's all very explicit in the transfer documents of exactly what they're taking and what they're not, and their later actions don't retroactively change it.

Because, sometimes, people lie. Evidently, some lifetime VPN systems actually work. I would not have predicted that. However, most of them could not afford anything close to that usage, and I'm guessing your provider no longer offers that option and is paying for your usage from subscriptions of others.

For the same reason, when I was recently advertised an Android phone from Aliexpress with 16 GB of RAM, a terabyte of internal storage, running Android 15 on a high-end processor for about £60, I knew that this was not what I'd receive if I chose to buy it. I would have liked that to be true, and I was tempted to order it if only to see what they actually sent as I could always deny payment for the fraud. It's still illegal for them to do that, but it can pay to opt out of a crime at the start rather than hoping that you can recover your investment after it's completed.

Re: How is this legal?

If they are, in fact, two independent companies, then there is no case for fraud against the new ones, but depending what happened to the old ones, there might be. If the old ones sold off the VPN business and took the money, that's fraudulent conveyance, and customers and creditors can sue them. If they sold off the VPN business and are still operating, then customers can sue the old ones for breech of contract, depending on what the contract said which might go either way. If the old ones became insolvent, there's no fraud and no case. If the new ones and the old ones are the same people, that's more obvious fraud.

Re: How is this legal?

That's what happens when companies merge everywhere, but they are not claiming to have merged. They are claiming to have acquired the assets, not the company, and possibly not all the assets of the old company but just those for the VPN service. If, for example, the old company was going through bankruptcy, it would be normal to sell off the assets to make money to pay for the liabilities if you couldn't find someone willing to take them both. Of course, if a company tried to sell off their assets and not pay their liabilities outside of bankruptcy, that's illegal. Canada isn't special in this case; the same things are legal and illegal there.

Re: How is this legal?

From the sound of it, the previous owners didn't sell on the liabilities, so either the previous owners still have those contracts and could be pursued for the service, or more likely the previous owners went bankrupt in which case those contracts are probably void*. The new owners would own the domain name, the VPN software, and a list of former customers who can become current customers. Of course, if the rumor that these are the same people is true, that would be fraud.

* Depending on the text of the contract, the insolvency of that original company could probably be argued means the lifetime of the company has ended and thus the contracts expired. If they didn't say that, then the company would simply be unable to meet their obligations and the contract holders would be considered creditors, but given the prices of the contracts, creditors unlikely to ever see any of the money that liquidation generates. Either way, it's not coming back and there's almost nothing you can do about it.

Ransomware scum have put a target on the no man's land between IT and operations

Re: Back to basics

It will probably take some massive disaster caused by an attack. So far, most criminals don't want to destroy this kind of stuff. They want money, and they won't get it if things are blowing up or bodies are dropping. They get it if all the business computers don't work, so they still mostly aim at those. Most companies have never had an incident like this, so they're going on reports of what happened to someone else, which are much easier to ignore than things that happened to you. Meanwhile, security people are busy running around attaching locks and tape to these systems in the hopes of preventing what they can, and those methods actually work a lot of the time*, so people who look at the effectiveness of current methods don't think they have to do anything more.

* Because, although we're nearly always telling users and employers otherwise, a lot of attackers are opportunistic, and if you have better security mechanisms than someone else, they'll go hit the someone else rather than work at breaking through yours. We're not lying about needing to have good absolute security, not just in comparison, because those attackers will try to hit as many places as they can, but a medium level will often knock out a lot of the attempts. That doesn't work when dealing with someone who actually wants to cause destruction rather than earning the most money from the least effort. Those people can and will break through much more complex things to do it.

Re: It's not rocket surgery...

If you decide you want to run your own cabling, it would seem more logical to run it outside but alongside the pipes rather than in the pipes. That would fix most of your concerns. Why you need that and whether there might be a less complicated way to obtain your goals are still legitimate concerns.

Meta's still violating GDPR rules with latest plan to train AI on EU user data, says noyb

Re: Consider...

Most of your "what would have happened" examples have no plausible link to having privacy regulations at the start of the internet. I'll agree with your first one: if you want the AI to be a little better, training on reliable data that's relevant would help. I don't particularly care whether it gets better, but I do care about having my personal data considered free for the taking, so I don't mind that tradeoff. Also, I don't think most of my personal data would be helpful training data. Also, even if you did want it, nothing says that the AI companies should be able to get whatever part is useful for free just because. But apart from those three problems, the general theory is right.

That's more than I can say for the rest of them, such as:

"no comment forums like this one" Why? This forum is funded by the paper it's connected to, which is funded by advertising. Advertising would still be permitted. It would be less targeted to your browsing history, and instead it would be targeted based on what's on the site, which worked pretty well in the past, and might actually be better because a lot of targeted advertising isn't targeted well. Why would they be unable or unwilling to run a forum?

"you would be charged per e-mail sent": Why? Because all the email providers that don't use the mail for tracking charge per month or per gigabyte, not per message. Some of them are also free, using the untracked mail as a loss leader to attract people to other businesses. What would change?

"governments would licence and whitelist all websites": Why? Just why? How on earth did you get from privacy regulation to that?

You made up all these things, including the many I didn't bother to list. Almost none of that would have happened.

Linus Torvalds goes back to a mechanical keyboard after making too many typos

I do wonder about the many people who post here who have trouble finding a keyboard, because I've got a keyboard much like yours. It's from some no-name manufacturer and I might not be able to find that specific model again, but it's basically what you asked for:

Very standard layout, has a numpad which I could have done without, brown switches, backlight is configurable to all white (I turned it off). I wanted a mechanical keyboard, but I didn't see any significant advantages from the expensive manufacturers, so I went for the cheapest normal-shape keyboard with the kind of switches I like since I don't want my keys making any more noise than they're already going to make with the extra travel. That keyboard has been working fine for the last six years and I expect it will continue to work fine for many more. It cost me about £40, so I'm wondering whether there's some other feature that people are looking for but they consider so obvious that they aren't mentioning.

After more than half a century, the voyage of Kosmos 482 is over

Re: no mention of radiation

Other Venera probes were powered by more conventional chemical batteries because Venusian surface conditions weren't great for the same kind of RTG power sources that we use on Mars. That doesn't prove that this one didn't try it, but it seems likely they'd do the same thing they did with previous probes on which the batteries were sufficient for their experiments.

Amazon tested warehouse robots and found they're not ready to replace humans

Re: Capitalism is dying

Nobody said that it was always a utopia, just that mechanization did not eliminate jobs leaving people with no option. The industrial revolution was horrific in many ways, but before we put on our rose-colored glasses, the time before it wasn't very nice either. You've focused on people doing the same work. Things got worse for them, with the guild weaver earning more money and living a better life than the factory weaver. The point of the guild, of course, was to make sure that there weren't many guild weavers, and those who could have woven but weren't allowed to had to do other things, things that were often paid worse than and with similar danger levels to the mill workers they would be decades later. They just did them with less commentary because that's how people had lived for a long time, so lots of injuries and deaths was not seen as in any way surprising.

That doesn't tell us a lot about today, but one thing it might is that, when people predicted massive losses of jobs from a technological advancement, they were often wrong. We're not just talking about 1800 with its scattershot record-keeping either. The same predictions were made throughout the 1900s with advancements like more efficient assembly line manufacturing, less labor-intensive materials, and of course, computers which were going to do everyone's job any year now. The first is now seen as heralding one of the most laborer-friendly periods, but that was not obvious at the time and many fought against it on the assumption that lots of workers would lose their jobs, nor was it inherent in the technology, since it's a combination of societal and technological factors that determines how a new situation will be felt. Therefore, maybe we should wait for evidence of massive losses of jobs without replacement before we say it's different this time.

Nextcloud cries foul over Google Play Store app rejection

You're right that it's a global requirement, and wrong about everything else. It's a global requirement so it depowers Android applications, making them store as much data as possible in the /data/data folder that users can't access, and user-installed apps can't access, but you know what can access it? That's right, Google sync software and Google backup software, installed and running as root. Now, why do you think they put in that requirement. It's not enough to just tell us that they did. If there's no actual security benefit, then that answer is not good enough. They aren't going after Nextcloud here. They're going after a bunch of different apps, of which Nextcloud is only one. They've taken similar steps to reduce access to users' data through any mechanism other than OEM-installed system apps, for example deprecating ADB backup without replacement.

But maybe this is Nextcloud's problem because they should be using some more secure method. Let's see what other applications use. There are convenient privacy reports that show the permissions apps use. Dropbox has READ_EXTERNAL_STORAGE and WRITE_EXTERNAL_STORAGE, the ones that implement access to all files. OneDrive, Box.com, Amazon Drive (write only for some reason), all have these permissions. But maybe these are just all lazy companies. What does Google use? Google Drive uses exactly the same permissions to do it.

I don't think Google is targeting Nextcloud here. I think Google is trying to target everybody, but if they did the same thing to the big players like Microsoft or Dropbox that they're doing to the small ones, they might get some real legal attention. But if you think this is totally normal, you can explain why Microsoft, who has thousands of programmers at their disposal and updated their OneDrive app lots of times since Android 11 was released, gets to use these.

US, China agree to roll back tariffs – but only for 90 days

The reasons differ depending on whose reason you're thinking about. Trump's reason is because he is into mercantilism. That says that exporting is good because you get money from other people and thus have more money, and importing is bad because you give money to other people and thus have less money. It doesn't work because it fails to recognize that there's a lot of stuff with value that isn't money.

His supporters probably have a bunch of different reasons, and some of them like the mercantilism idea too. The other important one, though, is many people who have, either in reality or in their perceptions, lost a quality of life they once had or expected. People who expected that there would be easily-obtained and well-paid employment and found that these were not as simple as they expected are unhappy with their current situation. In some cases, that decline is real, whereas in others, they're actually able to obtain work that compares well in compensation and difficulty to their previous options, but they would prefer it to be easier and are under the misconception that it was better at some previous point. Various other factors, such as the increased prices of expensive things like houses and education have exacerbated this problem. Politicians, searching for a simpler explanation for this situation than the many realities of economic, technological, demographic, and geographic change, have decided that manufacturing in China is a convenient thing to blame it on, because it was at least part of the answer, so it can be stated as the whole answer. Therefore, anything you do that hurts China would have to be pushing the US back to the ideal times of the past where their lives would be better.

US Copyright Office found AI companies sometimes breach copyright. Next day its boss was fired

Tell you what. Try that argument in a court. When you are convicted, that will prove how much your opinion matters when there is a law that says these things.

There is code I want to use. I have access to it; somebody posted it. It is copyrighted to a company who has not released it as open source. Guess what will happen if I try to use it for commercial purposes? Maybe they won't notice, in which case I'm fine. Maybe they will, in which case I can be sued. The fact that it's online, whether they intended it or someone posted it illegally (it's unclear) makes no difference at all. no matter how much I might want that to make a difference, it doesn't. Your opinion doesn't matter because this is about facts.

No, and if the license sneakily says you're not allowed to read it after you've read it, that's a reasonable excuse. But this is not that. This is being allowed to copy and use it as you like, which is when you have to look for the license. If you didn't know that, this is your official notice. Not everything you can read is free for unlimited and unrestricted use.