Posts by doublelayer

The graying open source community needs fresh blood

Your stereotype might not be helping. I'm a relatively young (adult, working in industry for a while) programmer, and yes, I too was taught C and assembly and systems programming in general. I don't use the assembler now because my code runs on servers, and if it switches from X64 to ARM servers we don't want to have to compile it again, but I can, have, do, and will write C programs whenever it is useful. If I switch to somewhere where writing in assembler has a point, I know the concepts involved, and while I won't start as quickly as those who already do it, I can do that as well.

I also know enough not to assume that C is always the right hammer for any nail that comes along. Obviously for things like the Linux kernel, you will need a language that compiles to efficient machine code, and there are other areas where that is of paramount importance. Even there, C is not the only such language. However, a lot of software, if written in C, will run in less CPU time and take a lot more development time, and in many cases, CPU time is so cheap that the improvement is irrelevant. People who make assumptions that those who choose to use something other than C do so because they are not smart or knowledgeable enough to use that tool can often be wrong.

"Digital native" never meant "knowledgeable assembly language programmer". It only meant "can probably use a computer successfully without needing support". Don't assume that the people saying it are attributing skills they probably don't understand themselves.

PowerToys bring fun tweaks to Windows 10 and 11

I think the "preparing" message comes when the process is still recursively listing all the files. It should and probably does copy the first few while searching, but it doesn't even know how many files there are yet. Since the next version will count the number and size of files remaining to copy since the time estimate is so frequently wrong, it would be annoying if the files left box kept going up.

Stop installing that software – you may have just died

Re: How did you get into this room?

Probably some education consisted of the line "Now that you can plug in inside this area, we want you to always plug in here and have that door closed while cleaning this area. Yes, we know that wasn't an option before, but now it's important."

Singapore's banks to ditch texted one-time passwords

Most of the time, they're using one of a few standard and open protocols, so they can run on desktops. All you have to do is identify the protocol in use and get one of the many open source implementations that has a desktop client. There are a few reasons why people generally don't, and that also explains why the most used authenticator apps from large tech companies don't have desktop versions, but you have about ten options with a quick search of GitHub. Finding one you trust can be trickier, but if you want a desktop version, you can have it.

There is a chance that some business will build their own authentication protocol which is not a standard, and if they do that and force its use, then you will be limited to whatever they allow. However, they often outsource this to something which supports those open protocols and you can use them even if they have a proprietary version.

Smartphone is already many folks' only computer – say hi to optional desktop mode in Android 15 beta

There are a few legitimate reasons:

1. The USB port requires support for displays and some don't have them.

2. The SoC may have limitations on screen size and refuse to support something larger.

And the ones that will actually matter a lot of the time, but not so legitimate:

3. If you don't let one phone do it, then maybe people will choose to buy the more expensive one that can.

4. The hardware can, the software can, but you need something like a kernel update and nobody is going to rebuild a newer kernel for this phone. Why not, they don't take too long to build, and there are phones with the same chip that already run the latest kernel so you don't need to wait for the SoC manufacturer to do something? If this operating system had been built competently, this process could almost be automatic. Nobody knows, but the manufacturers just won't do it. You can sit there explaining all the reasons why this should be no problem, but no matter how much you do so, the new component will never arrive.

With users mostly happy to keep older kit, Macs just ain't selling like they used to

Re: The no upgrade gamble

As it happens, I do. I can upgrade the RAM and storage as well, just as on many other laptops.

The throttling thing is for iPhones, not, as far as I know, Macs, and even on iPhones it tends not to be a significant decrease in CPU speed. They generally get about seven years of software support, which while worse than Windows or Linux, is significantly higher than your number. For now, OpenCore Patcher does a pretty good job of letting users ignore the cut in support, but that may not work as well for ARM devices and almost certainly will not be able to do a thing when the latest Intel models run out of OS support. I have a Mac that Apple decided to cut off at Mac OS 11. It's running Mac OS 14 fine, though the battery has seen better days.

It is fair to say, however, that people who don't use OCLP to do what I've done see a significant degradation when their OS support lapses. It's not just not getting feature updates. It's not even the worsening record of security updates quantity or speed. Many applications drop support for Mac OS releases quickly, especially including anything that Apple wrote.

Babel fish? We're getting there. Reg reviews the Timekettle X1 AI Interpreter Hub

Re: What a useless product

My preference would be to have an app if my phone is capable, thus avoiding the hardware, and to have a single purchase price, thus avoiding the subscription. I just pointed out that there are many markets where my preferred model basically doesn't work because people are unwilling to buy software but accept paying for hardware and subscriptions. That doesn't mean this hardware device will be successful. I think a lot of people will continue to use Google Translate and ignore the existence of this thing, and for all I know, their system might not have much improvement over that anyway.

Also, I think you might have one thing wrong:

"just look at Rabbit's owners going into desperation mode selling them for $199!"

$199 was the original release price. It's the same cost that it ever was. I didn't want one then, I still don't want one, and I'm not sure the company will continue to exist for long, but your sentence can be correctly paraphrased as "just look at Rabbit's owners going into desperation mode still selling them for the original price!"

Re: What a useless product

The only way it could work, in my opinion, is if the hardware is designed to fix problems that existing translation apps can't. For instance, if they're building this because it lets them include more and better microphones, thus decreasing background noise and improving the speech recognition, then they have a case. If they've got new local models, but those models don't run fast enough without a good NPU and they've built their box to have one, then they have a case. If instead they built a device so they could sell it as hardware instead of software, then I get their intent*, but it's as disappointing as you say.

* I think there are some markets which will buy hardware but very little or nothing for software. A lot of consumers have gotten used to a world where software is free, which makes it hard to sell software without going to a subscription model and annoying everybody.

Privacy expert put away for 9 years after 'grotesque' cyberstalking campaign

Re: A Simple Matter

I don't think the logic that committing a crime when there's the remotest chance of punishment is indicative of mental illness. There are two options that are both more likely in my opinion: pragmatism and stupidity. There are some criminals that actually analyze the risk of getting caught and decide that the reward of the crime is greater than the risk. Sometimes, their analysis is correct from their perspective. We also have another group that thinks this is the case and are wrong because they didn't think it through. I don't think there's a mental illness that clearly applies to either group. Both are immoral, and the second are stupid, but immorality and stupidity are not generally classed as psychiatric disorders.

Re: Privacy Expert?

I have to agree with you. At first, I was expecting the story to involve someone who had studied complex surveillance tools and consulted on how to get rid of them, which would also provide an unhinged person with plenty of experience on what to use to commit crimes. That doesn't seem to have been what he did. Using email addresses with his name, his personal phone, and various other obvious traces suggests that, in addition to being criminal and possibly mentally unstable, he wasn't an expert on anything privacy related or he could have made his crimes harder to pin on him.

Twitter grew an incredible '1.6%' since Musk's $44B takeover. Amazing. Wow

Re: Amazing. Wow.

You need to learn a few things, such as what coding workers do and what non-coding workers do. If I leave my employer and my code falls over when I'm not there to babysit it, my code was bad. If my code was even mediocre, it should keep working the same way it did before, meaning that if input doesn't go outside specified bounds, it should keep producing valid output. Code continuing to run after employees leave is no proof that the programmers were not needed, because what does suffer when you fire them is the ability to quickly make changes to that code. If it had fallen over, it would have indicated one of two things: the programmers were very bad or the system managing them and their output was very bad to the extent that they couldn't produce good code under it.

You should also learn that companies whose main product isn't tech don't just have programmers. You could fire lots of people and still have the programming teams intact. You'd have lost the people who gain advertisers' money, the people who prevent the kind of stuff appearing which drives off the advertisers, the people who make sure that it is properly working in different countries with different laws and languages, the people who make sure that bills are being paid (both your bills to others and others bills to you), but your programmers would still be there and your platform shouldn't fall over. That is not the recipe for a functioning business that gets all its money from advertising.

If you think your system would fall over if you fired more people, so therefore you have the right number of people, you're doing at least two things wrong.

I'm not sure there's an objective way to answer that question. I don't have an account either, so all I have is subjective reports from some people. Those reports universally say that there was a large increase in spam and bot activity, in some cases causing the users to leave. However, we must assume that any of my friends who had Twitter accounts, did not see a spam flood, and are thus just fine with their accounts probably didn't feel the need to tell me this, so that would bias the results in favor of more spam. A reporter could collect a lot more anecdotal reports, but might not be able to give an answer that works for everyone. What we can say from the data we have so far is that some subset of people have seen a worsening of spam, so there has not been universal improvement and may have been no improvement at all.

Re: Bots

He complained about bots before he bought the thing. Whether those bots actually existed or were his attempt to avoid buying something that wasn't a good idea to buy and he knew it by that point is harder to prove. It's also not clear whether changes he made had any effect on the number of bot accounts. In general, if they had as good a defense as you're suggesting they do, they would have included that defense and some attempt to back it up.

Coders' Copilot code-copying copyright claims crumble against GitHub, Microsoft

Re: Slackers

Possibly GitHub used Slack before Microsoft bought it and interactions between the two may still use Slack, or did at some point the lawsuit contends is important. I'm fairly sure that most chat inside Microsoft would use Teams, but they have a lot of subsidiaries and they may have invested in a Slack-based workflow.

Windows Notepad gets spell check. Only took 41 years

On my machines, no, I don't, or at least very rarely. On others' machines, yes, absolutely. When I'm helping someone fix something and I need a text editor, whether that's to edit a configuration file or just to store some notes about what I'm doing, they are unlikely to have Notepad++ installed in the first place and I'm not going to install it while I'm there. Even if they do have it, I'm not going to mess with their setup. So in those circumstances, Notepad gets used.

Epic accuses Apple of foul play over iOS access, wants EU to show DMA red card

I have spent the money on an iPhone. It's mine. I own it. I decide what runs on it, without paying Apple for the privilege. If they try to deny me that ability, I try to make it a requirement. The same reason the person who built your house cannot decide what you do in it. Even if they still own it and you're renting, they still don't get a lot of choices about what you do in it, though they have more control. But I did not rent my Apple devices from Apple. I bought them.

Re: FFS

That doesn't mean that interfaces should be made different in every way. As long as they don't deliberately say that it's the Apple one, then you can figure out which one you're in by looking for the logo, name, or information. Or you could go to the home screen, pick the one you want to be in, and be sure you're in there. This is always assuming that Epic's interface will look exactly like Apple's even though it's unlikely to do so because some of the things on Apple's store interface don't make sense in an Epic one. In none of these cases would design of individual UI elements be relevant.

Re: FFS

"I think in this case with Epic Apple wants to insure Epic's app store looks different enough that there is no possible confusion from customers."

I think it's more likely that Apple wants to throw as many blocks in Epic's way because they don't like them or anyone else who wants a separate store. They have done exactly what you don't like the EU doing: making vague rules for how you could maybe do the things the DMA says you're allowed to do, rules which already violate the spirit and possibly the letter of the DMA, then say that everyone who has tried is in violation of those vague rules to delay them getting anything done. The various arguments about how you can tell someone that your website has cheaper prices because it avoids the tax, whether you can only do that in one place, whether you have to specifically ask for Apple's permission for the layout of that screen when you tell them, whether you have to pay Apple and how much, all of these things are ridiculous actions taken in, at best, malicious compliance. The further investigations suggest that it wasn't compliance at all.

Breaking the rules is in Big Tech's blood – now it's time to break the habit

Licensing is the wrong approach

Licensing for open source software is both necessary and prudent. Licensing for every file that gets posted to the internet is not the right approach here. This is true for a few reasons. First, AI companies have already tried to ignore that and are fighting to do so in court. Microsoft and its AI providers are major culprits with their Github Copilot product, but that is not the only bot to have scanned lots of licensed open source code and then ignored the licenses. They have already acted as if licenses don't count and they're trying to somehow make that legal. I don't think it will work.

The more important reason is that licenses should not be necessary to deny permission to scrape. If Microsoft forgets to apply their boilerplate license notice to some code and I get a copy, I do not have permission to do whatever I want with it. The same thing should apply to other types of work, and surprisingly enough, national copyright legislation and international copyright treaties confirm that it does. If we establish a precedent that, if you didn't stick a license on your content, any treatment of it is fine, we're making things more annoying for no good reason, and if we're assuming that copyright law will still apply where explicit license terms haven't been applied, then there's basically no benefit to doing it in the first place. A license that requires them to do something they don't want to do should be no more powerful than a license that denies them permission unless they negotiate with you, and that's what we should have by default with normal copyright, no explicit licenses required.

Cancer patient forced to make terrible decision after Qilin attack on London hospitals

Re: I often wonder....

You're right, we should dump any operating system that Qilin's ransomware can run on. So that's Linux out then. There's a Windows version too. Macs for everybody?

In this incident, you are correct, but not everything works the same way as writer's block. There are indeed situations where the last negative event is the cause of the current situation and preventing it before that event occurred was either impossible or impractical. To blame the damage caused by the malware on striking employees is flawed logic, but to claim that a strike won't have negative effects of a different sort is also flawed. You may also want to consider that the statement where the action was mentioned wasn't trying to blame it for the specific incident covered in the article or for the damage caused by ransomware in general. At most, the strike was being blamed for a more difficult recovery from those independent events. You can easily disagree with that allegation, but not if you interpret the statement as trying to blame striking workers for something the ransomware broke, because then you're fighting against something with no defenders.

How tech went from free love to pay-per-day

Re: Speaking the copium

You are continually failing to recognize the points that we have been making. The point is not about whether Linux as a system got worse or was always bad. I agree with you, from the article you linked, it used to be more capable and now you have to do the annoying steps listed to make it capable again. But great, both before and now, you can do some types of things with it. But that's only what you have to do for the software to run at all, and there are times where that software is not available.

I don't work with graphics, so let's consider something I have a little more experience with: audio production. If you're a professional audio producer, you probably have a lot of tools you can choose between. Some of the more common ones include software like Avid Pro Tools at the center, but you have lots of alternatives. Does ProTools run on Linux? No, and if you're trying to convert a business that has already sunk a lot into Pro Tools, that can be a problem, but as an individual, you don't need it to. If you're looking for software with similar functionality, you have some choices like the open source Ardour or the not open source but it runs on Linux Reaper. People use those in professional settings, so if you just do a Google search and don't work in the area, you'll think that audio production is just fine under Linux, and it is if you're using the basic tools. Some people are. If you're making a podcast, either of those tools will be more than enough. A lot of people, however, are using plugins to these programs. They're relatively standard. You can buy one (nearly all the most popular ones are commercial and proprietary) and attach it to Pro Tools or Reaper or probably Ardour (not as many supported, but it works with most of them). Problem solved? No, because those plugins fall into four categories:

90%: Windows and Mac OS versions

4%: Mac OS version only

3%: Mac OS version only, and it hasn't been updated with an ARM build yet

3%: Windows version only

So when someone says that they want you to do something which isn't in the default set of features, but there are ten commercial plugins that could do the job, what will your excuse be? There are people, and unfortunately you seem to be one of them, who are motivated to claim that open source tools can definitely do that. The real answer is that it often can't because there are only so many people who want it and many of them don't care that the software that does it costs money and doesn't offer them the source because they can't write code anyway, so they don't need the source. I don't blame people for not knowing this, because they don't work in the area. However, when I don't work in an area, I try to avoid being very confident about what exists and what doesn't because, when I'm wrong and I probably will be, they will not trust me because I'm putting my preference for an operating system over being honest about what they can do.

Re: Speaking the copium

I have now read that page. Did you read the page? It does not help make your "just as useful, just as powerful" point. For example, try to make those work with any of these statements:

In all these years, I have never seen the GNU/Linux distribution landscape regress so far away from our needs. It is almost impossible to find a distribution where you can professionally run and set up our most basic tools: Creative software, graphic pen tablet, color calibration. And I tested a wide range of GNU/Linux distributions to make this guide! The choice we have in 2024 is super limited.

A small group I was a part of saw this coming for years and we were vocal about it. All of these issues were discussed to avoid the situation I am describing today. But it wasn't enough. We were told that our use was niche, our needs were diminished. Only a small percentage of what was deemed "good enough" has been done and released as is. Presumably, the decision was made to get the advanced functionality down the road, incrementally.

Yes, this guide exists to help to get around a lot of this, but these quotes are only about system things. Anyone doing this kind of work who has not used Linux before has to trust that, after they work around all the problems described in the post, the software they install will actually do what they need it to. The post you linked describes exactly what worries us about overconfident assertions such as the ones you've made:

The risk: This ecosystem might repel professional CG artists who are just trying to install their first Linux distribution alone. After testing what is currently installed by default, they will see the GNU/Linux alternative as a joke.

It hurts, it's a bad result regarding everything I've advocated on this blog for over a decade. This situation is draining my energy...

Re: The ribbon

I've seen no evidence from your comment or anything else that the ribbon was an improvement rather than just a change. By now, I've become as used to it as I'm going to, although I mostly don't have to use software which has one. I imagine that switching back to menus would annoy a new group of users all over again, and on that level, I can agree with you that it's a complaint that's become stale. However, you will need to do more if you want to convince me that the ribbon did anything good. To me, it's basically just a menu, but less organized because it put too many different kinds of controls in it. Instead of a list of items, each of which could pull up more options, they tried to put all the most common controls at level one, meaning that the less common controls got shunted into an ethereal place where you have to hunt for them, and sometimes they move and something else slips there instead. That doesn't necessarily mean they are worse than a menu whose contents might also be confusing, although I still prefer those.

Re: Speaking the copium

FOSS is certainly not all about supplanting Windows or proprietary, and the people who actually make it know that. However, to say that assuming it is misguided may be going too far. There are a lot of supporters of FOSS who do speak as if there is a struggle against proprietary versions. Whenever someone announces that, for example, a big organization is going to switch their desktops from Windows to Linux, people respond as if it's a victorious battle against their foe and that anyone who didn't do the same was making an obviously stupid choice because everything would be better if they did. People who watch this would be hard pressed to make that align with your characterization of "uninterested".

We know that these are often separate groups. The people who make this stuff are often not the ones describing unrealistic evil plots by Microsoft (it's always Microsoft, why can't some other company be evil some time) to ensure that Linux dies next year. Those who are not as directly involved in multiple parts of this may not understand this. Someone nontechnical recently told me a joke which, unfortunately, I can't remember well enough to tell in a funny way, but the point was that Linux users were more annoying than all sorts of stereotypically annoying groups like telemarketers and Apple users. Those who don't know us but know a little probably have a similar thought in their mind, and it will take more than your statement to prove why this is not the case. Worse still, if they see that kind of argument and try using it, they may conclude that not only are there a lot of annoying evangelists, but those people are wrong because they said that an insufficient piece of FOSS software was as good as a proprietary one and it isn't.

Usually not, but they advertise a lot, probably because the people who sign up don't use their service very much. In addition, they can at times be edging closer to the line. They often talk a lot about privacy without necessarily providing much themselves; yes, your local network won't be able to intercept your data if you've used it properly and they can't be blamed for people who don't, but if they keep tons of logs on your usage, they're not private. They can at times use other methods, for instance a VPN service whose advertisements focused a lot on watching region-locked content, but all their endpoints were well-known and blocked by a lot of content providers. They're not automatically scams, but I wouldn't trust everything in those advertisements.

Switzerland to end 2024 with an analog FM broadcast-killing bang

Re: Digital AM - Digital Radio Mondiale (DRM) - can't censor radio like internet

They've had decades to try to make something that could receive it reliably and they failed. Any receiver you can buy likely hits at least two of the deathly triad of lasting an hour on battery, costing an order of magnitude more than something else, and not being able to pick up the signal anyway. In theory, it's a great idea, but there's only so long it can be the next new thing with one of the key parts of the system absent and no indication that it will appear any time soon. It has been used for some purposes, for example long-distance communication to a small number of more expensive receiving locations, and it can still be used that way. For the intended use, broadcasting to a large number of people, often people who cannot afford expensive receivers, it's not working. They've had years to turn this around and we're still pretty much in the same place we were in 2020 or 2015 or 2010.

Tech luminaries warn United Nations its Digital Compact risks doing more harm than good

Re: That horse has already bolted

As they pointed out in that explanation, they don't use it and want to continue that way. I think they've overstated how important it can be, but their reasoning is not flawed. I hope that it continues to be practical not to use it, but we should also consider that the effects of not having an account can change. We shouldn't assume that it will always be as easy to avoid as it currently is. If we assume that, we might not pay attention when we could stop something dangerous from getting started.

Nintendo sues alleged Switch pirate pair for serious coin

Re: On one hand...

"are you mistaking the rant against Nintendo for being anti-homebrew in *all* its products as some form of defense of these specific Switch "pirates"?"

I'm not sure, but I'm definitely doing one of the following two things:

1. mistaking the rant against Nintendo as some form of defense of these specific Switch pirates, or

2. correctly interpreting the rant against Nintendo as some form of defense of these specific Switch pirates.

If it's not trying to defend them, then I've got it wrong, but I submit that it could have been made clearer, perhaps by actually stating that because we were discussing these particular people. Then again, I'm trying to figure out what you meant to communicate by putting "pirates" in scare quotes when referring to the people the article was about, and I'm not sure I have correctly interpreted that either. To avoid your ire, I will not state my assumption and you can specify it if you feel like doing so.

Re: On one hand...

"You could write a hypothetic firmware patcher that could save a bricked device and they would still come after you."

And if you hypothetically did that, I would have a lot more sympathy and, if you were in the United States, you would be able to argue that you were operating under the exceptions to the DMCA and had the right to do that. They did not do that, and defending them with the vehemence that we would legitimate actions will only weaken our ability to make the case that there are legitimate purposes, strengthening Nintendo or anyone else if they try to remove the exception.

Re: On one hand...

That argument would work better if they had made a homebrew modification that made piracy possible, but that was an unintended side-effect and they did it for completely legitimate purposes. It would have worked better if, even though they intended to use it for piracy, they said it was an unintended side-effect, etc. They chose neither of those options, meaning that your defense of them is falling a lot more flat than you probably intended.

The rest of your defense isn't helping either. Your characterization of their antipiracy measures, which I wouldn't be happy with if I owned their hardware, as trying to "police coding and learning" when we both know that's not what they're doing, does not help your point. Nintendo is not locking down their hardware because they want me to be unable to learn how to code. They are doing it because they want to charge lots of money and make it difficult or impossible for people to play the games without paying them lots of money, and that's something they have the right to do. For those two reasons and many others, I do not have Nintendo hardware. Incidentally, their use of a pirated copy of an emulated game, while humorous, is not itself piracy because they own the copyright to the thing.

EFF wants FTC to treat lying chatbots as 'unfair and deceptive' in eyes of the law

I agree that no new law should be needed here. I think they're asking for one because, when you have human agents, there are some classes of mistakes which aren't automatically on the company. If a person on the phone misreads a number 1234 as just 234, whether that's to my benefit or theirs, that's usually something that just gets corrected, not resulting in tons of legal consequences. A law, if passed, should make it clear that there are no honest mistakes from an LLM, and any hallucinations are their problem. Otherwise, I expect that court cases will eventually end up building the same thing in case law, but we'll have to endure many companies trying the Air Canada defense: "the chatbot said it, so it's not on us".

FreeDOS and FreeBSD prove old code never dies, just gets nifty updates

Re: Why?

It could help if you didn't misconstrue what people say when you disagree with them.

"They only do it in the biggest, most populous country in the world -- also one where most people are on average quite poor -- so it doesn't count?"

Did I say that it doesn't count? I'm questioning its frequency. If they do it in only one market, that lowers its frequency, especially if they don't sell many of them there. Now, from your comment, I'm questioning something else you say. Did you see that I pointed out that the prices were bad? I can get new computers with equal or superior specifications in India for a lower price. So the average being quite poor definitely does not count if FreeDOS doesn't decrease the price at all. Similarly, if it's supposed to run a Linux version, why not install the Linux version? That's just as free and less likely to confuse people. If Microsoft's licenses mean they absolutely can't charge less than a Windows version, why would people buy these to install pirated Windows on when they could spend the same or less and get one with a licensed Windows on it? Your rejoinder brings up more questions than it answers.

Re: Why?

Whether or not that's the case for them, I have to wonder why they would sell a computer with FreeDOS on it. Maybe I am not thinking of a customer, but I imagine that people who want to and know how to use FreeDOS are probably able to install it themselves. If it's not cheaper, why not include software the customer is more likely to want? Windows is pretty familiar, most people could figure out what to do if it booted a Linux that started in a GUI, but FreeDOS would confuse most buyers I know.

Mostly, though, I was trying to determine if this is really still a thing. I'm not even sure if people buy these models much in India, and I don't see them anywhere else.

Re: Why?

"A lot of vendors sell budget laptops with FreeDOS"

Do they still? I did a quick search, but I only found one company (HP), they're only doing it in one regional store (India), and they're definitely ripping off their customers because the prices and specs are terrible (27k inr, £255, $324 US for a low-end Athlon and 4 GB of RAM). I can get laptops with the same or better specs in many countries that includes a Windows license and, from more manufacturers, ones that ship with Linux preinstalled.

Brace for new complications in big tech takedowns after Supreme Court upended regulatory rules

Re: It's actually good.

And that's what they did. They wrote a law expressing what they wanted to happen and specifically named a group which was supposed to encode that law into specific regulations and keep them updated. It's remarkably similar to software: someone writes an HTTP spec, which tells people how to talk to HTTP servers, but they leave the decisions about what it will do in certain conditions to the software writers so it can better match intended behavior. That means, for example, that when a new change to communications technology happens, a regulator can approve that change without having to take the whole spec to the legislators and say "I know you don't understand any of the stuff in this, but we would like you to paste it into a law and then pass it".

The normal procedure would be to check whether the regulations passed actually match the authority the law gives. If the law says that the regulator can require fishing boats to carry regulators, but doesn't give them free rein to charge them for doing so, then their attempt to add that regulation doesn't match their powers and they would lose. That's not what happened here. Instead, the courts eliminated that power entirely, meaning that until legislators pass laws that include every detail of how something operates, the court can basically decide at the time whether to approve or deny any regulatory action based on no knowledge of the situation and with almost no limit. The power they're so worried about someone else having they've just taken for themselves. The situation wasn't good before, but it is no better now.

Antitrust cops cry foul over Meta's pay-or-consent ultimatum to Europeans

Re: They screwed up

It's not paying to avoid ads. They could easily have that model and not violate anything. It's the pay or we get access to and permission for all your data at any time, which would already be a problem, and then it turns out that if you pay, they still collect and use all your data anyway. Having a paid ad-free service is something they can definitely do, and maybe they will consider it if they are ever penalized sufficiently to get them to stop collecting excessive amounts of data. I had hoped that GDPR would get us closer to that and it didn't, so I'm not optimistic that the DMA will end differently.

Supreme Court orders rethink on Texas, Florida laws banning web moderation

Quick summary:

"The courts decide that the government can attempt to influence free speech and moderate content."

Translation: They should not be forbidden from having any contact between government and social media, including providing non-binding, non-incentivized*, suggestions. The government is allowed to suggest that some piece of information is wrong, and social media is free to ignore them and leave it there.

"The courts decide that the government cannot create laws to uphold free speech and ensure the playing field is level."

Translation: governments are not allowed to pass a law forcing social media to keep up messages, whether they want to or not.

Do you notice the big difference? The first case involves communication and optional suggestions. The second provides mandates and punishments. If the first case involved government forcing comments to be taken down, that would have been struck down too.

Your choice to describe forced publishing as "uphold[ing] free speech" suggests that you don't need help understanding, you just don't like it.

Re: Waters still muddy?

A publisher has a lot more free speech than a common carrier does. A common carrier is supposed to carry my message, no matter what's in it*. A publisher can choose what it is willing to publish and what it's not. That reduced freedom is why common carriers get special protection in law and publishers just have the normal freedom of speech.

* Not that common carriers are forbidden from blocking. Email providers are often considered common carriers, but if I'm accused of sending spam, they'll come down hard on me. One of the services I've used is very concerned about spam and, while they've never done it to me, they have some scary messages about extremely quick punishments for the smallest of spam reports.

CISA looked at C/C++ projects and found a lot of C/C++ code. Wanna redo any of it in Rust?

Re: Pay no attention to Dunning and Kruger

Relying on the user to solve the problem isn't always the right approach either. Imagine that I built a car with a little switch that would lock the steering wheel at an extreme point, so if you twisted it too far, it would get stuck and send your car in circles if you didn't flip over. People ended up buying these, making a mistake, and crashing into things. I responded by saying that, if they were better drivers, this wouldn't have happened, and I'm right. You still might decide that my car design was more at fault than the drivers were.

We try to improve the quality or safety of things made by tools all the time, and it doesn't always take the form of telling the users of the existing tools to be better and blaming them when anything went wrong. Sometimes, we tried to improve the products by improving the tools, and sometimes, defects were more due to tool problems than user problems. We can't expect that every problem can be fixed by having a better tool, nor can we expect that all problems can or should be resolved by chastising the users.

Re: The tools are wrong.

The difference between a string library of your choice in C and strings in the standard library of a different language is that strings are very common types that lots of things use. Yes, when the code is entirely mine, I can use the library and be completely covered. When I have to connect to other libraries, they will not be using my library of choice. Likely, they won't be using any library at all, because they don't want the dependency risk, meaning that many parts of the program will be using the basic subset included in the C standard library. Is this necessarily a problem that we need to solve? I don't think so, but when you make a statement like "String handling in C is entirely down to which library *you* decide to use", it does not help when it's wrong.

Yes, that is a good thing. My statement doesn't mean that all suggested improvements are necessarily better. There are many valid reasons to suggest that C is fit for purpose, and they can be debated. That's not what I was objecting to. I was objecting to the simplistic logic that any vulnerability in a C program is the total and undivided fault of the programmer that wrote it, and therefore that there is no reason to consider changing it.

Re: The Rust Evangelism Strike Force...

While possible, it is unlikely to give you privilege escalation because GCC doesn't need to run with elevated permissions, is not a suid, and therefore probably doesn't have any more access than the program calling it. If you're already running and you want to elevate, you need a vulnerability in the kernel or something that already has those privileges, and those would be more dangerous targets than a compiler. That doesn't make the compiler bugs completely safe, just less important than a kernel bug or one in utilities that run with different privileges. The one situation that makes sense to me is running on a different machine, so a malicious program could spread itself to a machine on which you compiled code.

"A poor workman always blames his tools", but it does not follow that "he who blames his tools is a bad workman". It may also be that he who blames his tools has bad tools. Of course, the statement worked a bit better when workmen chose their tools, rather than having about three that are capable of doing the job and the selection among them was often made long ago. Dismissing every argument about improving the tools with this logic simply guarantees that our tools, whether they're good or bad, will not improve.

Re: The Rust Evangelism Strike Force...

"how can a buggy compiler hurt me?"

A lot of the answers you've gotten have considered intentional flaws in a compiler or bugs that generate the wrong result, but if you specifically mean memory safety issues, it could still allow an attacker to generate code which executes something when you compile it. The risk is not as high as some other things, since if you are compiling code, you'll probably run it yourself later, but it might be another attack method, for instance one that can infect a build server rather than your development machine. I agree with you though that, as memory safety vulnerabilities go, there are things that are a lot worse than a compiler you run manually.

Microsoft CEO of AI: Your online content is 'freeware' fodder for training models

Re: Copyrights as a structural obstacle

"On the moral side, it is funny to read pro-copyright opinions on the same forum where ad-blockers are considered acceptable."

Copyright has no influence on what I allow my computer to display. They are free to run ads. They are free to try to prevent me from seeing their content if I block those ads. I am free to strike things from the document that gets printed to my screen. Ad blockers are not the same as piracy, and I don't care whether you're a pro-piracy or a pro-copyright person who makes that claim; it's equally wrong.

Re: So we need DCMA 2.0?

"The conceptual problem I see is: how is my reading and internalizing a web page manifestly different from a LLM being trained with the same page?"

This again? Every time, some argument like this is made, and each time, it does so by either misunderstanding or misrepresenting facts. Starting with:

"In neither case are the page's contents reproduced or stored the LLM or my brain (I don't have an eidetic memory.)"

They are stored and they are reproduced, often accidentally. First, they are stored in the training archives, without permission. That is an accurate, byte for byte storage. Then, they are partially stored inside the LLM. True, I can't, even with access to the model, run a command like "llm-extract book-title" and get it back, but it will often print from it verbatim. This has happened, over and over, across models and sources, relevant to the query and not, and it is only somewhat less now because code has been written to minimize it because it makes their crimes too obvious.

"A fairly simple example I would consider is where I train a LLM on the entire Public Domain corpus of The Gutenberg Project say from an offline resource (eg their 2010 DVD.)

From my reading of Gutenberg's T&C I think I would not be in conflict with any of those provisions."

You would not be in conflict with anything, even if you downloaded them fresh, although if you're going to, Gutenberg would rather you used something like their Kiwix versions so their servers aren't stressed and that way you can have the full archive rather than the subset on DVD. This is specifically because the work they distribute is not copyrighted. You can do whatever you want to that data.

"Posing rhetorical questions I would ask what moral or ethical lines will I have crossed at that point? And when I provide free, open access to my trained LLM? Finally when I place a paywall in front of my LLM?"

No lines at all. Public domain training content is fine to use for all purposes, commercial or otherwise. It's other content where those lines appear, and they appear at the start. Training your model on content you don't have the right to is both unethical and illegal.

"Finally how does one legislate ethics and morality? Extant attempts are without exception cures disastrously worse than the disease."

That's what law is. Laws are always intending to codify our concepts of ethics and justice. They have lots of downsides, but unless you think that no law is better, we've already decided to try.

FCC wants telcos to carrier unlock cellphones 60 days after activation

Re: change network provider

That kind of fraud is a problem with any installment plan. If you use one of those split payment things or even a credit card, you have the option to not pay back the bill you said you would. This isn't new and there is no reason why phone companies should have any different experience than anyone else. If someone does it, they have a lot of different ways to collect on the debt, and if they don't want to, they can outsource part or all of the process. There is no reason for them to stop having an installment plan with that minor risk, and if they did, people can use other methods to split the cost. Like the phone contracts, doing this will often mean they pay more than the price of the phone, but they will pay less per month.