Posts by doublelayer 9378 publicly visible posts • joined 22 Feb 2018
I would always answer the other way. Admittedly, I've never been asked that so blatantly, but whenever asked some question about whether I would leave, I interpreted it as "Would you like us to set a disloyal flag on your employee record?" and answered in order not to. The good part about that policy is I never had to decide what I actually thought while filling out their survey.
The TPM is used for more than just locking access to the BIOS. It exists to provide some processing and cryptographic information that is resistant to tampering for multiple levels of firmware and software. It is useful in obtaining that goal, despite having other downsides and many users not needing it. It is more complex than a BIOS lock.
"In post-rentier economics (essentially return to what pertained at the time of Leonardo da Vinci) the end product - be it an idea or digital artefact constructed from an idea - has no monetary value whatsoever."
Ah, Leonardo da Vinci. A man who famously never needed anyone's patronage to produce his work, and whose original art is valued at zero because the copies are exactly the same. If you're going to argue about IP rights and the benefits and drawbacks thereof, it helps not to use examples that directly contradict your statements.
Romania doesn't border Russia. It will be one new NATO border they'll gain if they take Ukraine. Add in Finland (well, possible NATO-joined Finland or at least more NATO-aligned Finland). It's obvious that taking Ukraine will only increase their threat surface with regard to NATO, that they know this, and that they don't care. In other words, their defense excuse was a lie.
I mostly agree with you. However, I take exception to this part: "Realistically, we're simulating intelligence the best way we know how".
No, we're not. Collecting terabytes of text someone else wrote and writing a search algorithm isn't simulating intelligence. We have that already. It's called the "I'm feeling lucky" button on Google, and how many people use that? It's obvious what a text prediction model isn't doing, and if you need to simulate intelligence, you'll need to do at least some of those things. A child can make the connection that teachers are usually human, if they teach mathematics, they're still teachers, and therefore understand questions even if they don't have the answer. This model can't do that. There's a way to check whether it understands the question. Ask a five-year-old how many teeth a math teacher has, and they will likely say "I don't know" because they're unaware of the 32 number. Or maybe they'll guess, but they will use intonation to indicate that they're doing that. GPT won't do that.
On a different level, it's just a hypocritical statement. They say that people shouldn't develop their own model because "it would be massively environmentally costly, compute costly, and we should be trying to share resources as much as possible", but they can develop their own model despite those costs. Basically, what their argument boils down to is "We can build our model, and you should pay us for ours instead of doing one of your own because you will be harming the environment".
This shallow attempt to put down competition might bother me except the product's so obviously useless and I really don't want people to waste resources making more of it.
"The need to tightly control public information is a modern thing."
No, it certainly is not. Let me give you one good example from the time period. You know the Spanish flu? Why is it called that? Because it started in Spain, of course. Except it didn't; it started in the central United States. Soldiers training there caught it and spread it around to other troops, who went to other countries and spread it there, and eventually it was everyone who spread it everywhere. So where did that name come from, then? The answer is that Spain put out a lot of information on the disease and other countries did not. The other countries did not because they were censoring for the war effort. Spain did because it did not participate in the war effort. People got their information about this disease from the one country in the area that wasn't censoring, and the information divide was so severe that the pandemic got a misnomer that is still used a century later. Censorship is not new.
"We should opt to disappear. That way the earth will regain its balance sooner than later without another catastrophic extinction event."
I hate to break this to you, but nuclear war would be a catastrophic extinction event, and probably not for humans. Lots of other life forms would go extinct though.
Also, what balance are you referring to? The planet does not have a balance. Ecosystems vary at all levels, from geological to behavioral. Human activity has been causing a lot of change, and there's certainly a lot of good reasons we should act to stop doing that, but don't pretend that, if we succeeded or stopped existing, that change wouldn't continue. Nature is not a school physics example where everything comes into graceful balance if people stop messing with it. Nature is big, complicated, and in constant flux.
China wouldn't want to destroy Hong Kong's democracy. Hong Kong's usefulness as an international center depends on not being like the many other cities in mainland China that have most of what you would need but don't attract the interest from other countries. No chance they would change the laws there to be essentially the same as in those other, rejected cities and back up those changes with violence.
They also would have no reason to take over and maintain control over the Xinjiang region--it isn't part of China, the people there didn't accept their presence, and it has little to gain by holding it. If they did have control over that region, they wouldn't annoy the locals by locking up some of them without reason. They certainly wouldn't be stupid enough to do that to hundreds of thousands. After all, genocide only makes it more likely that someone decides to oppose you with violence. I would expect a very peaceful Xinjiang.
China is a dictatorship with a leader for life who has all the same problems that Russia's leader has. If his desire for new territory is strong enough and he thinks he can get away with it, he will take what he can. It's not about defense, just as Russia didn't invade Ukraine for defense (Ukraine's military couldn't take over Russia and wouldn't be suicidal enough to try, and by taking it, Russia is only increasing the land border they have with NATO and NATO's level of concern). It's not about culture (Russia doesn't think Ukrainians want Putin ruling over them and China doesn't think Taiwanese want Xi ruling over them). It's not even about resources. It's just about what you can have by using that wonderful toy, the military, that you haven't been able to use very much since you took power. The land first, what can be stripped from it in due time, and any extra clout you might have with someone else who fears you more now.
At least in this case, it won't help. Copyright is 15 years. This means that, should you get a copy of the service manual, by 2030 it won't be illegal to have it. If you had a copy of the manual now, it would be illegal, but would that really stop you? I'm guessing not. The problem is that you can't get a copy of the manual right now. Copyright shortening doesn't help.
You've also shortened patents (though from 20 to 15 years so not as much). This should help with the parts themselves, right? After 15 years, it's legal to make generic versions of their inventions. If you don't have the designs for those parts, however, you may not be able to manufacture replacements. Most of them aren't available because they weren't patented in the first place (parts that aren't inventive don't get patented). Even for those where you have the designs or can create your own, the problem is that the system won't accept them. The software (for which you don't have the source) has ways to check for valid parts by using a cryptographic key (which you don't have) or by requiring an unlock code (which you also don't have). Patent shortening doesn't help.
Cutting copyright and patents, even if you did it right now, would only partially help the problems with the right to repair. Doing so also brings some potential downsides, including more secrecy with innovation. You assume that your taxes pay for all the innovation, and you assume that, if legal protections are removed, businesses won't compensate with weirder home-built protections. Both assumptions are incorrect. Jumping to seemingly easy answers appears fine, but only by finding the real problems and addressing them correctly, as these farmers are doing, will actually get you to your destination.
First problem: It's big, complicated, specialized equipment. The other manufacturers have been building that for decades. You have been building it for never. It takes a lot of knowledge to build that properly. You will need to design the stuff, get a manufacturing capacity, and then test it a lot. That is not cheap. It's not like the Framework laptop, where they just had to design the enclosure and boards with standard components used by twenty other manufacturers; you're looking at a far more complex business and they can build in bulk and you can't yet. Where do I go to get someone to put up millions for that, especially considering that this was only the first problem.
Second problem: You have to sell and support this everywhere. You may start in one place, but even if you do, you have to be able to sell the equipment to the farmers who want to buy it. That requires people who can demonstrate its benefits, answer questions, and get it out there. When you've done that, you have to support them. You have to have repair parts for sale, repair techs for those who don't want to do it themselves, people to diagnose problems at short notice. It only takes a few failures for farmers to start telling their friends that your company talks about the right to repair because you don't bother having any repair services, and then fewer people will buy.
Third problem: Scaling is hard. If you can build one big farm machine, you still don't have plans for all the other big farm machines that are in demand. I can't tell you which ones you need, but I do know there's more than one of them. Now you have to start doing all the sales and support work you just figured out in lots of places. Keep in mind that even the large, established players don't do that everywhere, and you can see that it's an uphill climb to have a company with the same opportunity to sell as they have.
Fourth problem: They don't want you to win. Up to now, I've only pointed out the problems you have running your company without them even noticing. Let's say you managed that. You can design, manufacture, sell, and support lots of nice equipment. The companies that currently make a lot of profit selling competitive equipment don't want that to keep going. They have more money than you do. They may decide it's time to prevent you from succeeding. They could give your local partners incentives to stop selling and supporting your stuff and switch to theirs. They could cut the prices on theirs, knowing that they can make it back by charging for repairs and you can't (yes, it's worse for the farmers who buy from them in the long run, but not everyone knows that). They can get some of your stuff and research the problems with it, then use that to convince people not to buy. If you're luckier, they come to you and offer just to buy your company. You get money and stop being in their way.
It's not impossible to do, but you need a lot of resources, including time, money, and effort to pull it off. The rich investors don't particularly want to pick a fight with rich companies when they could fund something with fewer enemies. Not many people want to run that company either, knowing that they have to work really hard just to be one of the pack.
Except, through a mechanism unexplained, those portraits had memories and personalities. This system won't have that, so it will either just say what it was told to say, making your deceased loved ones into puppets, or use some GPT-style bot to decide on things to say, turning them into clones of a malfunctioning robot. I don't think the business spent much time deciding whether this was a good thing. Most likely, their customers will get something they didn't expect which will be disconcerting or harmful.
I love how Nvidia failing to update drivers for Linux kernel compatibility is now Microsoft's fault. The Linux-related comments are on Nvidia and I'm not here to defend them. While I'm not here to defend Microsoft either, I must also point out that the comments about the TPM are missing the point about what that is there to do. I think that requiring one for Windows 11 is useless and generates unnecessary obsolescence, but nobody claimed that having it would in some way prevent malware. You appear to be complaining that the chip doesn't do something that it was never designed to do, and if it did do you still wouldn't support. This leads me to think that, instead of raising actual problems, you are just angry and would like to blame Microsoft for anything that doesn't work as expected.
I try to not have to recover. I've had HDDs that likewise failed catastrophically, and that wasn't any better even though the constant clicking made it clear that I could not bother with the normal hardware recovery methods. By the time a drive is failing, there's a good chance it's not going to work at all, so I don't trust that I can get things back at that point. If it's important enough, I have the disks set up in a RAID configuration and have copies of their contents elsewhere.
There are a lot of parts and powers of the United Nations. The sanction regime against Iraq was implemented by the UN Security Council, which voted 13 to 0 in favor of a set of sanctions. That is a strong vote. A vote of that nature would fail now because Russia has a right to veto resolutions in the UNSC. The general assembly, where they cannot do that, doesn't have as much power as the Security Council does. Had one of the permanent members vetoed the Iraqi sanctions, Iraq would probably still have been sanctioned by the others, just without the UNSC stamp of approval. The fact that all five permanent members and eight other countries thought they were warranted only strengthened the coalition.
I am not here to defend the UN's effectiveness. It is a bureaucratic system with advantages and disadvantages, and because it relies on member countries to do everything, it is weak. For example, even when it passed the sanctions on Iraq, some members could opt not to follow them. It also doesn't get to decide on its own what sanctions are legitimate (the sanctions on Iraq were proposed by UN members, but only gained UN force after members voted to support them). The UN is not a final arbitration system, nor a world government, nor anything else that would give it the power to decide that one sanction is legitimate and force the revocation of another. It is a system for coordinating international action only.
"The Koreans were basically asking the US's permission to continue selling stuff to Russia. Which is pretty much kowtowing to the US's decision, since Korea is supposedly a sovereign state,"
You misunderstand. Basically, what happened boils down to this:
Korea: If we export phones, will you be angry with us?
U.S.: No, that's fine with us.
Korea: Would you mind making that official?
U.S.: We'll write it in the sanction description.
Korea: Thanks.
They can do whatever they want. They wanted the U.S. to support their action, and the U.S. did so. Had the U.S. disagreed, Korea could export anyway and deal with any complaints raised by the Americans.
"According to you, a free, independent country can not chose to remain neutral and must instead take orders from US"
That's not what they said and you know it. They disagree with what South Korea is doing and has done. They did not say that South Korea's right to do so should be revoked. I disagree with your choice to make blatant misstatements in service of a national stereotype, but nonetheless I would object if you were banned for saying what you did.
"I'm not a huge fan of the U.N., but isn't this the sort of thing that it is supposed to take care of,"
No. The U.N. does not have the authority to decide what sanctions you use. They wouldn't be supposed to do that anyway because they aim for neutrality. So on no account would the U.N. get involved in what is sanctioned and what isn't other than votes to approve or disapprove which everyone votes on and nobody cares about afterword.
"instead of leaving everyone kowtowing to the US's decisions?"
South Korea is an American ally. They can ignore the U.S.'s decisions, but they choose not to. The U.S. is going to do what it wants and South Korea has chosen to work inside it rather than go on its own, which it would be perfectly in its rights and able to do.
"Sometimes analogue's simplicity is it's strength. Assuming no disaster, those negatives will likely outlive me."
Don't give analogue media the credit when the problem is physical construction. Your CDs don't stop working because they're digital. They stop working because the materials used to manufacture the user-writable disks weren't designed to last a long time, either on purpose or through oversight. There were lots of analogue media types that were similarly fragile, and you're not considering them now because they have mostly been discarded as broken. There is digital media that is designed to last a long time without being copied to something else (which with digital media is easier than analogue).
In addition to there being lots of problems with cars, they also do a lot less than a computer does. Users have so many problems because computers can do a lot of things. If it was as simple as a word processor and no other applications, you'd see a lot fewer support requests. When you demand a box that can run any program that has been written for the OS and processor and run them all at the same time, with the same files, with some level of security between them, and with sufficient options that a technical user can make that box do any number of tasks, you end up with a much more complex system than a car that can go, stop, change speed, and turn.
It isn't unreachable. Quite easy from a normal internet connection. It might be unreachable from Tor exit nodes, because it's using a redirection service (found that by looking at the scripts pulled in from the first HTTP request). Those tend to block Tor because they don't trust people coming from there. Or perhaps, since you don't appear to have checked normally, you just lied about that. I see little benefit in bothering to check.
We certainly could do more. I am surprised to see that we have done what we have given our tepid responses to the other crimes Russia has committed recently, so I may start off with a different attitude to you, but we both agree that there are things we could be doing that we are not. I still take exception to those who imply that our countries have done nothing or will do nothing when that's obviously false, as well as with people who extend that argument into hypothetical areas that would be clearly different.
They have the right to do so. There is no international law against censorship, at least anything that's enforced or phrased any more clearly than "censorship is bad, signed by censors". Depending on their local ISPs' control over the networks and internal systems, they could try doing that already. I wouldn't be surprised to hear that they are trying it but finding it harder than they thought to implement at short notice.
You clearly didn't understand what they said. I'll summarize it. If nuclear weapons are involved, the internet situation isn't the problem you need to work on.
And by the way, we wouldn't do things for Putin if he threatened to nuke us. If he threatened that, we would threaten to nuke him in return. It gets kind of scary, but that's what has happened each time and would happen this time as well. People are cowardly when it's someone else suffering but can turn strong when it's themselves at risk. Those who don't switch get replaced when public concern gets high enough.
"The problem is that when one nation uses that same network to attack another nation, that's when the neutrality gloves need to come off. In order to maintain that stability you need to defend it from serious abuses like the very one being done by one nation against another."
The problem with that is that Russia's information war would not be stopped or even dramatically harmed by carrying out Ukraine's requested action. They have propaganda sites ending in .ru and with servers based in Russia, certainly, but they also have propaganda sites ending in .com and .ua with servers in lots of countries. As each one is identified, the hosts can be targeted and requested to shut down the site or revoke the domain. Even if they wanted to set up this shutdown and had the authority, it wouldn't solve the issue. Doing it, however, would have some dramatic side-effects.
I think nearly everything should be done to get Russia out of Ukraine and charge them with war crimes. If an action will cause significant harm to unconnected things and also not work, let's not waste time trying it. Let's spend our time finding an alternative which corrects the latter and ideally the former.
It depends what data is included. It's usually a database of all the users passwords, but if those passwords are hashed, they need to break each one independently. They get them one at a time, but since they have the cracking computer running all of them, they can get many done in a day if they're starting with the low-hanging fruit. If the company was stupid enough to store them in plaintext, then they get all of them without doing any work.
The other problem with captchas is the set of users who don't appear to have a clue what they're for. They put them in weird places. I had one that required me to fill out a captcha on every login. Not even just logins from new places (that is also not a good use), but every single one. I think they viewed the captcha as a 2FA solution. That business lost me as soon as I could migrate my stuff off them.
"Given a sentence for not disclosing a password or encryption key might be less of an offense than what they'd have if they could have a nice rummage through your files."
What crimes are you expecting they'd decide you committed by looking at your files? I agree that I don't want them to look at them, but serving up to five years is a strong penalty, exceeding many normal crimes. Having a secret delete password is destroying evidence, another crime they can charge you with. I probably have the same view as you, that such laws are illegitimate and would be better repealed. While they're there, however, it's important to know what the law says so you can act accordingly.
Well, at the risk of causing a debate over which countries are civilized or not, you are not correct. Some countries do protect passwords in the way you state, for example the United States, but in others, such as the UK, failing to disclose passwords, encryption keys, etc is a crime in itself for which you can be imprisoned. This is the case even if you are exonerated from the original investigation.
My response would be somewhat different, namely that either they need to work on making it easier to update Android or they need to have a regulatory incentive not to have such a short planned obsolescence time.
Google has been talking about lowering barriers to updating Android for years. And, as useless as many of those attempts have turned out to be, they do appear to have made some improvements. Still, it's annoying that they have a situation like this at all--after all, nobody needs to ask whether the new version of Linux (kernel or distro) will support a particular old laptop. If the laptop had drivers already, it probably still has them and you can go get them. If it didn't, you may have network, sound, or graphics issues, but the problem will be the peripheral hardware, not that Intel or AMD no longer support the chip from 2009.
And I do tend to blame the manufacturers of the devices to some extent. There are projects that manage to update Android past the typical cutoff point. They don't need to wait for the SoC manufacturer to do everything for them, and if for some reason there is one where they do, then they should know not to use that one. There's a simple way to check which one Planet has turned out to be: how were their security patch level updates? They don't need SoC assistance for those, and if they stopped releasing them, it's clear they moved on and let their product's Android version die.
And this isn't cryptocurrency, it's a straightforward fraud. Had this guy said he had a bot that would trade on the stock market, it would have gone the same way.
I've said this before too, but a Ponzi scheme is a very specific kind of fraud. Cryptocurrency has seen a lot of fraudsters, often not doing a Ponzi-style fraud. Continually saying that it's a Ponzi scheme when it's a different one only makes it seem like you don't understand what Ponzi schemes are or what cryptocurrency is. It's as if I started calling all crimes burglaries; they would still be crimes and I'd still be correct that they're bad, but I wouldn't convince people because I'm continually using incorrect terms.
No, it really wasn't. By that point, petroleum wasn't as critical to the world as it is today. In fact, it went the other way--because of World War I, petroleum became more commonly used, for example replacing coal as the fuel for most naval ships. The participants in the war weren't major petroleum producers. The closest you get is Russia, which produced it but whose production wasn't targeted, the Netherlands, which produced it from their colonies in Indonesia, and the Ottoman Empire, which didn't produce it but could threaten the place (Persia) where the British were getting theirs. The land that changed hands wasn't taken for oil wealth. It turns out that the land lost by the Ottomans had a lot of it, but that wasn't known until later.
"Arguably, without FTP, or something like it, The Internet (as used by consumers) would never have been built."
The key part being "something like it". I don't have much objection to FTP, which is probably because I never had to write an FTP client, but if he hadn't made that, someone else would have used a different protocol to move files. It's kind of obvious that, if you have files and you have a communication system that can send data, you will eventually want to put those together. FTP is important historically and in many modern-day situations as it's one of the oldest protocols commonly supported, but many of the protocols used by those consumers have superseded its function in a different and often simpler way.
Not the part where you want to get data from a different system or put your data onto one. That requires the machine-to-machine bit. That is one of the levels of the internet, but if you want to limit the definition only to OSI levels 3 and 4, it's still critical to lots of systems and has seen many refinements. It's not done, and it never will be done, but it's impressive how well it tends to work.
On that basis, it could as well be a folding phone. In fact, it's probably more likely to be talking about that because the phone mostly has input devices on the outside (touchscreen, cameras). Smart glasses, while they may also have cameras, will need a lot more effort put into the output devices (screen that is easily seen by the wearer without distracting from the environment, same for audio because most glasses don't go over the best areas for bone conduction). I'm not sure how much you can outright lie about what your product is before a patent doesn't apply to it, but as none of these three concepts is patent-worthy anyway (the Pi 400, many Android folding phones, and Google Glass respectively as prior art), it probably doesn't matter that much to them.
I like the way you think, though.
"Anyone noticed that _nobody_ is building a device like this right now?"
I know of a few places making such things. Running Mac OS, no, but the hardware exists. However, you've asked for a different product so you can bring your own keyboard, and there are lots of those. Tons of small desktops exist out there. Depending on exactly how small you want them, you can quite easily get ones that mount onto the back of a screen and have Intel or AMD processors sufficient to run Windows or Linux with pretty good speed for average workloads. There are even a few computers still using the form factor of the old Intel ComputeStick, but I wouldn't recommend them.
Apple can build a computer in any shape they like. It won't make it patent-worthy, and if it's what you've asked for or what their claims appear to describe, it won't be something otherwise unobtainable.
Yes, but they still need a computer that is trusted to access that data. If the company has purchased laptops, the user just has to move that one to have an identical environment without needing a second computer. Most companies I've dealt with don't trust users' personal computers for the required access.
As for phones, I'd like that to be true, but the software often restricts the utility as a desktop device. I'm not only talking about things that require specialist software on the client, but also more normal office tasks. I have never seen a company attempt using phones in the role in which laptops and desktops are routinely employed.
As I understand Apple's concept, it's just a laptop without being useful when the user doesn't have desktop peripherals, which seems unhelpful to me. The only possible benefit I can see is if they were able to make the hinges work so well that the keyboard and all the computer components could be pocketed, thus making it more portable than a general laptop. I don't think that's happening, and even then it wouldn't be that useful.
If they argued that, it would fail just the same. They would have to have patented it then to claim ownership. If they did that, protection would have expired in 1997 and anyone could copy it. So that would be a fun argument to watch them play out.
Me: "a user who already backs up their data"
Reply: "And I am Florence Nightingale. If they do, it's on to an USB stick at best. Which will get overwritten with the ISO to reinstall their system."
Granted, but this isn't really a solution to that problem. Failing to back up is a problem in lots of cases, and reinstalling the OS is the simplest of them--the data's still there and you can back it up just for that case even if your routine backup option is lacking. This argument is a lot like if I said "Never use SSDs, only mechanical disks, because most SSDs will immediately remove data when it's deleted but mechanical disks won't so you can use undelete tools much more reliably". It's technically correct, but there are advantages to the alternative, and if people are relying on that relatively fragile protection instead of properly managing their data, it's just a matter of time before it backfires.
The rest of your examples are good ones but fall into the category I previously mentioned where you are doing it for a specific goal, readily understood by the user, and requiring some technical skill to use.
"On every laptop I've used I always had /home on a separate partition, and just NEVER have I run out of space on / or /usr due to stuff happening there."
Yes, that's probably the most normal situation. After all, you're unlikely to suddenly decide you need tens of gigabytes new software one day. However, if you ever did, the partition situation could make it hard to install that. If you kept a bunch of space around for the case where you did, then you're unnecessarily restricting the space available for /home.
In my case, on the systems where /home is sharing with /usr, I don't pretend my usage is normal. I have an old device with a 128 GB drive which is used for a bunch of random, at times experimental purposes. I decided to test out the Gitlab CE software by installing it on that (for those who haven't done this, it brings its own Nginx version, a bunch of Ruby libraries, is rather large in itself, and is designed to store data outside the home directories). When I was done experimenting with that, I removed it. Having one partition gave me the flexibility to do that without repartitioning either time. You don't have to do that, and if you don't or have a larger disk (I got the laptop for free from a user who wanted it junked), you probably won't encounter the situation very often.
That said, try explaining to a standard user why they should do that. Not with the separate disks situation, not with a server, just a standard user with a single disk that all the partitions are going to go on, a user who already backs up their data. What benefits can this provide that justifies the extra complexity, complexity that is absent on lots of other systems and optional here? The best reasons I've heard that apply here are as follows:
If /home is separate, you don't have to replace it when you reinstall. This assumes that you'll be reinstalling a lot, which they probably won't, and it also assumes that every installation will be smart enough to figure this out, not repartition, not create a new partition for /home, and correctly mount it later on (this last one is mostly an editing of fstab if it doesn't, but if you have to manually reconfigure your filesystems and don't get anything out of it, maybe not so useful). A correllary to this argument is that it's harder to trash the /home partition. The user can point out that the chances of breaking their home directory are already quite low, most likely ways will use the filesystem instead of the partition and won't care where the home directory is, and that they have backups to deal with both.
There are lots of good arguments for keeping /home on a different partition. In each case I can think of, it's a case where, if you ask "Why do you have it partitioned this way", the user will have a specific set of benefits that doing so provides them and that they know about. It's possible that there has always been a clear benefit that keeping the partitions separate provides, even when a user isn't doing it to implement a desired feature, but if there is, I do not understand it. I would welcome learning that lesson, but it should be better than "We do it this way because most Unix servers do it this way and, even though they have lots of reasons that don't apply to this, we must follow them".
Ok, I think I missed that with the talk about hidden folders. I missed it because configuration often does go in a hidden folder and ... Windows doesn't put the user data in hidden folders. It puts it wherever you set when you're in the save box. That different programs have different default locations doesn't change the fact that you can set any of them to save in any directory you please and most of them will remember your defaults for next time. I don't like it that Outlook thinks that it should save attachments into the documents folder. I treat saved attachments as downloaded data until I organize it, so the downloads folder is far more sensible of a default. So when I first save an attachment on a new installation of Outlook, I point it at the downloads folder and it remembers that. If I want it to save the attachments to a downloads folder on a different disk, I could select that as well. No hidden folders or forced organization involved.
The problem is that it really can't. Email clients can easily categorize by such things as sender, recipient(s), thread (subject line, in reply to, or checking for copies of previous messages), dates, and if they're willing to add even more storage, some keyword database for message contents. That's great, but it's also what I can quite easily do from the client's advanced search controls. I don't need the client to categorize my archives by sender because it's pretty easy for me to do that manually if it comes to that. The categorization that could be needed is more advanced processing of the message content to link related messages together. For example, if I am included on thread A about something, then start thread B about the same thing but with different people, and later am involved in thread C which reports a problem with the thing implemented by thread B, those things are related in my mind but will not be linked in the software. The intelligence to identify that link is usually absent, and if you try to build it in you end up with all the standard language processing problems and likely won't get useful results. If the messages were crucial and required frequent cross-referencing, I would perform that organization manually, thus having the certainty that I've done it correctly and in a way I can quickly navigate. When you consider that, at least in my example, all the effort spent on this autocategorization is for messages the user is keeping around because space is cheap and with the expectation that they will be unneeded, it is often not worth it.
"A non-trained person is unlikely to be reinstalling their laptop OS."
They don't have to. On literally every commonly-used consumer OS, the primary storage is treated as a single unit. If you are low on storage, you could delete some personal files or uninstall some software to free up storage, with the free storage being usable for either. You do not need to reinstall the OS to be in that situation. A user who does that on Windows, Mac OS, IOS, and Android is unlikely to accept Linux if they think they have to do it all separately.
"If someone trained does do a reinstall for the user they're not going to be happy finding it was on a single partition."
Not necessarily. In my opinion, a lot of systems for personal use should use a single partition. Programs are data, just like your documents. They're stored in different directories to keep them organized. That's good enough. There are many reasons to have separate partitions for servers, most of which don't apply at all to a personal machine. You can also have as many extra disks or partitions as you like with software on them if doing that has a benefit to you. Windows and Linux make it really easy to point to installed software that's on a different disk, even if it's not mounted at a typical path for binaries. Even if you know what you're doing, there's not a lot of benefit to having separate partitions for different parts of your root filesystem when they're all sharing a disk and all have to be present for your system to work.
A note, this doesn't necessarily extend to all parts of the system. Having /var on a separate partition to deal with something that doesn't handle it correctly makes sense. Having /home on a separate partition so it's easy to share between multiple booted OSes or persist after large changes does too. In both cases, it's a thing that is useful to a user who understands why they're doing it and can configure it themselves.
Take the placement of /home on a different partition. I've just explained two reasons you might want to do it. If you're going to make any big changes to your root filesystem, having /home partitioned insulates it from damage. You can install a new OS without having to move anything. However, you could always do what everyone else does: back up your home directory, reinstall the OS, and copy your files back. You should be backing up already, so it's not quite as bad as it seems if /home ends up sharing. Many of my systems do have /home on a different partition, but often because it's loading from a network FS or external disk. On the self-contained system that uses one internal disk with one OS on it, I have kept /home on the same filesystem so I can use the available space for whatever kind of data I want.
"You could just check programmically the folder was a sub of the desktop and not the desktop ."
It could have been in the documents folder, home directory, or somewhere else depending on how the user did it. You would have to just check it against a bunch of too general directories, and if you started down that path, the user could always find a new wrong one.
"or if it were me I'd have nominated a specific folder on the desktop for them to put stuff in ."
If you were doing that, just nominate the network folder the stuff is supposed to go in. The program is supposed to file things for the user because they were either slow or incorrect when doing so on their own.
The world isn't yet in agreement that Russia should be cut off. They haven't even managed to get SWIFT cut for Russia, though they claim the financial industry sanctions are of a similar strength. From that, which they've used before, to actively trying to cut off international communication which hasn't yet been used as a sanction is a big step. In addition, you would need a lot of countries to work together to get that done. If only one country does it, it only blocks routing of packets from them to Russia, but doesn't do anything to people who are willing to route into a country that isn't on board. Since Russia has allies*, the decision would quickly have to be made whether to include them in the blocking, ignore them as only a few types of people would be able to use them, or convincing them to change their mind. It's harder than it looks.
*Russia has allies: While they don't have allies that are also fighting in Ukraine with them, they have a few countries that actively support their claims (Belarus being the most obvious example). They also have a lot more countries that are willing not to do anything against them, including both China and Pakistan quite actively. Both are large countries with active internet presences through which Russia can route some traffic.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
