Posts by doublelayer 9378 publicly visible posts • joined 22 Feb 2018
And they deny that though it's Facebook, so it'd be entirely in character. Doesn't change my view though. I'm hoping that Facebook manage to trample NSO out of existence. If they could do that by spending massive amounts of money causing them to crash slightly afterward, that's just a bonus.
At some point, laziness is no excuse. Take a related thing. If I install equipment which is faulty and is likely to kill the people who use it, but I don't know that I've done so, that's negligence. Still a crime, but a lesser one. If I know that it's likely to kill but I leave it up, I've committed a larger crime. A thing which is known to cause injustice and is left in place is at least tacit acceptance of those consequences, especially when the alternative, removing the system, is such a cheap and easy action to take.
It all depends on the size and quality of training data and the effort of the developers. If you start with a nonrepresentative sample that hasn't been cleaned up, pump it through the training process until the percentages are high, then sell the result, you'll get something inaccurate most of the time. If you're selling a product though, the number of photos and high test scores are all you're quoting, so many companies do that.
With rigorous attention to detail by data scientists and machine learning experts, you could get something which is significantly better. However, it would be a lot more expensive and it would still be wrong often due to unavoidable problems like poor cameras. At this rate, most people have either concluded that they don't want to do something that will never be even close to acceptable or that, if they're going to be inaccurate anyway, no use spending a lot of time trying to improve. And there we are today.
"Physics isn't racist, it's simply physics."
Sort of. Physics isn't racist and nor are these cameras racist; they have no ability to change what they were programmed to do. Two things in this situation are racist though. First is the AI algorithms which have been inappropriately trained such that they're more likely to be incorrect about certain groups. That's not the program's fault as it's just performing mathematical operations, but it is a fundamental inaccuracy. The larger thing, however, is the use of all this stuff. If you use a camera you know won't capture people correctly and feed that information into a model which you know won't judge people correctly, that's a racist act. You are using tools which have the result of creating unjust circumstances, whether that was the explicit goal or not.
Not only did this not cause damage, but it is designed to prevent damage. Had that been the actual reactor, the company would have been very angry. Because the reactor melted down? No, because it failed safe and the company would lose money. That feature is designed specifically so a "Jerry" or someone doing the same deliberately can not cause a safety failure.
That's not how that works. They need to fix the code, but using a security hole is still illegal. Failing to lock your door means you're more likely to have your stuff stolen and you shouldn't do it, but it's still robbery if it happens. Facebook's failure to produce entirely bugproof software doesn't in any way justify the malicious use of those bugs.
If you get found out, that's a serious crime and you go to the prison the government concerned thinks you'll dislike the most as an example to the others doing it. In reality, you get ten bitcoin. A zero day is a zero day because nobody's got good information about who knows about it, so it's easy to tell someone about it without generating proof that you did so. Not only could they have found out about it themselves, but it will take long enough for someone to find out that they have it that, when that does happen, the exploit isn't useful anymore.
Laundering the crypto without getting found out and landing in the first sentence is an exercise for people who like committing crimes.
Because the people they've seen the malware used against are definitely terrorist material, aren't they? Stuff like this gets sold to governments which don't value their citizens' rights (or anybody else's for that matter). It's not a law enforcement tool. It's a tool of dictatorship, which can be proven by watching who ends up a victim of it.
I don't think that's correct about Windows or Linux. There's just 10 or 11 for Windows, and you spend about ten minutes deciding what license you need. For Linux, there are plenty of options to choose from, but you can decide on one and just stick with it. If you like Ubuntu, for example, you can probably bet on it sticking around for a long time. For Apple, I would say things are similarly easy except I don't understand your complaints. You choose a machine and run their OS, updating it every year unless you have something that doesn't update quickly. In any case, it's not a lot of work unless you want to really optimize things by repeatedly changing to something marginally better. If you want to pick something and not change it, you will probably be fine with a variety of options.
"My water comes out of a hole in the ground from a natural spring. Goes through my house and back in to the ground."
And does anything clean it before you drink it? Do you run your own filters?
"Percolates through the field and right back down the hill. Zero waste."
And do you filter the waste water before releasing it? Because not all the things that end up in waste water will be appreciated by the plants growing on the field, let alone someone receiving the excess.
"It sounds like the government can shut off your water and use that to coerce you."
Er ... I suppose they could. Of course if I was the government and wanted to coerce you, I'd use somebody from the military. Unless everything else you do is also independent to your house, water's not the only thing that could get cut off.
That's true, but just as phones went from luxury to convenience to requirement, mobile phones and smartphones are following that same curve. I would say that mobile phones are pretty close to required now, as people expect that they'll usually be able to reach you even if you're not at home and also that you can send and receive text messages. Smartphones aren't there yet because you still can do most things people expect if you can receive SMS and standard calls, but those are easily into convenience and I predict that they too will become effectively required.
The same things apply in other types of services. For finance, banks were once an option, and now for most positions, the option of getting paid in cash isn't available so you need one. Payment cards went from unknown to optional and now we're seeing more businesses rejecting cash payments. I don't like that last change, but that won't prevent it.
That depends. You can find some information about me if you do research because I published who I worked for. I did that because other employers seem to be confused if they can't find it. That doesn't prevent other aspects of my life from being private. I didn't publish my browsing history, my financial details, or my communications. I don't post photos of myself or places I go. Privacy can mean a lot of different things, and some information can be public without all of it being so.
"So you have to pay for these services (which are not required to use) in some way or other."
That's simplistic or just wrong. A lot of services are effectively required even though we once did without them. Yes, at one point there were no telephones, then they were luxuries. But try having no telephone now and you'll find some things won't work. Getting jobs, for example, can be difficult if they can't contact you. Phone may die and be replaced by email, but if you want to have neither phone nor email, you'll have trouble getting employed. It's subjective how inconvenient a lack of something can be before it becomes a requirement, but there are many things quite high on the inconvenience scale which are privacy risks.
A better method, at least as temperatures decrease, is to use the heating for building heat and make the building services budget pay for the power. Because that would entail reducing the power usage on the other departments, they'll have to add some more capacity so finance doesn't see an anomaly in their own bill. Using desktops to mine is even less efficient, but if that's what needs to be done to keep the budget in reasonable bounds, who are we to complain? Also it's probably not safe to complain.
Thank you. This clarified a lot for me. I remain opposed, but now that I know what you suggest, it makes more sense.
"What I know is that the unregulated 2.4GHz non-ISM use is a hot mess in saturated environments because people simply believe that more power (transmitter output) and more APs is more better."
That's part of it, but the major part in my experience is that there isn't very much of it. It has 14 overlapping channels of which the U.S. disallows 3, so it's not ideal for the use. It's not the user's fault that, decades ago, someone gave a relatively small band for unlicensed use. The larger 5 GHz band did a lot to improve this, and there has already been a third new band created for it.
This is, to me, the only problem that can be solved. If you didn't expand the bandwidth but you had an exam as you've described, I predict that you'd see either nobody following the restrictions or everybody using those bands and having the same level of collisions as before. If a single licensed user can set up an access point for other unlicensed users to connect to, then the ISP-supplied equipment, which represents a lot of user equipment, could be installed by licensed technicians, meaning you'd still have the same number of access points and they'd still have the positioning and contention problems. Your only hope is that basically nobody can use the band, but that only helps the few people who get licenses.
"How does one resolve that in an unlicensed space today?"
I abandoned 2.4 GHz in areas where a lot of other users were on it. I configured access points to only have 5 GHz signals running by default. Devices which couldn't support 5 GHz could still be accommodated by temporarily enabling a usually disabled network, but those are less common now. Not an ideal solution for everywhere, but fortunately I'm not a network admin.
"Amateur space cannot be used for commercial purposes, period. Furthermore, amateur space is not "assigned" to any one licensee."
And good things those are. The bands which are assigned to a single licensee are operated by businesses which sell that access to the public. Allocating more space for a private user's ownership when we have a suitable public space in which that user can already operate is not benefiting anyone. If you have WiFi crowding problems, then you can use a wire or you can expand into the new frequencies allocated for it. 5 GHz is significantly larger than 2.4 GHz was, and we now have new bands for version 6. This doesn't solve all your problems, but it's not the responsibility of spectrum regulators to solve everything for you by taking things away from the rest of the public.
You have suggested the plan for a restricted WiFi band. That means that you have an idea about who should be permitted to use it. I'm not asking you for the exact writing of the exams, but for its general intent. What things would you tell the person writing the exam? How would you establish whether a given person is deserving or not.
This is your idea. You proposed that a restriction be in place. You have to have some idea about who gets access, and even if that's just a baseline of some things that they'd be required to do, it would help elucidate your point. Without that, I only have your vague statements about the "unwashed masses", a phrase which I must tell you isn't helping your point in my mind. It sounds as if you're saying "There should be some people with the special access and I can only tell you that I'm one of them". So far, I have heard no beneficial result from such a plan other than you would get private space and you want it. All this seems to do is to take even more bandwidth from what you've admitted is an overcrowded band for the benefit of an elite class which you can't even identify. I'll vote against.
Using any wireless connection opens up the risk of eavesdropping, but if you let users use WiFi, that's already a risk you have to handle. In fact, if you let your users use a wired connection that's not the business network, you already could have that problem given the places where traffic could be captured. This is a known problem. Acceptable solutions are not allowing any external networks, allowing external networks only with a secure layer like a VPN on top, or deciding you know the risks and you'll take them. Changing from home ethernet to mobile doesn't change the calculation there, as by the time you got to allowing home network you probably needed an encrypted connection already.
"An exam and enforcement, to answer points one and three, respectively."
Slow down. That doesn't answer point 1. What is on the exam? Only people who can answer electrical engineering questions about radio equipment get their private WiFi? That's like the amateur exam. In that case, there are already amateur bands, use those. I get that you'd have a mechanism for proving whatever it is you're proving. I want to know what the qualifications would be.
"Long term I expect WiFi and LTE to converge."
I don't know exactly what you mean by this, but in most of the possibilities which come to mind, I don't expect that nor would I want it. The main reason that merging is impractical is that they're run by different people and connect to different networks. I can't just start setting up my own LTE equipment without licensing it, and if I did, I'd have a nightmare of access control. LTE and 5G standards are great for their purpose, namely having a few open internet networks for large area coverage. WiFi is great for its purpose: having a radio connection option for an existing network. Having LTE which connects through a private network is a recipe for never being sure whose watching your data or whether you're on the secure one.
As for the standards themselves, there's a lot more similarity, but this doesn't mean merging makes sense either. What does make sense is sharing stuff between standards, which can already happen. If WiFi comes up with an interesting solution to a radio communication problem, I have no doubt the next 6G standard will implement or enhance it. However, I don't think they'll ever change the two standards for public and private approach.
"I kinda wish that that there could be a WIFI equivalent to GMRS (at least here in the US, it requires a license to use)."
Tell me, if you were placed in control of the regulator and implemented this, how would you set it up? Specifically:
1. What does a user need to do to prove they deserve a license?
2. How do you, if at all, get device manufacturers to keep that band in their products rather than just drop it and only support the unlicensed ones?
3. How if at all do you restrict people from operating on licensed WiFi bands if they don't have a license?
Point 1 is the most important to me. I can't really think of anything you would do to deserve the license over others. "Educated WiFi users" is not a category that makes much sense to me, and it seems like you're just paying for the privilege of a private band, which seems to have no public good at all since there is limited capacity in that area. I get that it's better for you, but I don't know why the regulations should be changed for that.
"the only way evidence can be used to support a conviction is if that evidence supports that conviction. If they didn't do the crime, the evidence didn't support the conviction, some other factor - *human* factors - perverted the process."
I tell the court that the defendant was matched with images believed to be of the perpetrator, and the algorithm concludes that they match with a confidence level of 99.4982%. This was trained on a database of seven million photos of human faces. A nontechnical juror hears these impressive numbers and assumes the computer must know what it's doing to produce such a precise number and since it had such a large dataset. The juror is not familiar with the technology and doesn't hear the facts that make this less trustworthy, namely these:
1. The seven million photos were stolen off social media, meaning they were taken on very different cameras, subject to intentional and unintentional editing, and are of better-targeted at the subjects' faces.
2. The images were biased toward one ethnicity causing less accuracy on those with different facial features.
3. The software is using machine learning which can't really provide more information about how it concluded various things.
4. The program hasn't been rigorously tested on extra information because that would require retraining which is costly.
5. Machine learning models always produce really precise numbers.
6. Jurors tend not to know how easy it is to do machine learning wrong.
There are a lot of human factors there, but it's still the fault of the technology usage and can be solved by not allowing the flawed technology to be used as evidence (or at all).
"And once again: Is it worse than what has basically always been there?"
Yes. It is very much worse. Mostly because the computer has a lot less data. If you put someone in a line, the person can take their time looking at some people and try to be honest. No guarantee, but they can do it. The computer is attempting to do the same from a moving crowd, possibly getting only a few frames with moving subjects in them, and all without the instinctual visual knowledge about human faces that is learned to some extent from infancy by all humans with functioning visual organs. It's been tried, repeatedly, by different people in many places. It doesn't work.
Hong Kong is not supposed to be entirely subservient to China's central government. It is not supposed to be "just another province". This is not an opinion. There is an agreement by China and the UK which specifies that it's not supposed to be. China has given it the status of "special autonomous region" in recognition of this. And to be fair to Beijing, they used to adhere at least somewhat to that agreement.
Of course we object to China's repression of Hong Kong. I object to China's repression of anybody, but Hong Kong is a clear example of why. The people of Hong Kong were given no control over their politics, either by the UK or China, and they clearly want it. When they object to being part of a country which doesn't respect their rights, I entirely understand why and sympathize with it.
"what happened in Spain with the Catalans being the poster child for dumb and pointless repression -- its not as if the Catalans were going to go anywhere, they're stuck in the EU"
You will find that many do not support Spain's actions after the vote for independence. I would also add that, as undemocratic as those actions were, it involved a lot more justice than China's actions in Hong Kong let alone anywhere else they operate.
"As for Hong Kong's border regime its in the same kind of situation as Northern Ireland is with the UK/EU -- culturally its in one country but physically and practically its in another. Only time will fix this."
No, I can think of a few other things that could fix it. Independence could fix it. Northern Ireland had a chance to become independent and chose not to. They could have that choice again if they have changed their mind. Hong Kong was never given that right and I don't expect China to offer it now. Autonomy could also fix it. The UK doesn't force Northern Irish people to act more British, but mainland China does have policies intending on making culturally-distinct areas change to conform more, which hasn't been very popular among the general population.
I don't think the people commenting here are either denying the existence of or condoning China's human rights abuses. What I see them doing is questioning the accuracy of the claims of malware-laden hardware.
In addition, you will find it hard to purchase certain types of devices without paying some company in a country which has a repressive government. That's not an argument for ZTE, and their history of not updating devices and including bloatware on at least some of them means I'm unlikely to buy from them even if they release a phone instead of a window. If you know of a good way to ensure that electronics aren't built in a repressive country, I'm interested. Until then, it's infeasible to blocklist everything from China and still expect to have equipment.
It has 64 million points of color on the inside surface to give you a wonderful picture from the equally-sized sensor underneath it. Unfortunately, those points make the lens no longer transparent, but you are guaranteed to get that one nice picture no matter how you point the camera.
I wanted to include a second joke here, so I tried to find a unit whose symbol is P. Do we really not have any scientific units doing so? I thought we had used up most of the Latin letters for that purpose.
Under that interpretation, it basically doesn't say anything. I could reverse each line and it would mean about the same. Document the code, but don't let it interfere with you getting it working. Have a contract, but don't let that prevent communication. Have a plan, but don't be rigid. If you assume that both things are necessary, then you get a message that says nothing.
From what they've said and how I've seen it applied, I think they, or at least a lot of them, really did mean to reduce the effort spent on documentation and planning. Even if they didn't, it happened. It's bad that it happened. A balance is needed, and whether they didn't care or those who read it didn't interpret their commands correctly, the result can negatively affect both the customers and the developers. I don't have to blame them for this when a lot of blame can go to management and that's more fun, but neither will I laud their manifesto when it has had such a detrimental result.
The manifesto is indeed very different from its application, but that doesn't necessarily mean it's good. Here are a few parts of it which I have seen go badly:
Through this work we have come to value:
Individuals and interactions over processes and tools
Working software over comprehensive documentation
Customer collaboration over contract negotiation
Responding to change over following a plan
Sorry for the repetition, now I have to take some separately.
"Working software over comprehensive documentation": Yeah, everybody seems to love this. You know what happens? I have to be your comprehensive documentation when people email the developers asking for help (I don't know if not having any support staff is part of the company's agile plan or just something they do). I'm not calling for a weighty manual documenting every line or giving a paragraph on each option, but document what it does and how and keep that up to date. If the software works and nobody other than the devs knows how to use it, it's about as good as it having a bunch of bugs.
"Customer collaboration over contract negotiation": If you have a good customer, of course. If you don't, get that away from me. A bad customer can ask for lots of things that aren't going to work. Whether that's just adding useless features, complex extra requirements at the last minute, or even asking for the impossible, it always happens when you've already done the core stuff. The requirements should be set forth at the beginning, so you can decide whether you can do them. Asking for some minor changes halfway through is fine. Asking for a major feature halfway through is painful but sometimes there's a valid reason. Asking for an overhaul about 90% of the way through is something from which the devs need insulation. Which brings us to
"Responding to change over following a plan": Again, it depends what the change is. Maybe something previously required isn't needed but a new thing is. Respond to that change. Maybe someone had a good idea and you can implement it without pushing out other important things. You can respond there too. But in general, you should have planned for most of the likely changes and you should follow that plan. In that case, you don't have to respond to change every day, meaning you can give the appropriate consideration every time an important change happens. The way management usually messes this one up is to consider anything they decide to be a change to which the developers need to respond. They started caring about something on Monday, so now the devs have to drop everything for it. On Tuesday they don't think it's as critical as they used to, so now it needs to be dropped again. That attitude is appropriate only for bugs which have been newly discovered or found to be more damaging than previously thought. Otherwise, it's a method of moving very fast and going nowhere.
"For companies, there should be a certification to call oneself Agile. Sorting the wheat from the chaff."
I'd like to see some suggestions on how they'd do that. The main problem is that nobody knows what "agile" is. I've theoretically been "agile" for a while. It's not much different than not being agile. It comes with a variety of words which replaced other words and means the same. A meeting is now a standup meeting. Sometimes they're really short. Sometimes it's an hour long and I'm wondering whether I can fall asleep. Sometimes it's an hour long because we actually have an hour's worth of stuff that needs discussion.
To me, any business is going to write the code using the internal habits and culture. Announcing that they'll be doing this in an agile way just tells me what the various things will be called. It doesn't mean that anything is different for the people doing the work, and it certainly doesn't mean anything useful to preventing burnout. An agile company can still demand that someone write something impossible, stay up late to do it, or handle support requests without assistance. A non-agile company can still look at what their workers are feeling and try to prevent things. This is up to managers paying attention to the needs and status of their workers; we're all doomed.
There is Arch's Pacman which now retrieves ZST archives (a fun thing to find out if you ever try to update an old Arch image which lacks the ZST decompressor which is of course available as a ZST archive). There are a few package managers which exist primarily for embedded, including OPKG which has its own format. Although apt/yum and deb/rpm are the most common, there are several alternatives in modern use.
"Am I missing the USP (for the customer, not MS) here?"
Well, sort of but it's not a very big one. The theory is that, in March of 2020 when everybody was being sent home, some IT people had a conversation something like this:
IT1: Everybody here was on desktops because they were cheaper five years ago. What can we do to keep them connected now?
IT2: Buy laptops?
IT1: We don't have the budget for that.
IT2: Have the users come in and take the desktops home with them?
IT1: They might not be able to store them and all the peripherals. Also it would be convenient if we didn't have to deal with all that physical security.
IT2: Have them remote in from something cheap?
IT1: That's a larger support cost if we send a bunch of people something like a Raspberry Pi and they've never seen Linux before.
IT2: Well, I'm out of options.
IT1: You know what would be convenient? Having something where the real computer runs on servers elsewhere and the users connect to that, assuming nobody ever runs into connection issues. That solves the physical security issue without having to keep a bunch of offices with desktops in it.
IT2: We can do that on our servers. Running VMs is possible.
IT1: It would be better if Microsoft did it.
IT2: No problem. I'll just call them and give them the idea, then activate this time freeze device and we can continue our conversation in sixteen months when everyone has gone with one of the other solutions already.
Had they done this in 2019, people would have used it. By now, most perspective users have figured out some other way to deal with the problem.
All of that is supposition. You assume Samsung installs U.S. spyware. You assume Huawei doesn't install U.S. spyware. You assume that theoretical Chinese spyware wouldn't sell useless information. Do you have proof for any of this?
If we're going to state our assumptions without having any compunctions about proving them, here are mine. I assume neither Huawei nor Samsung install U.S. spyware, nor do they install Chinese spyware. I assume the NSA and CCP have plenty of exploits they'll use to target those devices when they want, but they didn't get manufacturer assistance with that. Have fun disproving that any more than I could disprove what you think.
"Genuine question: is it better to buy a superseded top-of-the-range Samsung phone that's now modestly priced, or a new mid-range model like this?"
It depends what you want in a phone. The top range may have better cameras or screens, faster processors, or more memory. However, if it's old enough, those will have become cheaper and the midrange will already have them. The modern midrange will have longer support time and will probably run a later version of Android. The previous top of the range is more likely to have support from replacement ROMs. Basically, decide on your budget and compare the ones available when you're looking to buy, because you can't guarantee what they will have and it all depends on your preferences.
There are a lot of Chinese companies making computers and a lot of non-Chinese companies making them in China. While it's theoretically possible that all those people are making compromised kit, you'd expect someone to have found something about it by now. I require more proof when dealing with such an accusation. If you can prove they produced something with a deliberate security flaw before, that would change things. I know of no such proof.
"No, it hasn't, it's just that many people never understood what makes a workstation a workstation in the first place."
Does that mean that workstation never was a meaningful term or that nobody understands what its meaning is but it has one? If you think that there is a definitive difference between a workstation and a normal desktop, what do you think it is? This could be as big as the great "what hacker really means" debate.
"A machine that can provide only a few cores, almost no cache, and 64 GiB of non-ECC DRAM is just an ordinary PC, at best."
No, it's not. I don't know where you think PC ends, but that's not ordinary for the range. You don't normally get 64 GB RAM in a PC, either a business-oriented or home-oriented one. Eight cores at 5.2 GHz likewise is not an ordinary spec. Compared to a server, of course, it's very different. Also compared to the top of AMD's range with their superiority in multicore chips. Still, that's well above average for a machine a single person is using.
"Yeah that's not helping any. Intel aren't making competitive products these days. If you want a quiet small form factor machine, you want something like a Raspberry Pi 4 or if you insist on x86 you're going to be looking at specialty low-power parts (see FitPC for examples)."
This is getting ridiculous. Intel's competitiveness versus AMD is less than it's been for a long time, but they have plenty of low-power chips. Especially if you're considering the Raspberry Pi 4 as a comparison. The BCM2711 isn't slow compared to those that came before, but many low-end Intel chips easily outclass it. In addition, if you're buying a low-end computer, the ones using Intel's chips are likely to have better thermal performance than a Pi, which needs assistance with cooling due to the size constraints and power of the new cores. AMD is getting some interesting chips in this range too, but this is an area where Intel has been competitive with it while the high end has gone to AMD; in the past few years, AMD has had few chips that could operate in the low power limit and still work well.
"These days my minimum threshold for "workstation" is 16 cores, 64 MiB of cache, 256 GiB of ECC DDR4, and 3x 4K60 DisplayPorts. You're not going to get that from anything in this line."
You're right that this line won't do it, but mostly I have to ask why you need that. Most of the time, "workstation" means single user. What are you doing that requires 256 GB of memory, and whatever it is, you know that few other people are doing that and many who are are using a server to help with the heavy lifting.
If you can run the management software on Linux, that probably helps a lot with the old software problem. It probably doesn't as easily handle the old hardware problem. You can use a USB-to-serial connector if it works for this equipment, but you could do that on something else. You could also create your own converter which uses the GPIO to connect, but again you're taking a lot of work to graft together a connector. If possible, it's a lot easier to run the Linux-based management software on a computer with the correct port already there because then you don't have to. If it's a particularly rare port, then you might test a converter to have as a backup, but RS232 is not that rare.
It depends. There is legacy hardware and there is embedded stuff. Both still use serial a lot. Admittedly, a lot of modern equipment uses serial over USB, but not always, especially if the equipment doesn't have USB support already. I've seen serial ports for console access on a number of relatively modern things and there will be people with even more of that around. Given that, why not throw a couple on the back; they're cheap.
I doubt it. I haven't really had complaints about the fan on my laptop during calls, nor have I seen others with that problem. The one I'm using put the mic on the opposite side as the fan, which probably accounts for some of that. Moreover, a desktop doesn't change the situation much. If you don't want the microphone near the fan, then you have to connect one some other way, and that way will work equally well with a desktop or a laptop.
Only if you accept that there is a "the grand scheme of things", whatever that means. In terms of those things we care about, it's not a blip. The planet doesn't care about what we do, but if we lived for a billion years, it wouldn't care either. If we sliced it in half with a laser just because, it wouldn't care. It's not that we're meaningless to things, they just don't have consciousness and don't have the ability to be interested in us.
Everything which does appear to have the ability to be interested doesn't think we're a blip. We humans of course are quite interested in our own actions, animals complex enough to recognize our presence have changed their behavior in response to those actions, organisms which don't have complex behavior have evolved to respond to our actions. If, eons from now, a different life form views the history of our civilization, it may well be a blip to them. It could also end up being wholly unimportant to them. For us, for everything else we know, it's not one.
They are driven by something with a CPU, and that could easily manage clocking (in fact, it almost certainly does). However, there's no way of knowing exactly how that was configured. Running the management system to avoid failures would make sense for a place that wanted to preserve their investment, but if they could obtain devices for relatively cheap, then it's also possible that they wrote the management software to run them as fast as possible to speed up their mining. Without knowing who was operating them, you don't know what settings they used.
No, it probably wouldn't. Some filesystems keep track of that, but not all do. What if something other than a browser downloads the file? What if it was in an archive that was downloaded? What if it was copied from a drive using something that doesn't have such a metadata field? What if it's currently on something without such a field? If you're doing this at all, you can't treat a file not known to have been downloaded recently as safe.
This isn't useful in this or most cases. The problem this is intended to solve is the use of untrusted libraries by a developer who doesn't audit the library for security or is dealing with a library of such complexity that it's infeasible to do so. If the dev inserts the checksum of an insecure library, your system won't catch it.
If you're really afraid that something will modify a library which your main application imports, you're going to have to do more. If all you've done is insert checksums into your main module, the person substituting libraries can just edit those checksums so they match again. You can do this better, namely by signing your code and not running code from unsigned random files.
People may sue a lot, but you usually need a reason and they don't have one. "He was lying and being a jerk" doesn't qualify. Certain other criminal things might work, but for a civil suit, you usually have to answer two questions to prove that your suit is valid:
1. How were you harmed by the action?
2. In what way was the action unlawful, including breech of private contracts?
If you can't answer both questions, your suit is likely to be thrown out and you can save yourself lawyers' fees by checking before you file one that's going to die fast.
Yes, that argument could be made, and that argument does work if your only metric is whether things could be worse. Yes, they could be. However, the actual metrics used will depend on your personal philosophy of how pollution is to be managed. Some people think that reduction in consumption is important, so Amazon's more efficient delivery of unneeded items is a negative for them. Some people think efficiency is needed, meaning some activities which are more efficient in getting the same result should be done instead, essentially asking "Could it be better" and if the answer is yes, basing their calculus on that. There are many other attitudes people can take for this, but if I tried to list them all, we'd be here all day. And there are a lot of people who think such things until they want something, and then all of that gets canceled.
I don't even know how my own philosophy balances out, and I definitely don't know yours. Still, Amazon's argument doesn't sound very good to me. If you accept carbon emissions as a negative, then it sounds like "We are doing more of the negative thing, but we made more money from it". Phrased that way, it doesn't sound as nice. Of course, that has to be balanced against the benefits the customers get from that process and the benefits they would have gotten from the next best option, somehow valuing your utility, and now we're into that part of economics where you just have to make up numbers because nobody can tell you what they are.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
