Posts by doublelayer

Here's who thinks AI chatbots will eventually be smart enough to be your coworker

Re: I have a question:

Also, you don't need any AI to do this. The easy cases are already fine to handle with basic software. I need a meeting with four people in my organization, so I go to whatever calendar setup we already have, add all recipients to an invite, and pull up the shared calendar feature and look for a time where everyone says they're free, then try that and see what happens. Of course, this mostly doesn't work, but not because they weren't free when they said they were; it's sometimes the problem, but usually the problem comes earlier.

When this doesn't work, it's usually for one of a few reasons that AI won't fix. It doesn't work because everyone's in some meeting and there is no time when all five of us are free that is feasible for our use case. Or there is a time zone conflict, and trying to keep this in normal work hours for everyone is proving infeasible. If I'm the only one on this continent, it's usually fine for me to take a meeting at a weird time so that it works for everyone else, but if I have a colleague in the same position, I would probably not do that without apologizing to them and getting their permission to do something like that, which an email probably won't do. Or it doesn't work because one of the four doesn't want random meetings scheduled, so they've blocked out multi-hour chunks as busy time when they are working but could be available, but I won't know that without asking them. How hard I have to ask them depends on the importance of this meeting and their presence at it, and their reaction will depend on the same thing. If it's me organizing it, it's probably pretty important because I hate meetings, which also means my colleagues are usually happier to attend something I've organized, but the AI won't know that, nor would their responding AIs.

This all gets more broken when I drop the "in my organization" part of the premise. The AI doesn't have access to the calendars of anyone outside the company, and if I have to meet with them, then we're dealing with a lot of possible ways to ask for and get scheduling information. The AI could probably deal with some of this, although I'm worried that it will get confused about someone's time zone data and make a complete mess when I'm not looking. There's a reason there are at least two companies I know of who offer a calendar reservation application as a product. An AI that lacks information will do it badly and giving the AI sufficient information to manage it is a sisyphean challenge, both in the "a lot more work than is justified" perspective and the fact that it will never end.

What comes after open source? Bruce Perens is working on it

It sort of would, but it would break basically everything. A Red Hat that charged for the service they provided and provided everything to the community without any roadblocks would also be disallowed by this license, since they just say that paying for an operating system is what you're not supposed to have done. So it doesn't really achieve the goals of freedom, just of not paying.

Consider how weird it can get. I have had a free Windows license. No, not what you're thinking, a real license code issued directly by Microsoft to my name in exchange for no money. They have programs for educational and charitable use that have such things. Technically, the one I used did have some money involved, but it wasn't mine. Some of them can be entirely free to the person and the organization. Theoretically, I am free to run that code on that Windows installation, am I not? Similarly, did I pay for Mac OS? I did a while ago when Mac OS upgrades cost money, but since 10.9, those are always free. I just bought a computer from Apple and the OS happens to install on it if I boot to media*. So doesn't that count as a non-paid-for OS? In practical terms, we all know that I did pay for it, but I don't have a purchase agreement for the software or a license code for it, so I can argue that I didn't. Loopholes like this are why a license needs to be clear about what they're going for and specifically write to that effect. Oh, and no matter how they write this intention, it will still be a non-open-source license and I won't be very pleased about it, but that's a separate issue.

* Mac OS installing from media: This claim is only valid for Intel-based Macs. ARM-based Macs can do that only if the hard drive is not damaged and has firmware on it. Be careful.

CLIs are simply wizard at character building. Let’s not keep them to ourselves

Re: GUI-Based Admin

"Imagine you have 1,000 servers, and have to double-click on the server's icon to open up its admin GUI."

Not the right comparison here. Imagine you have 1,000 servers, and have to activate your SSH client, decrypt your public key, and enter the root password to elevate to run your script on each one. That would be painful too. That is why nobody who has to do something on a thousand servers does that. People who operate on that scale have a lot of tools to take an action on a large set of hosts. That can at times be a CLI program (take this script and run it on all hosts), but frequently is not. There are GUI-based tools that allow you to change the configuration of a set of servers too, and many of these are popular because they allow the admin to visualize the effects of a change: how many machines have completed it, how many errors have happened, are error rates changing, etc. However, there are even more interfaces for such things, including ones that deploy all those changes as code in version control, which often have multiple frontends with different capabilities, but those frontends are mostly for retrieving information, not setting it. You seem to be alleging that GUI administration involves handling one computer one by one and CLI administration is not, but neither generalization is correct.

Re: -h or --help

So your complaint is about the lack of --? That's really not the important part in all of this. Lots of commands have different syntax for their parameters. For example, do you do --param value or --param=value? It doesn't really matter, because the man page tells you that. Similarly, --install or install makes no difference. IP is different, but it doesn't really change the mess you're going to have regardless. IP could be split into different programs, and each program's arguments could be simpler, but then you'd just have thirty commands for doing something to a network interface and the complexity would be figuring out which of the thirty you needed. It would be no easier than figuring which of the many functions of IP you want and how to invoke that.

Re: Intuitive GUI? My arse.

I made two points in two comments, and I think you may have mixed the two to disagree with them. The pdftotext example wasn't about a GUI being superior, it was to demonstrate why "man -k" is not a functional solution to finding a command you didn't know about. And no, it did not work well for me in my example. I only found the command because I already knew it existed and was able to figure out what the search term needed to be to make it come up. I pointed out why several logical choices that a user who didn't know the command existed would make completely failed to find the command. I had a second example which I decided not to include because it would have been similar, just a lot longer and less successful. You are correct, however, that the GUI would not necessarily solve that problem.

However, while a GUI won't make it obvious what every installed program does, it does make local option discovery easier for the unacquainted user. You can pull up contextual menus on an object and get a list of actions that can be taken on the object. This is quite useful in finding out what actions you can take, whereas the command line often doesn't have that because they expect the user to understand which commands you can send streams to and what they do. The original example of how to shut down Windows demonstrates this. I don't know how long it took them to think that the big button opens a menu and maybe the shutdown option would be in there, but since there was no other system menu, it probably wasn't very long. Once they opened that menu, they could explore all the other things in there and, knowing what options they had, start to use them. If you don't know what options are available to you, a GUI usually tells you that information more quickly than a CLI does. That doesn't make the CLI bad, it makes it harder to learn quickly, but since the CLI also comes with a lot of power and convenience once you've done that, it's still a valuable skill to have.

Re: How I see it

"There probably will be times when you really do want to click yes or no to every instance of a word you are searching for and replacing; but the rest of the time, you can't beat sed."

Or the "Replace All" button? Which doesn't require your editor to close the file before you do it.

Re: -h or --help

"That’s against the basic philosophy of Unix: “one command, one function”, but it has become the norm for modern tools, and we’re stuck with it now."

I think that makes the most sense. A package manager should be able to manage packages, which means that it should be able to install and remove them. Is "apt install" and "apt purge" really that worse than "apt-install" and "apt-purge", because that's what you would get if you insisted that the app-install command be unable to delete a package. I don't think the philosophy was that each minor difference in function should be handled by a completely separate program, but that programs shouldn't succumb to scope creep away from their initial function. Installing and deleting packages is one function. Starting, stopping, and reporting on a service is one function.

Re: Intuitive GUI? My arse.

"GUI design is opinionated...which means you don't always get access through the GUI to all the possible features and combinations, only what the developer decided that you needed and was willing to expose."

That's all interfaces. Developers of CLI tools have exactly the same problem. They run out of single letters for flags, the new option will conflict with an old one and users will complain if any part of the interface changes, the new option requires them to come up with a complex command syntax and write a parser for it and they really don't think the users will use that interface enough to justify that work, or they don't want to write the hundred lines that the new option will require to reject invalid combinations, so they leave it out. That is if they can, for example the other options are available in a config file somewhere or they're writing a frontend to something someone else wrote which has the option. Whenever you run a program, you are going to get the options the developer chose to put in it. A GUI doesn't necessarily have any fewer options than a CLI, and the best way of knowing which one has more options is to know which one came first because it probably got the canonical set, while some developer figured that their version could skip something that they thought was unimportant.

Re: Intuitive GUI? My arse.

"I'd go one step further than a big red button on the PC itself and go as far as to suggest a dedicated key on a keyboard to "restart current app" would be great"

I will take the opposing side there. The user can learn to close applications. They can learn that there is no restart Outlook button, and that when you say "restart" you mean exit and relaunch. Or you can say "exit and relaunch". Either is much simpler than adding a key to a keyboard that is completely unnecessary.

If we start adding more keys for every situation that comes up some time, the desk will have no space for a mouse or papers because it will all be a keyboard with buttons like restart app, clear print queue, actually check for network interface having a connection rather than have the user assume that their internet is down, enable remote access for the helpdesk to show you what to do, close remote access because a scammer told you to push that button, figure out what application the user wants to use to open that file instead of the OS default because they don't know how to change it, etc.

Re: Intuitive GUI? My arse.

The man -k command is a pretty rudimentary search facility which usually doesn't work. It can't handle any kind of complex query, and it falls back to hoping you can find a keyword that only matches the command you're looking for with it being quite easy to get nothing or twenty unrelated things instead. Here's an example. I've installed a command, pdftotext, which does what it sounds like. Imagine that someone new has sat down at my terminal, doesn't know that command exists, but wants to do that.

man -k "convert PDF to text"

convert PDF to text: nothing appropriate.

Well, maybe the quotation marks are not right here? They wrapped it because it is a command line argument with spaces in it, but when you do that in a search engine, it means to literally find the phrase. Let's try without.

$ man -k Convert PDF to text

30-systemd-environment-d-generator (8) - Load variables specified by environment.d

Git (3pm) - Perl interface to the Git version control system

RAND (7ssl) - the OpenSSL random generator

__fbufsize (3) - interfaces to stdio FILE structure

__flbf (3) - interfaces to stdio FILE structure

[over a thousand more results removed]

Fine, so maybe a phrase isn't the best search pattern. We don't want a thousand results. What if we use the smallest term possible:

$ man -k PDF

pdfattach (1) - Portable Document Format (PDF) document embedded file creator (version 3.03)

pdfdetach (1) - Portable Document Format (PDF) document embedded file extractor (version 3.03)

pdffonts (1) - Portable Document Format (PDF) font analyzer (version 3.03)

pdfimages (1) - Portable Document Format (PDF) image extractor (version 3.03)

pdfinfo (1) - Portable Document Format (PDF) document information extractor (version 3.03)

[8 more results]

This would seem to help a bit, after all I only have thirteen PDF handling tools installed here and they all came from the same package anyway. However, what happens if I run the same command on, for example, a system where I have a lot of LaTeX tools, which output to PDF. You get pages even with a terse search term like that. The search facility is very broad and the only way to restrict things involves regular expressions, which can work (man -k "PDF.*text.*convert") or can fail (man -k "PDF .* text .* convert"). Yes, the latter search pattern doesn't work because the description we're looking for has PDF in parentheses. This worked because I already knew the right command and could work back to a search pattern. For a user who doesn't know what the command's description has in it, they will have many more problems.

Re: Intuitive GUI? My arse.

My suggestion is to write a quick script in any language of choice that looks something like this:

for directory in path_environment_variable:

for filename in directory:

manpage = output_from_command(man filename)

synopsis = manpage.parse_line_starting_with("Synopsis")

print(filename +": " synopsis)

The problem comes when you have commands that don't structure their man pages like that, don't have a useful synopsis when they have one, or the many things in there that don't have man pages.

Writing code in the comment box doesn't work well. It was indented properly when I put it here.

Re: Intuitive GUI? My arse.

A CLI has a claim to being the least intuitive interface to the unfamiliar user, however. If you already know a command exists, you can probably look up a help page on it, but if you're asking a question like "What do I need to perform [action]", you don't have any way of finding that out quickly from the interface. You could look through every binary available on the path on the assumption that it's probably one of those, but those aren't organized in any way, have names that don't tell you anything about what they do, and not all of them will even have man pages. Invoking them to hope that they all have a --help argument and won't, for example, run some program that ignores the parameter is dangerous. Don't get me wrong, I spend most of my day in CLIs and I'm posting here while I wait for another CLI command to finish, but that works a lot better because I have had years of experience with it and I'm using a system where I already know the commands I'm using most often, know what they are capable of, and have DDG at my side in case I need someone to tell me the incantations related to something I haven't worked with before.

New York Times sues OpenAI, Microsoft over 'millions of articles' used to train ChatGPT

Re: It's all about profit

Or to limit the actions others may take with the work. For example, to have a restriction on how something can be distributed or used. I can require people using copyrighted code to release changes as open source, or someone using copyrighted text or artwork to only use it in noncommercial situations, and I have those rights because of copyright. It can also limit where the work can be displayed. For example, if I write something on my website and I want people who read it to look at other things on that site, whether because it could earn me money or not (it's not), I can restrict others' right to put it on their website instead. Those things are not necessarily about profit, though they often have an option to have a commercial benefit as well.

Study uncovers presence of CSAM in popular AI training dataset

Re: Problematic associations

They probably didn't, since they were referring to geographic differences. There is a lot more traffic going through countries like the UK, US, and Australia than there is in others. That traffic will represent the inhabitants of those countries more than those of others, regardless of the ethnic background of the users concerned. There is, for example, more likely to be data from people of African ancestry now living in those countries than those of African ancestry living in various countries in Africa where internet access is limited to a small subset of the population, even though the size of the latter group may be higher than the former. Similarly, the traffic generated on the African internet is likely to be biased towards countries like Nigeria and Kenya with a lot of internet infrastructure rather than countries like Chad or Eritrea which are quite lacking. This is a pattern that an AI trained on the internet will be repeating, along with many other patterns. Depending on what you want the model to do, these patterns, either that one or something else, may be desirable or undesirable, but ignoring them and expecting the AI to bypass them is a fool's errand.

"May as well just randomly scrape images off random sources."

Yes, that's the plan. Then, if you can be bothered, hire cheap labor to filter out some of the worst stuff. Then just train on the remaining mass. That's the models we have now. They contain stuff nobody wants in there, they contain illegal versions of works that the AI companies don't want to pay for, they contain complete gibberish, they contain personal information, and the AI companies are fine with it because they still look sort of authoritative when they make up something.

FTC bans Rite Aid from using AI facial recognition in stores for 5 years

Re: Skull measuring

"It is surprising how many people fail to recognise the official political philosophy of this party."

And one more of them, since what you've put in there is inaccurate. No, they did not plan or try to destroy capitalism. Having the word "socialist" in the name does not mean that, and you can look at speeches or actions to demonstrate otherwise.

Millions of Xfinity customers' info, hashed passwords feared stolen in cyberattack

Re: LOL! Just another day in the cloud

I'm not sure about the other times you said it, but you do know that A) this could happen equally well whether the server with the database is running on the cloud or in one of their buildings, B) this was initially caused by a bug in networking equipment, so if they were running a cloud-only network it wouldn't have happened (this time, but probably eventually something else would have done it) and C) this is a company with a lot of servers in their own buildings, so it probably wasn't running on the cloud? By attempting to blame the cloud for what was actually a system configuration problem, you risk proving that you don't actually understand what the cloud's security implications are.

Re: In a more direct way...

Do you want to explain how a company makes money by having customer details taken by people who will use it for fraud? If it's the obvious: the fraudsters pay the company for that information, would you care to cite a single time it's happened? The fraudsters don't want to pay for that when they can steal it. They can steal it because the company has been lax with their security because they won't suffer too much. Those things work together very well for those who sell stolen personal information. If you want to prove otherwise, you'll have to do more than suggest that I'm stupid; I might be, but you have done nothing to prove that your allegation so demonstrates.

Re: So my theory is...

In an indirect way, yes, because they spend less on securing things and have little or no consequences when a breech happens. Directly, no. They're not getting money from having a breech. I'm not sure if you meant that, but others have expressed that view before and it is ridiculous.

Musk floats idea of boat mod for Cybertruck

Re: "You know, can't all vehicles briefly act as a boat?"

I don't know, Tesla was pretty good at taking something that someone else invented, making a modification to it, and getting the credit while the actual inventor who achieved it first was relegated to a "who is that". Then again, although almost all of his accomplishments were derivative rather than revolutionary, he was making them through his own knowledge and effort, so Musk still doesn't get that far.

Re: 7,000 lbs?

I'm interested to see how many cars you can find in any category except 6. There may be a few things in 5 which can be shoved, but I'd like to see a category 2 car. Oh, and was there supposed to be a 4 there? You might need a new category system when user-carried is not part of the plans, replaced by user-carrying.

UK officials caught napping ahead of 2G and 3G doomsday

Re: G for Generation

I'm not sure that was ever claimed. 3G, for example, was not ever attempting to get 3 Gbps to any device, and it got well over that if we're looking at the network as a whole. The fact that when they started using those names they kept incrementing the number suggests that it was always meant to stand for generation.

Zuckerberg hunkers down in Hawaii to wait out apocalypse

Re: What these type of people dont realise.......

Dirty secrets are less important after some unspecified world-ending event. If you have the choice to get some supplies from the bunker because the guards are siding against the owner of them, would being told that the guard was fired for embezzlement really change your view on whether you would like to eat? If the secret was much more severe, you'd also have reason to question whether the allegation was even true (so you're telling me you didn't have a problem hiring a serial killer as your guard?) which wouldn't help either. I don't know what scenario we're really looking at here, but it clearly involves a significant amount of societal collapse. For example, you can no longer earn the support of people by paying them, presumably because money is now worthless, and in such a scenario, personal information is not so important either because everyone is focused on survival. How we get into that scenario is another question, but not one I think they have an answer to.

Re: Why not?

I have this feeling that if the event is sudden enough that you know you should go to your secret emergency location, it is probably also sudden enough that the people working the airports where your plane is decide they don't need to be working today, or if they do, that your flight is less important than the emergency transport going through that airport. Unless you have a private runway and a pilot who doesn't have a reason to want to go somewhere else, air transport is probably insufficient. Similarly for a ship, they usually take a lot of people to run properly and those probably are not sticking around when something bad happens so you can get going when you say now.

Re: Why not?

I do wonder how they expect to get from where they live to where their secret bunker is. Sure, sometimes they live in the above-ground part of that place, so if "the event" happens while they're there, then they might be fine. If they're in California when "the event" starts off, it's not going to do them much good, is it? This also depends heavily on what exactly the event is, but I'm not sure they've thought far enough to have any useful answer to that question. If I had infinite money and for some reason couldn't spend it on useful things, I'd consider something like this, but I'd have answers to the questions that caused me to do so.

Google Groups ditches links to Usenet, the OG social network

Re: Another brick in the wall

I think you're looking at this in the wrong way. Email won't be killed off. If it goes, and I think that's going to take a long time, it will die a natural death. Anyone is free to set up a mail server and use it, either using the traditional protocols or using some of the security overlays designed to make spam a little harder which are also available as open specifications. Some providers can make it less interoperable, but they can't take it from you entirely. If email dies, it will be because people have decided that they no longer want to communicate that way, not because someone else made it impossible. True, if the free mail providers discontinue their services, that will make more people abandon it, but those people could have found a different provider and continued to use the protocol. If you're afraid of decentralized things dying, look at the users, or more importantly the former users, rather than the competitors. The competitors can't kill them. They can only entice users to do it for them.

You don't get what you don't pay for, but nobody is paid enough to be abused

Re: Problematic pattern recognition

Any legal or policy reasons will apply equally well to recording in person or remotely. If you need consent from all parties to record where you are, that applies whether there's a computer in the middle or not. If the policy says you don't do it when we're talking about the dodgy stuff, it says that no matter how the company does it. If you're going to ignore those things and do it anyway, then it comes down to the ability to make the recording, which is in most cases easy. If you are not going to do it, then you are back to trying to get things in writing. Where you are located is unrelated to this.

Re: Problematic pattern recognition

Of course, but my point was that working remotely or in person makes no difference to that situation, since you can record with equal ease in those situations if you've decided to do so. It may not always be the best idea, but if you're worried about recording, remote working is no guarantee that it will come in writing. Our friend elsergiovolador is a major adherent of remote working, but I'm afraid it leads to attempts to praise it whether deserved or not.

Re: Problematic pattern recognition

It's completely unrelated to WFH. If they want to avoid a paper trail, they'll do a voice or video call if they can't have a physical meeting. Either way, there will be no recording unless you make one, and you can make one if it's in person just as easily as if it's online. It's not hard to turn on your phone's voice recorder before having a meeting in person if you decide you need to. I know that you like remote working, but ascribing benefits it doesn't actually have doesn't really help make that argument.

Science fiction writers imagine a future in which AI doesn’t abuse copyright – or their generosity

Re: Regulate the prompts?

"The problem with this scenario is not at all that the data was obtained illegally, but that you're committing a weird kind of slander by creating the impression that I'm the one doing or saying these things."

I covered this. That's a clear argument if I actually say that the video that appears to be from you actually is you. If I don't use your name, but it just looks like you, then I can respond that of course I've never claimed anything about you and therefore slander doesn't apply. It is still wrong and it should still be illegal based on not having permission to use the recordings of you without permission, and thus the illegal access to the training data is quite important.

"What, would you ban celebrity impersonators?"

As I pointed out, the impersonators and the software are not really doing the same thing, even though saying their intentions are the same is a superficially valid argument. I pointed out the comparison specifically to say that it was different in multiple ways from the programs that make faking speech and video easy. Did I fail to make that clear?

Re: Wonder

If we decide that modern AI is useful, then that's exactly the situation we're in. My view is that it is not sufficiently useful to justify exempting training anyway, and that even if it was, the companies that intend to make a lot of money from the result of the training can pay for their training data. We usually don't get exemptions for being useful. No matter how useful the code I write, I don't get my requirements for free. AI companies don't deserve an exception.

Re: Why worry so much?

And how will you find that place to publish? If people are attracted to a publisher because they have avoided the AI-generated books, then what stops a bunch of authors from sending their AI-generated books to that publisher for consideration because, if they manage to fool them, they'll benefit from the better reputation? Any publisher that is successful at gaining that reputation will have to deal with a large quantity of LLM-written books anyway.

I'm not sure there's anything we can do about this, but I'm certainly not willing to take the destruction of copyright with that as the end benefit.

Re: Science fiction or fantasy?

No, it's better. It's a future where rich people who want to pay so that everyone else gets great work for free just pop out of nowhere. Surely, if those people are there, they'll also be generous as well. If we're leaving reality, why not paint the best utopia we can come up with? The writer of that proposal certainly isn't restricted by reality.

Re: Short sighted sci-fi

Well, not a sci-fi writer exactly, but the short story "The Great Automatic Grammatizator" by Roald Dahl comes close. I won't post a link since I'm on copyright's side here and it is still found in libraries, but it's almost certainly online as well.

The companies who decide that having a work in their training data is better than not having it would pay the person who made that work for the right to include it. They would be free to ignore anything like that and only use free data, and if they decided that they want to use more content, they decide what content and then pay for the stuff they want, rather than trying to decide that everything is free if you happen to be an AI company, a rule that I'm pretty sure they would not agree with should I get a copy of their models.

Re: Regulate the prompts?

Consider something else I can do with AI. There are some pretty good programs that, with some effort and a lot of GPUs, can create models of a person's voice and appearance. Let's say that I can find some videos of you speaking online and I use this software to create a model of you. It looks like you. It sounds like you. It is under my control. I can make a virtual person of you say whatever I want. I could use this in an illegal way by actually using it to impersonate you, but let's say I don't. Still, whenever I want a video made, it would appear to be you saying and doing whatever I stated. Should I be allowed to do this if I can find the video online, even if you didn't intend it to be up there, even if you would have tried to have it taken down if you knew it was there?

That's basically what these models are doing. It sounds cleaner because it's text and because it's using data from more than one person, but it is still using data which was obtained illegally. I can pay you to get the right to make a virtual you, and several people have agreed to that, but doing it to someone who has not agreed should not be permitted. The human parallel is also there. Making a computer representation of you is similar to hiring a lookalike actor and a vocal mimic if we only look at the intention, but the realities of the two are very different, both in strength and scope.

Europe inches closer to insisting gig workers are treated as employees

Re: > No competition, indeed

If we're getting annoyed at common misinterpretations, I can complain about people who assume that average is supposed to mean the mean and complain when people use it otherwise. For example, when discussing IQ which is specifically built around the median, average probably does refer to the one that's intrinsic to the measurement, as useless as that measurement is. When not using such a measurement, the meaning of average is not clearly defined, but in informal speech it is more likely to refer to a combination of median and mode than to the arithmetic mean. That doesn't make their point any more useful or less cliched, but I'm afraid your rejoinder is pretty cliched as well.

"During a session, requests can carry a session identifier by annotating href and XHR request-URIs with query parameters, and forms with hidden form fields, simply by injecting those when serving content."

That is very easy to break. For example, if this is an online store and I've sent you a link to an item in the store, but you already added items to your order, then using my link will not retain your session identifier and you'll lose your progress. Weirder things might happen when people unwittingly include their session identifiers in links sent to others. Probably most of them will have expired by the time anyone uses it, but there's a chance that if Alice shares a link and Bob clicks it right away, Bob might find himself having more access to Alice's account than Alice had planned. Ways to work around this are of course possible, for example embedding an IP address in a session token and invalidating it if a different IP address is used, but that would risk having a leaked session token including personal information and would now mean that a user on a laptop or mobile device finds themselves losing stored progress when they move to a different network. There is a pretty good reason for not including session-specific data in URLs, and if you're not putting it there, then it has to go somewhere else.

Re: losing his job

I'm certainly not trying to be a contrarian. I think I'm failing to understand the distinctions that are so clear to you. For example, losing your contract and losing your job don't feel all that different to me. Using myself as an example, in either case, I am still a programmer, I still have the ability to go find someone else who wants to pay to get some code written, but I don't have the relationship that was bringing in income anymore. Therefore, if failing to show up early in the morning can get me fired or have my contract canceled, it feels like a similar level of power to me. The mechanics of how I do the job are different if I'm a contractor or an employee, but I will have similar requirements if I want to keep earning money from that company. If I'm a contractor and the company cancels it, what I am left with that an employee wouldn't have is the structures I set up to do the contract, for example the corporate entity through which I work, but that's not a lot since I, as an employee, could also just set one up if I decide to start being a contractor.

There is clearly a difference between contractors and employees and should be one. I'm not opposed to trying to classify gig workers as employees, although my opinion on it varies with what specifically they're doing. I just don't think the test of how much freedom they have works very well because it seems difficult to quantify that. If I'm working as a contractor for a massive corporation, someone trying to apply the rule probably won't think that I have an equal level of power with that company, but size doesn't necessarily equal the amount of power they have. Many of the other tests that are used are things that a computer could do given some defined inputs, or at least a human could do without having to require too much common sense, while the level-of-control test would appear to require a lot of subjective determinations.

For example, who has more freedom: a person who is tasked with going to a specific location at a specific time and decides in advance whether they'll take it or leave it, or one who gets individual jobs and accepts or rejects them at the time? Using those descriptions, I would say the former is the more restricted, but those are descriptions of a cabling contractor who has to install cable in an office when it's not in use and a rideshare driver who gets to choose the rides they do, and with those clarifications, the driver seems to be the more restricted. You could use a different test, such as whether there is an intermediary involved, but that still wouldn't change all of the ways the worker does control their situation, arguments that those businesses are sure to bring up when this test becomes part of a court case. That is why I suggest we find a clearer way to make that determination.

Re: test applied objectively

I'm not sure that comparison works. I can easily tell my boss that I don't care about his requests, at the cost of probably losing my job. A contractor has a similar situation. A contractor that's told that they have to attend a meeting with people to determine the requirements of the project or to provide feedback on an intermediate stage will not usually get to just refuse to do that and keep the contract, but does that mean that the contractor is not really a contractor in that situation? Similarly, the mechanic can look at the task I asked for and tell me to go away, and I can look at my boss's tasks for me and quit. Those end up looking very similar: I don't do the work, I don't get paid, I probably don't get any more work from that guy. So what test do you use to determine whether a choice to leave, which basically everyone has, is that of a contractor or an employee?

Re: Not difficult

"Funnily enough one of the other criteria is how much freedom the freelancer has in undertaking their work. If you act like a boss, you're an employer."

That was their point. Specifically, that a term like that is kind of vague. If you hire a contractor, you usually still have some requirements about what that contractor is going to do for you, including some pretty detailed descriptions of the job you intend that contractor to complete and quite often tasks they'll have to perform that place some restrictions on when they do it. For example, a contractor hired to write a system will receive a lot of information about what the system has to do, on what resources are available for the software, and a list of meetings they'll have to attend to get information which will partially set the work hours involved. Does this qualify or not? If it doesn't, then what does? It is not a test which can be applied objectively.

Google Pixel gets privacy mode to keep your selfies safe from prying repair techs

You can try telling the repair tech that, since it's only hardware, they can swap the screens without any login information, testing by seeing if the lock screen comes up correctly. I'm not sure if they'll accept that, but you can try it. Some other repairs might need them to test things that are only available when the device is running.

FCC really, truly won't give SpaceX nearly a billion bucks for Starlink rural broadband

Re: I don't want to be mean to lobbyists and PACS and lawyers but ....

"if you are going to offer subsidies for rural service why not just refund the rural users some part of their bill for whatever service provider they choose?"

Because that will just make whatever provider of any quality that currently exists put up the prices, collect the subsidies, and provide nothing more than they already did. These subsidies are only there for building something better, meaning that companies that want them will have to make some improvement, or at least pretend to well enough to fool the regulator.

Re: It is only Rural if they say it is.

I'm not sure why you're using Nigeria when speed data in various parts of the US, which is at issue, are available for a longer time. For example, you could use this article or the map it's referring to. One problem that might have is that as more users sign up, they will either have to send up more satellites or decrease speeds for everyone, probably both. As the linked article states, they're already having to do that in some parts of the country where there are more users. If subsidies are used to add more users to the service right now at the cost of decreasing the speeds for everyone and not necessarily having any long-term plans, then the FCC might prefer a different method.

Re: Internet in Brazil was very expensive, until competition.

"No-one in the UK would put up with over $100/mo for a sim only mobile service."

Do they have to? I see plenty of much cheaper mobile plans from US providers, including options from $10-$40 per month, with the higher levels including unlimited data though there's probably a catch to it somewhere. Also, we're not really talking about mobile service, but home internet. I don't expect that the FCC's rural proposals will be perfect, but they're intended to provide faster service where there is currently nothing, and if such a service could compete with Starlink, then customers would at least have a choice of two providers, and those providers might try to win customers from the other one in some way that would benefit them.

Think tank report labels NSO, Lazarus as 'cyber mercenaries'

I will try one more time. Governmental abuse is one problem. The availability of tools for that abuse is a different problem. You can have governmental abuse without those private companies if the governments themselves write it, and that would be, or rather is, bad as well. Since those private providers of intentionally abusive tools distribute them to more governments, it tends to enable more abuses because the cost of developing them is spread around. The governments concerned would have been willing to commit abuses anyway, but the tools make it easier for them to do so and to have stronger effects when they do.

If I was trying to "deflect", I wouldn't keep saying that governments are committing abuses, as I have in both preceding posts. Meanwhile, I have yet to see any real point from you, other than somehow deciding that NSO and its ilk are unimportant. They are contributors to the problem, and thus we discuss them.

Surprise! Email from personal.
[e-mail] is not going to contain good news

Re: BTC price

It wouldn't be very accurate, since it would at best track criminals cashing out and victims initiating payments, both of which come after the original crime. It also doesn't very much because people gambling on it causes a lot more noise, so fluctuations related to criminal activity are absorbed and somewhere between very hard and impossible to notice.

Re: but it's not always as easy as that to implement.

"You'd really like to trust your personal phone to Microsoft (for example), not to screw up an update? Really?"

You mean of the Microsoft Authenticator app? Yes, I'm completely fine with that. I've had that authenticator, along with several others, for years. The app has never broken, let alone caused damage to anything else on my phone, because can you name any app update that went through the normal procedures* that bricked a phone? At worst, the authenticator would break which is more my employer's problem than mine, but that's never happened with that one or any of the other ones I've used, and I've used at least four for various purposes.

* I'm referring to a normal app with user-level permissions, not for example one installed as part of a rooted device's firmware. I have broken, although not bricked, a device by messing with those, but a) that wasn't an update, it was me deliberately deleting and replacing files in the /system/priv-app directory which you can't expect to be perfect, b) authenticator apps are not in that directory and don't do anything like that, and C) they wouldn't have the ability to do that even if I was able to replace them with code intended to brick your phone because of app sandboxes for user-installed apps.

Re: "Many of these victims reported [MFA] was not enforced"

Which points out how it fails:

"IT shouldn't be able to mandate MFA without approval by the C level."

IT: We should have MFA.

Management: We're busy. Go away.

IT: There are major security risks. We can prevent them or at least significantly reduce them.

Management: Fine, go do it.

IT: We'll need some money and you will all need security tokens on you at all times.

Management: How about we skip this idea?

IT: That's dangerous, and it's not that much money.

Management: Bring it back up at the next planning meeting.

Admittedly, everywhere I've worked full time, IT has and it's gotten implemented, but you can see how it isn't quite as simple as we'd like it to be. I'm not comfortable letting IT off the hook when the management has pushed back, but neither can IT bear all the blame in that situation.