Posts by veti 4417 publicly visible posts • joined 25 Mar 2010
I had my first ever spam call from an AI, this week. It said its name was "Chrissy" and it was doing some kind of survey, but it was definitely a robot, so I didn't feel bad about hanging up on it.
Human trafficking is moderately risky and fairly expensive - at least, compared with robots. This is one problem that AI should be able to take care of, probably quicker than law enforcement could do it.
It's not about "helping solve a difficult problem". It's about not making unnecessary assumptions, or at least being aware when you are making them so you can think about their effects and test for them.
This, incidentally, is why I worry when I see and hear about "activist" workforces putting pressure on their employers to take sides on (this or that). Because that's a tell for lack of diversity in the workforce. The assumption that all "right-thinking" people see things the way you do is the root of about half the evil in our world.
They controlled for that. The study shows that it's more stressful to sit through the same presentation onscreen than in person.
Of course that's not the whole story. There's still a lot to be said for telepresence. But it's a part that hasn't been objectively measured before, which is interesting.
If you sell a book, you have no right to specify who can or can't read it, or how, or why. That's *not* one of the rights copyright gives you.
And I for one will fight any attempt to create such a right. It may happen, through some sort of backdoor licensing step as you suggest, but if it tries to come to my jurisdiction, I'm prepared to travel down to parliament and camp in the lobby to make them put a stop to it.
Well, one thing that's hard to understand is your jump from "the datasets contain copyrighted work" to "OpenAI is commercializing copyrighted material". And then you go on to claim that, apparently because of this, it's "violating copyright law", which is a whole other step that doesn't follow logically from the previous one.
If you sell copies you make of a book or other work (that you don't have the right to), it's copyright violation, sure. But OpenAI isn't doing that. It's selling access to a system that has (probably) been trained on this book, among many others. But unless it actually regurgitates significant chunks of text, it's not clear how that's a copyright violation.
Copyright law creates certain very specific, clearly defined "rights" around 1) reproduction, 2) adaptation, 3) publication, 4) performance and 5) display (the "five pillars of copyright"). To make a case against OpenAI, you'd have to demonstrate to a court how it's doing at least one of these things with your work.
The "emerging economies" point out, not unreasonably, that it's the developed countries that have produced nearly all the CO2 to date, and them talking about emissions in the developing world smacks more than a little of (a) pulling the ladder up after them and (b) rampant hypocrisy. If America and Europe and Asia, with all their wealth and infrastructure and knowhow, still can't bring themselves to cut carbon emissions, then what gives any of them the right to demand that Africans do it?
Yes, sure, providing clean energy to the developing world would be great. But unless you can suggest a way to provide it at the same (or lower) cost as coal-fired power stations and in unlimited quantities, you can't blame them for taking as much as they can get and also building more fossil-fuel-powered capacity to top it up.
That's why it's the west that needs to clean up its act. It has all the advantages, it needs to develop the technologies and methods to provide a decent lifestyle for everyone with net-zero, or something close to it, emissions. Once we've shown it's possible, then developing countries can adopt (or improve upon) those innovations. But until then, it's not reasonable to ask them to bear the burden of cleaning up our mess.
Yep, they're back.
It's sad. Back in 2016 this happened. This forum had usually been a reasonably balanced, sensible place, but suddenly you couldn't say anything disparaging about Trump without people jumping down your throat. People who would have been (were) completely silent if you'd said equally (or more) rude things about him a year or two earlier, mind you.
In 2020, for whatever reason, it didn't happen. Sure there were some Trumpists around, but not many, certainly not enough to compete with an overwhelmingly anti-Trumpist consensus.
But now, they're back. Climate denialism is one of the big fronts on this particular forum, so every story that touches on it will draw them out. And I assume it's only going to get worse for approximately the next 12 months. Oh well, we've had about seven years of near-normality; now hunker down, they'll mostly be gone by December 2025.
Aristotle was a great thinker and in some ways the most scientifically minded of the Greek philosophers. (In the sense that he believed the truth should be inferred by observing reality, rather than by making it up in your head like the Platonists.)
Your quarrel is with the medieval scholastics who took him as an authority rather than a foundation. The forerunners of modern Biblical literalists. In other words, people who missed the point so widely, they had to pretend they were aiming for a whole different target.
Downvoted for asking for specifics. Typical.
Look, we can sit around here moaning all we like about how iniquitous it is and should be stopped, but to stop it we would need to persuade a court, or a government, to do something. And for that we need specifics.
If you can say "$CAR downloads $DATA from $PHONE_APP", where $DATA is legally protected, then you've got an argument we can make to a court. Then we can take it to the next step, which is demanding accountability from car manufacturers and phone users. But as long as you're just wringing your hands saying "they could be taking anything, we don't know", you've got the square root of sweet fuck all.
(Note that the specific items of data that qualify for protection will vary by jurisdiction. A single phone number, for instance, probably doesn't qualify, but a list of them might. A number plus an accompanying name might. That's why we need to be very clear and specific about what data is being taken.)
Can you show how such data might leak?
Asking because phones in my experience have ways of classifying data and controlling what leaks how. Can you show that a car is likely to have access to (data that would be considered illegal to leak, however that's defined in your jurisdiction)?
Austria prefers not to join NATO, that's their decision to make and nobody (in NATO, at least) is going to force them or threaten them over that. The security landscape of Europe looked very different in the 1980s to today.
Czechoslovakia and Yugoslavia both broke up - one amicably, the other violently - on their own terms. Sure there was outside interference (particularly in Yugoslavia, which became another proxy war between Europe and Russia), but nobody invaded or tried to conquer either one of them. Their governments simply came to the conclusion that their old structures weren't working and needed to change. That's something independent countries can do. See also "Union, Soviet".
Brexit proved that the UK was independent even when it was a member of the EU. And it's still a member of NATO, so according to your own logic Brexit made no difference to "independence" anyway.
This is the second time in as many days I've seen somebody using common English words in a very, very eccentric way on this site. I can only conclude the Russian trolls are back in force, and they're putting trainees on us.
To "deliberately" turn a war into a quagmire, you'd have to be controlling both sides.
Now, perhaps you think the Russians are deliberately playing along in order to allow their enemies to learn about their cyber warfare tactics, possibly so that they can develop their tactics as they go... But otherwise, you'd have to believe that the Ukrainians, with their barely-coordinated American and European supporters, can fine-control the pressure they put on their enemy to ensure it's always just enough to keep the conflict going, without the risk of ending it.
I don't think anyone has that much control.
Netanyahu fostered Hamas to be the most visible face of the Palestinian cause, because it lets him off the hook from having to deal with them. You can't be expected to negotiate with someone who says up front that they want nothing more than to kill you.
Netanyahu can't afford to negotiate with any Palestinians, because it would destabilise his political alliances. But he's under constant Western pressure to deal with the PLO and Fatah. Fostering Hamas was a way of deflecting that.
Hamas has built very little international support. That (for whatever it's worth) goes to Fatah and the PLO.
And that is one thing Hamas hopes to reverse with this war. They'll make the Palestinian cause so toxic in the west, that the PLO's support will dry up completely, leaving Hamas as the undisputed champions of all Palestinians.
Because sure, Hamas hates the west and it hates the Israelis, but what it hates most of all is Fatah.
I cant say I have seen anything from Trump to suggest he would do the same (I may be wrong). I suggest he would be more likely to try and dismantle this huge violation of America's rights.
Considering he's the one who signed the present iteration into law (the FISA Amendments Reauthorization Act of 2017), I think you're being unreasonably generous there.
Looking down this thread, it looks like the trolls are back in force. Starting to feel like summer 2016 all over again.
This is what you get for trying to violate the rights of foreigners. The authors of the 14th amendment were wise, when they said that the same legal rights apply to every person "within its jurisdiction".
Then lawmakers make these laws on the basis that they're meant to be used against foreigners. Quietly ignoring the fact that it would be blatantly unconstitutional to discriminate between subjects on the basis of nationality, even if there were some technically feasible means of doing it, which there isn't.
Always remember, when lawmakers say they want to do something to foreigners, what they're really demanding is the right to do it to anyone. This includes you.
I can see Microsoft's plan having some effect. At least campaigns (but no-one else, apparently, for some reason) will be able to show when the materials they produced have been tampered with. That's - not much, but it's something.
Meta's proposal - is meant to achieve what, exactly? It only affects paid ads, which are only a small part of the problem in the first place, and it won't even affect them very much. Can't see how it would even stop the sort of A/B testing that the IRA pioneered in 2016. All it needs is someone willing to lie.
I think I have the right to "override their T&C's" because (1) I never agreed to them, and (2) they themselves are freeloading on the open standards and structure of the internet, which is the same technology that I'm "exploiting" to block their ads.
They want me to stop freeloading on their service? - fine, they can build their own delivery system using their own technology and set their own rules. Until then, they can fuck right off.
Their demographic research tells them that 50-something bald men very often have adolescent daughters who routinely use their parents' PCs for whatever, and who are horrifically insecure about their appearance and particularly their hair. And far be it from them to allay such insecurity, when there's so much better money to be made by stoking it.
1. The only legal standard that makes any kind of sense is, "if it would be illegal for a human to do it, it's also illegal for a computer system to do it, otherwise it's fine".
2. Nothing in copyright law allows rightsholders to place any restrictions on who or what can study and learn from their work. As long as they don't (copy, perform, distribute, adapt or translate) it, they're golden.
I'm sorry for the artists, but they're on a losing track here.
Every new Windows system launches and shuts down quickly. Then they gradually slow down as cruft gets added through updates.
When 8.1 was new, I was delighted that it would give me a usable desktop in less than 20 seconds from a cold start. It made a welcome change from Vista, which would often take 20 minutes to get to that stage (no exaggeration, I actually timed it more than once). But by the time it finally obsoleted (only ten months ago, in fact), it was up to over a minute. I'm sure Win 11 won't be so snappy in five years' time, either.
Funny, but that was pretty much my experience of trying to "upgrade" from Windows 8.1 to 10. It did actually work, eventually, but the poor machine was never the same again, and eventually gave up the ghost about six months later.
8.1 was a perfectly good system. 10 gave us back the start menu (which was nice), but introduced many new annoyances (no, Microsoft, I do not want your idea of news or weather in my taskbar or start menu, that's not what those things are for, and I certainly don't want to search from there). But the Internet hivemind had decided "8 bad, 10 good" - without even trying to distinguish between 8 and 8.1, of course - so here we are now, everyone convinced that 10 is the system of their dreams and it'll have to be pried from their cold dead keyboards.
10 was successful because it followed a failure (8). 11 is a failure because it follows a success. It's all about perception, nothing more. There is no real reason to stick to 10, it's just the devil we know.
OK, let's take those one by one.
Lockdowns were a terrible idea.
First of all, "a terrible idea" is a value judgment. That is to say, it says more about your values and priorities than anything objectively measurable in the outside world. So it's hard to call it "deluded".
Having said that, I would say that the delusion here is in believing you can make a simple blanket statement like that - well, at all. Practically every "lockdown" was different, implemented in a different environment with different rules and circumstances. Some of them worked, some didn't.
New Zealand's first lockdown, for one, was undeniably a brilliant idea. Thanks to it, we had 15 months of almost-normal life and almost-full-strength economic activity (minus tourism, which is an important but still only small part of the whole economy) while the rest of the world was going through wave after wave of shutdowns and deaths by the million. Other lockdowns in other places, and a later lockdown in NZ, certainly failed in varying degrees and for various reasons, but that doesn't detract from the breathtaking success of the first one.
Masks don't work to significantly reduce the spread of respiratory viruses.
"Significantly" is a weasel word, of course. Most people who have studied the subject say that masks do make a difference in some situations. The statistical evidence is not clear, but anyone who claims on that basis to have definitive knowledge on the negative side is - yes, I think "deluded" is a reasonable word there.
Covid probably came from an accidental lab leak.
Possibly. Possibly not. Again, use of the word "probably" suggests an unwarrantable level of confidence in your conclusion. Delusion.
Covid vaccines are nowhere near as safe or effective as we were led to believe.
I don't know what you were led to believe, but I'm not aware of any major revisions to the information I saw about them.
Net Zero is a terrible idea. Even if the whole of europe ceased to exist tomorrow, our emissions would get swallowed up by China.
Again with the "terrible idea" thing. It depends what you think the point is, or how you think reality is likely to unfold in either scenario. I suggest that for "the whole of Europe" to achieve "net zero" would involve pioneering technologies, social and economic changes that would quickly be adopted by other countries, including China.
Religion is stupid (OK that one goes back well beyond 2020)
Again with the value judgment. What do you think "religion" is trying to do? I'm not sure how you can go about answering that, but when you have answered it, then you can rationally assess its actions and measure them against its success criteria and defend the statement that "it's stupid". But I suggest not here, because no-one is going to read a 10,000 word essay on the topic, which is about the minimum it would take.
Of course it would entail attributing volition and intelligence to an abstract idea. Not sure what you'd call that... But you've already done that much with your statement, so clearly you've made your peace with that bit.
Yeah, it's hard to see the amount as a deterrent.
But the degree of trust required to give this information to His Muskiness... as we say in New Zealand, "Yeah, nah".
I kinda get why the Philippines might be a reasonable place to trial this, but why NZ? I didn't think we were a hotbed for spammers or bots. Is it just because the market is so insignificant, if it dries up completely that's not much of a loss?
Tools, yes. Hamas is chiefly funded by Iran, plus some of the Gulf states, none of whom give a shit how many Palestinians die.
I'm sure they do want to provoke an equally brutal response from Israel, and I'm reasonably sure they'll get it. Bibi's government is childishly easy to manipulate that way. But what goals they have in mind, is a bit more speculative.
They're diverting some attention (and aid) from Ukraine, to Mr Putin's glee. They're creating division in the west. They're bolstering Mr Trump's chances of winning another term, with everything that would imply. Perhaps most important for their long term strategy, they are politicising support for Palestine in a way it hasn't been for a generation now. If they can make the Palestinian cause sufficiently toxic in the west, they will bolster their own position and claim to be the only real representatives of that cause. Then, PLO RIP.
You're almost there, but you've missed one crucial link in the reasoning.
"People keep trying to mash various shaped pegs into the round hole that is Excel" - this much is true, and well put. What you're missing, however, is that this keeps happening, despite the ready availability of any number of otherly-shaped holes they could be using.
It follows that any solution of the form "we should stop doing this" qualifies as "well, duh" advice. Sure we should, much as we should stop eating meat, but we won't. So the question "what can we do to prevent/control/mitigate the resulting damage?" is a good one. I don't think the article comes up with any very good answers, mind, but I can respect the attempt.
Scenario: you receive a report of a hack against a system where your software may have been deployed, and it sounds like it may have been exploiting a vulnerability in your software. You're not sure about either of those points, but circumstances are suspicious.
What exactly is your responsibility at that point? Are you "aware of an exploit" yet? Can you afford to simply ignore the report and wait for someone else to do the hard work investigating?
OK, so it's three days later, and someone publishes a preliminary investigation into the hack. No one has actually pinpointed your component yet, but the circumstances are piling up. You're now about 80% sure it was present on the system, and if it was vulnerable that would explain what happened.
Are you "aware" yet?
And whenever you are "aware", what on earth makes you think just sending an email is going to cut any ice? Name an official notification system anywhere in the world that relies on email. No, there will be a form, requiring you to specify the program, the version, the component and other details that will require serious expert investigation to uncover. So whenever you are "aware", you've got a ticking clock to discover enough about it to fill in this form, whatever it's going to look like.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2023
