Well... yes... that's kind of what explicit consent means!
In our case, it's a signed agreement. Before we can send an email.
And about once a month, someone "opts out" by mistake by clicking a link in an email (or via some webmail's Junk function which then asks if you want to visit the unsubscribe link) and I get stroppy enquiries about why they're not getting emails and weren't told they'd unsubscribed.
Why? Because I *can't* email you to tell you that you'd unsubscribed... you'd unsubscribed! And if you didn't notice that you did so, with the confirmation screen, then there's nothing I can reasonably do.
Ironically, in the last few months I have received several dozen pieces of unsolicited commercial email to an unadvertised address, without consent, concerning "How to get ready for GDPR", GDPR conferences, GDPR auditors, and even people claiming to help me form my own GDPR policies.
I find it absolutely hilarious - who on Earth is going to touch the GDPR companies that can't even follow the rules themselves?
That said, it's just a return to common sense. Did I ask you to email me? Specifically YOU? No? Then why are you emailing me?
GDPR lets me give the same response as I would to someone knocking on my door. Do I know you? Do you have legitimate business that required you to wake me up?
No? Then fuck off, and never darken my door again.
Dealing with from the IT end has also been enlightening. We hired a member of staff just to get us through GDPR. They went through all my systems and processes. Pretty much, it doesn't affect us.
Explicit consent before sending email? Check. People able to stop such email on demand? Check. People able to request the data that we have on them? Check. Data being held only as long as necessary? Check.
Because most of this stuff was just obviously what the Data Protection Act required anyway. And being a good business.
All the changes that have happened are to do with things like paper records (nothing to do with IT), etc. and databases that are outside IT control (e.g. our alumni list was hand-managed on paper, they've since digitised it because GDPR doesn't distinguish how you store it, so there's no longer any advantage to avoiding the DPA because you're not storing it on computer), and formalising policies that were already in place.
Actual IT changes necessitated? None. I've updated a bunch of software which now have GDPR deletion/anonymisation features (but we won't use those for a long time because pretty much we only store what's necessary and stuff which we need to keep anyway) and things like "obtaining and recording explicit consent" features.
GDPR = DPA + case law. If you've been keeping up over the years, GDPR is no shock. If you haven't.... well, you've been at risk for quite a while whether you think so or not. It only needed one stroppy customer to take you to court to expose practices that judges have been saying you MUST do (to be classed as "reasonably protecting the data" even under the previous DPA) but that just weren't codified in an actual law.
About the biggest pain in GDPR? Gathering all the GDPR compliance statements from everyone else we deal with. (Hey, Apple! Are you done yet?!).
Either Amazon, or Whole Foods, are losing money on this. Maybe it's market-capture. Maybe it's just bringing the profit margin ever-lower.
But if you want to compete you can only do two things: Do the same, or ignore it and let them suffer losses that you're not willing to suffer.
If you're only paying for Prime what you save every time you go into Whole Foods, they are making a loss of 11 times that, even if you only go once a month.
In the same way that Prime Video (despite "giving away" movies to all their Prime subscribers) isn't knocking Netflix out of the market, a deal like this won't knock everyone else out of the market. Netflix still has three times as many users as Amazon Prime Video does. And they are paying SPECIFICALLY for Netflix, not getting it bundled as a freebie on something they already buy for other reasons.
Amazon is certainly a big player and scary competition to have, but they have tried to penetrate all kinds of markets and failed in many of them. Same with Spotify vs iTunes vs Prime Music.
Hell, I used Amazon's "Fresh" delivery service precisely once and wouldn't touch it again. In the UK, it's basically one of the cheapest supermarkets (Morrison's) and the service is pathetic compared to any other supermarket's online service.
Personally, I'd keep doing business. Hell, maybe even approach other providers and try to do a similar deal. But I wouldn't panic about anything other than staying competitive and providing a better service.
Nintendo made their first loss in 30 years at one point.
And they've been going for 129 years.
There's no suggestion that they would have lost money consistently for their first X years of business at all. In fact, when they started to look like they were losing money, they changed tack and moved on.
Apple - made $118m in its second year, and the first year wasn't a loss.
Shell - I can't find enough documented with just a quick search by they certainly aren't an Amazon-like "let's fritter money away in huge chunks now to own the market in 20 years time" kind of business. Hell, they use to import sea-shells. Again, when it's not profitable, they moved to other things.
Facebook - granted. But there's nothing to suggest that Facebook or any of those companies are worth what they say. WhatsApp was priced at something ridiculous and they just bought it up and have done nothing with it.
Sorry, but I can't see that it would be a huge loss to demand "Hey, companies, you need to make money as stated in the primary aim of running a company" (in the UK at least, last time I formed one, they have to be run with the aim to produce profit).
So you already have bikes which are at the top acceleration they can be, and at a speed that all the others are at.
So... erm... what's the point, exactly? It's surely then NOT all about the speed again.
It's a bit like Formula One - everything could be very much faster but you're just not allowed to do that and each year the restrictions get tighter, so you're into people winning by 0.001 of a second (i.e. small enough that a light breeze could make all the difference).
This is the bit that, to me, makes such things very boring indeed. Now, if you said "you can only spend so much" or "you only get a litre of fuel", then it gets interesting again... people can innovate.
But if an electric bike can go no faster than a petrol bike, because of safety and convention, and all the enthusiasts buy that kind of bike, then what does it matter what bike you have at all?
I could sort of it see working, if say all the tickets available were for the quiet-periods, unsold seats, etc., if they were buying up the seats en-masse for a cheap price (saving the theatre having to advertise them, etc. as they're already "sold" as far as they are concerned), or something but... MoviePass don't seem to be doing that.
The Wiki article literally says that they load the cost of the movie onto a pre-paid debit card, which you use to buy the movie. So you pay $X a month. And they give you $Y each day. Unless X > 30Y, I can't see how they ever could make money. And when it is, nobody would bother with it.
But just... giving away a movie ticket every day for the price of a movie ticket once a month isn't that appealing a business model, especially once you stick a middleman into it, apps, pre-paid card numbers, etc.
At what point were they going to morph that business model into something that actually makes profit?
"Moviepass - the subscription cinema ticket service -"
It really wouldn't be that hard to include that in the summary.
P.S. most of the time when I have to Google something, I end up going off Slashdot to read about it instead. Especially if it's some US-specific thing.
If you want to provide content, rather than just parrot articles from elsewhere, days later, with not even any moderation, editing, clarification, or summary, then Slashdot really needs to pull its finger out.
"MoviePass has been burning through $21.7 million per month."
"MoviePass for the three months... lost $107 million, earning just over $1 million from marketing deals and $47 million from subscriptions."
It doesn't take a genius to work out that 48/3 = 16 and thus you're earning $5.7m less every month than you're actually taking in.
Your business model is literally worse than "Let's burn $5.6m in cash each month on a big bonfire".
We honestly need - as a planet - to put in laws that stop this shite. I know the shareholders are the main ones being burned, but if after a year of operation you can't show overall profit (or at the very least, contracts GUARANTEEING that the profit will be repaid and progress towards fulfilling those contracts), you should just be shut down. And then again each year. Hell, you shouldn't even be allowed to get to things like IPOs etc. at that point.
Yes, I have owned a business. It wasn't large, I don't claim that. But I never made a loss, not once.
I know that with shares, it's a case of "who's left holding the hot potato" and up until that point people can profit from nothing more than hype, but we need a way of stopping people just taking investments, burning through them, without a single profitable year.
Yes, that would mean no Amazon (or at least, an Amazon starting up in a very different way). But it would also mean that there wouldn't be a thousand Amazon wannabes all doing the same and losing people's money, because it's not JUST the shareholders who lose out. Everyone from employees to suppliers to the taxpayer loses out from such things.
Electric cars (and bikes) just confirm what I always knew.
Nobody's actually interested in the speed, or the acceleration. What they want to do is make a noise and be loud and "sound" cool.
Seriously, Harley Davidson could make an electric motorbike that out-accelerates all their other models (they are actually doing that). All the bike enthusiasts I talk to laugh at that idea - they don't even mention "range" or "battery life" (I think they have a hard time saying "battery" to be honest, if it's not full of some environment-destroying fuel, they can't play Mad Max). They just think the idea of something near-silent is counter to why they buy the bike.
They don't really admit that, but that's all it's ever been about. Not "Hey, I have the faster car". Not "I love the speed". Nope...it's how I can get everyone's attention and who's looking at me?
The cars are the same. Line up a dozen sports cars and nobody will look at the electric model. Even though it will out-accelerate the $200,000 supercar (and let's be honest, any race where you're just at top-speed all the time is boring... an electric car would win in a drag-race, in a rally, on a track, etc.).
Look at the motor-sports and electric cars don't really figure. Even the "electric formula one" kind of things get zero attention. But hey, put regenerative braking into something and use that to boost the performance, that's okay because it makes a lot of noise still.
Racing is literally about "who can be noisiest, messiest, cause the most disruption, and nearly trash their car" not "who wins".
For years, consumer cars have gotten faster and faster, but nobody really notices or thinks it matters, because they've also become quieter and quieter. Everyone drives what would have been a Formula One car back in the sixties, but now those kinds of cars are "granny cars", because they don't make lots of noise.
Sorry, but all those "car enthusiasts" that I know spend more time polishing and waxing, and bolting on ridiculous addons to their car than they ever do tweaking performance. Hey, unless you get a modchip that makes the car noisier and smokier...
A child missing warning has almost NOTHING to do with public safety on the scale you're talking about.
That child's safety, yes. But I don't want a text every time he crosses the road not at a crossing, walks along the top of a wall.
It's an ENTIRELY different thing to flood alerts, tornado alerts, etc.
We just don't have missing child alerts like that in my country. They are on police-force websites, missing-child sites, people copy/paste them to Facebook if they're local, but unless it's something incredibly drastic then they are certainly not forced down anyone's throat. There has not been a time when *everyone* was informed indiscriminately about a missing child in such a manner, even with a few that made the news.
Waking up an entire city at 3am because of a reported-missing child is a ridiculous solution and abuse of the service - and just as the article states, it makes people ignore ALL the alerts.
So, yes, if someone starts pressing the panic button and forcing through junk alerts to my phone where *my* life isn't in danger if I don't receive them, then there's going to be a problem. And, note, this only applies if I can't turn them off. If I don't consent to getting such alerts, then I don't consent. Let me burn to death in the forest fire, drown in the flood or be taken away by the tornado.
But if you class "little Johnny 200 miles away has been hiding in the garden and we can't find him" in the same manner, then I'm going to turn them off and charge you the cost of waking me up for absolute nonsense.
I'm more surprised that you're all familiar with such alerts.
The day I recieve one of these, I will not only disable it (or ditch the phone for telephony entirely and go full IP and mute everything else), I will file a formal complaint and initiate a lawsuit if they just say 'tough'.
My country surely has the same facility but unless it's literally 'London is radioactive, stay the fuck away' it shouldn't be used. And people say *we're* a police state. Honestly, you could trigger outright riots and vigilanteism with that kind of crap.
The transaction signing algorithm can be swapped out in an ordinary update. That's been done before and will be done again.
But Bitcoining mining relies on proof-of-work, using hashing-to-create-a-hash-of-zeroes, and it's pretty fundamental. I'm sure that other proofs-of-work are allowed but they would need pretty drastic changes to the way that all Bitcoin miners operate or are optimised, whereas transaction signing wouldn't as they are much rarer calculations only necessary for verification of the ledger, not billions-upon-billions of proof-of-work calculations.
Hashes are actually one of the best ways to stay QC-safe.
At the moment, we use our existing encryption algorithms to generate hashes. Instead, most of the quantum-safe encryption algorithms use hashes to build themselves.
The reason is quite simple if I can use an analogy. It's not 100% accurate, but good enough to make most people understand.
First - a hash. You take an input, you generate a "mini-mash" of it - you jumble it up and cut bits out in a predictable manner until you get something that is absolutely tiny but built from that original input.
The same input will give the same hash every time, because you do the same thing every time. Yet millions of different inputs might give you that same mini-mash (because they are much fewer hashes than there are data-sets, so by chance they overlap sometimes - a hash collision) but that hardly matters in real life because the chances of those other inputs being valid Microsoft Word files, or containing the same secrets as your files are infinitesimally small.
Quantum-computers attacking conventional encryption works like this: - you "build a circuit" that performs the same encryption that was used (e.g. AES, ECC, etc.). - you plug in the ANSWER (the encrypted text) into the end of it. - by some magic of physics, it instantaneously determines the only possible inputs that could have ever formed that answer. Thus, it works out the SECRET INPUT (i.e. the keys) that was originally used to encrypt it - all in one "tick".
As such, QC defeats traditional encryption entirely. Every encrypted text/web session is one tick away from compromise with zero effort required and only tiny amounts of time expended.
But when you apply that technique to hashes, there may not be only one possible input. In fact there may be an infinity of inputs that give the same hash (because the input can be any size, right? So the mini-mash of a entire novel could the same as the mini-mash of "123" or the same as the mini-mash of a dataset as large as the universe).
As such, the QC can't determine the answer - it gets all the answers and doesn't know which one's right. To know which one was right, you'd have to check them all... and you're now back from "working out the answer instantaneously" to "checking all the possible combinations one at a time".
So instead you can build QC-safe encryption by using hashes upon hashes upon hashes upon hashes. Now any possible inputs a quantum computer may determine is lost in an infinity of other inputs... and it's no longer as simple as "just give us the only input that looks like a Word file" - you have to check them all.
As such, hashes are the basis of much more security, based on their "unknown but potentially infinite amount of data" turned into "a small set of characters" property.
Crypto-currencies use hashes a lot (Bitcoin is/was basically built upon "keep hashing different things on the end of this string until you get a hash of 0 out of it") and so may be the last thing to fall to QC.
In the same way that QC turns cryptanalysis on its head, to solve the problem of QC we turn hashes and encryption on their heads.
So all your web browsers and disk encryption programs have got a quantum-safe algorithm in them already, then, and you're using it, right? So that your data is safe for the changeover they're talking about.
I think you'll find this is IBM warning that they - as a company trying to build quantum computers at the moment - see them coming in the next five years, which means we should have moved 5 years ago.
It's a warning that is going unheeded.
No. Elliptic curves are not quantum-safe.
What we have already, you can take and massively increase the key size but that doesn't make the TIME spent any less. It just makes the QC that cracks them "larger" and thus harder to build. Increasing AES / etc. keysize will give us a couple of years past someone making a viable QC. After that we have... what? Nothing in place, certainly nothing commodity, certainly nothing that an ordinary user can use.
It's think it's quite well-known that things just aren't this simple.
1) The common cold is really several hundred different types of the same thing. You can't cure them all, and if you try, another few will pop up. 2) Though the summary mention asthmatics, etc., the common cold is pretty much not worth treating. Go away, take some tissue, it'll be gone in a day or so. Thus treating it is really a waste of time. 3) Almost all cold remedies are "over the counter", not "prescription". It's just not worth treating unless it gets more serious. This means people are WILLING TO PAY and the things they are paying for are pretty much USELESS. They relieve symptoms, not cure it. And that's if you're lucky. Many are no better than placebo. 4) Because of this, they are a huge source of income - cheap and easy to make and regulate, people buy them all the time and use them for every single cold, and do so VOLUNTARILY. No doctor is prescribing them, nor an insurance company covering their purchase (at least in a sensible country). 5) By "curing" the common cold, you are then taking away a LOT of easy money for the pharmaceutical industry, which people choose to pay VOLUNTARILY, for no real good reason, every time they have a cold, which helps - but also hurts - nobody. When that revenue stream dries up, guess what? They'll move to something else or up their prices on other things.
Paying for cold remedies funds your cancer drugs, in part. But you don't have to pay that. You choose to. For placebos and the equivalent of some honey in hot water. It's a stupid thing to do, the remedies literally do nothing worth doing, and they are perpetually present in the population - by the time you are symptomatic, you've already got it from and given it to everyone else.
In that kind of environment, a cure is a nonsense. You'll just force something that's almost completely benign to mutate, and it'll do so very quickly given the number of strains, the transmission rates and the tiny amount of contact required to catch them. And you couldn't cure them all.
Nobody of normal health loses even a day of work to a cold over their lifetime. (Flu, etc., are VERY different beasts).
I speak as someone who DOES NOT medicate with anything non-prescription. I don't take headache tablets, indigestion tablets, cold remedies or anything else. I have literally bought one of each to keep in my bathroom cabinet for guests... they will likely be thrown out unopened when people comment that they look like they are out of the Ark.
Those things cure themselves, or they have simple remedies that don't even need anything beyond what's already in your kitchen.
But people willingly paying for those placebos are paying profits to a pharmaceutical company that would make up for them elsewhere. On things you DON'T want to make people pay extra for. Like asthma inhalers, and insulin testing kits, and so on. Things that are very common, quite cheap to make and not very dangerous so don't require a ton of testing.
Curing the common cold will just give you worse colds that are harder to cure, and more expensive medication. That's pretty much why none of the big pharmaceutical firms have bothered to do it.
"AI" that we have is no-fucking-where near actual intelligence at all. They are large statistical systems, often blurring expert systems, human-fed tuned heuristics, statistical analysis and genetic algorithms in one huge mass of junk.
Notice how AI peaks early, and then plateaus forever. It's easy to train it how to tell an image has a banana in it, but then further training - even to billions of images - doesn't improve it much. And retraining or further refining its training (e.g. find bananas AND apples) leads to utter failure.
Because it completely loses such "inference". The way we build them and teach them makes them form one of two things - human-fed rules and follow them unthinkingly, or arbitrary rules that we can't inspect.
Is it thinking it's a banana because it's yellow? Because it's bendy? Because the image is mostly yellow? Because the center pixels of the image are yellow? Because of the presence of a crescent shape?
And thus we only ever end up fixing it in one way: Telling it what ISN'T a banana enough that those rules are forced to evolve. That's the point at which is falls down and can't "untrain" an assumption that its initial generations were formed entirely upon, and which every subsequent training has reinforced a billion times.
We don't have AI. It doesn't learn. It doesn't infer. It cannot determine the patterns or rules for itself. It just blindly and statistically forms arbitrary "superstitions" about what it's been told which sometimes can be convincing for a short period on simple examples.
If you feed a pigeon, in a sealed box, at random times it will form such superstitions about the pattern of feeding. If it was scratching its feathers when the last random feeding happens, it'll start to scratch its feathers when it was food. If, by chance, that random feeding happens again, you'll see the pigeon scratch its feathers whenever its hungry and get confused when it's then not fed.
AI is not only doing this, it's even dumber than this. The pigeon will eventually unlearn and form a new superstition, quite quickly. Eventually the pigeon will give up trying to predict the system having established that it's random. AI can do neither.
Everything we refer to as AI currently isn't. Nothing is actually thinking for itself. Inferring. Reasoning. Detecting patterns. It's not even a short-term, hand-fed pigeon.
Now, how we go about making machines infer is another matter entirely, but our current approach is drastically wrong and we've been convinced by our own superstitions. Watch someone demonstrating Siri etc. to their friends for the first time.
Privately, they've tried "What is the weather tomorrow" and that pretty much works every time. So they demonstrate that. Then they believe that it's actually understanding those words so they either try themselves, or lead others, to improvise around it with a freeform query. And before long it all falls apart. They don't notice that they've TRAINED THEMSELVES to talk to Siri, not that Siri has learned to understand them. But they convince themselves that it's somehow intelligent.
I honestly believe that if we want to stand a good chance of AI happening, we have to stop small, short-run trainings of something in a limited scope and just go for an artificial life-form. Rather than train a box for a year to understand text and then give up, restrict the box to X amount of nodes, etc. and constantly try to feed it towards the answers we want it to give, we have to look at our own and other species children.
We need long-term, large-scale projects that just sit and listen to the world. We don't interrogate them for the weather, but we watch for signs of patterns. Does attention get focused on a piece of new data that's unusual? Does it start to discard data of its own accord? It takes a baby several years of intensive, personal, focused training after MILLIONS of years of evolution to get close to becoming an actual human (leave
Digital radio surely encompasses DAB, Radio over IP, satellite radio, and all kinds of things.
The true distinguishing feature here is "is it just a waveform over the airwaves?". Digital technology allows error detection and correction, compression, efficient use of spectrum, etc.
To be honest, it's only a matter of time before AM/FM disappear for anything but amateur use, but the problem I have is this: we shut off analogue TV broadcasts a long time ago now (2012). The line given to us was that those frequencies would then be used for things like WiFi, 3G, digital TV, etc. thus freeing up a lot of space that we could use for more stuff in narrower channels.
I'm not entirely sure that's happened at all. There are token services running in that band but nothing of any real significance.
I'm therefore not convinced that EVEN LOWER frequencies (thus inherently limited in their maximum bandwidth) are going to be at all worth cleaning up.
Make one app. That has messenger in it. With a bunch of options of what I want it to do (run all the time for messenger, read my photos, etc.).
Try and not make it an app that literally sucks up all my storage just browsing (my gf filled her phone up twice to the brim, when we looked it was all data stored in the Facebook app - removed the app, reinstalled, all was fine again)
Then, maybe, just maybe, I'll consider installing it. But JUST that. Nothing else. No other apps to do the same thing. And, no, you really don't require (or will ever get) one percent of the permissions your current apps demand.
To be honest, the fact that you DELIBERATELY break the Facebook mobile website to remove messenger (when "View as Desktop Site" shows it perfectly well but in a not-nice format) pisses me off more than anything. You are literally trying to force me to use the apps and I have no interest in that.
You know what happens when you try to force people to use products/services they have no interest in? They go elsewhere.
Another 5 years and Facebook will be like MySpace is now.
But there is no way on Earth that everyone who claims they can hear the difference I stated actually does. It's a standard "Oh, yeah, I can see the 8K from here" bollocks line.
There are people who can remember every single detail of every day of their lives. There are people who can recognise a face in a crowd of millions given only a five second glance at a photo. These things are all possible.
But 99.9% of people do not exhibit such skills, whether they claim to or not. And MP3 vs raw audio is one of those things that's been double-blind tested a thousand times over and people can't tell.
I pity you if you can. It means your music will NEVER sound right. It's as simple as that. But similarly, why should I have to enter into the argument every time it's mentioned when I - and the other 99.9% of the population - can't tell the difference.
But nobody has ever been able to prove that being MORE active than normal is somehow magically BETTER than just being active.
Same for everything - not having enough vitamins is bad (malnutrition). But eating more of them than a normal person requires doesn't turn you into Superman.
And it works in reverse. Eating too many fatty foods is bad. But it doesn't mean that cutting out fatty foods entirely is any better than just eating sensibly.
Pressure has nothing to do with atoms bouncing, per se. In fact, at high pressure, in solids, etc. they can't do that at all. They exert a force against other atoms.
In a gas, sure, the "atmospheric pressure" is the result of atoms moving around each other, but even there - gravity makes the pressure on the bottom of the container greater than the lid. You might not care, it might only be a small difference, but it's still contributing to the pressure on the floor.
If you have a force exerted on an area, you have a pressure.
You just write them a polite letter that says "go away" and then tend to stop bothering you until someone else moves into that house again. Or you buy a TV (the shops have to dob you in by law).
Slashdot Mirror
But they did apply.
If you're handling EU data, they've always applied. To get that data you SHOULD have had to sign the same kind of guarantees / waivers as anyone else.
Ignorance of the law is no excuse, and if you're collecting data on EU citizens, of course it's liable to EU law just the same.
If GDPR doesn't affect you now, the various DPAs never did.
If it does affect you, the various DPAs always did.
Well... yes... that's kind of what explicit consent means!
In our case, it's a signed agreement. Before we can send an email.
And about once a month, someone "opts out" by mistake by clicking a link in an email (or via some webmail's Junk function which then asks if you want to visit the unsubscribe link) and I get stroppy enquiries about why they're not getting emails and weren't told they'd unsubscribed.
Why? Because I *can't* email you to tell you that you'd unsubscribed... you'd unsubscribed! And if you didn't notice that you did so, with the confirmation screen, then there's nothing I can reasonably do.
Ironically, in the last few months I have received several dozen pieces of unsolicited commercial email to an unadvertised address, without consent, concerning "How to get ready for GDPR", GDPR conferences, GDPR auditors, and even people claiming to help me form my own GDPR policies.
I find it absolutely hilarious - who on Earth is going to touch the GDPR companies that can't even follow the rules themselves?
That said, it's just a return to common sense. Did I ask you to email me? Specifically YOU? No? Then why are you emailing me?
GDPR lets me give the same response as I would to someone knocking on my door. Do I know you? Do you have legitimate business that required you to wake me up?
No? Then fuck off, and never darken my door again.
Dealing with from the IT end has also been enlightening. We hired a member of staff just to get us through GDPR. They went through all my systems and processes. Pretty much, it doesn't affect us.
Explicit consent before sending email? Check.
People able to stop such email on demand? Check.
People able to request the data that we have on them? Check.
Data being held only as long as necessary? Check.
Because most of this stuff was just obviously what the Data Protection Act required anyway. And being a good business.
All the changes that have happened are to do with things like paper records (nothing to do with IT), etc. and databases that are outside IT control (e.g. our alumni list was hand-managed on paper, they've since digitised it because GDPR doesn't distinguish how you store it, so there's no longer any advantage to avoiding the DPA because you're not storing it on computer), and formalising policies that were already in place.
Actual IT changes necessitated? None. I've updated a bunch of software which now have GDPR deletion/anonymisation features (but we won't use those for a long time because pretty much we only store what's necessary and stuff which we need to keep anyway) and things like "obtaining and recording explicit consent" features.
GDPR = DPA + case law. If you've been keeping up over the years, GDPR is no shock. If you haven't.... well, you've been at risk for quite a while whether you think so or not. It only needed one stroppy customer to take you to court to expose practices that judges have been saying you MUST do (to be classed as "reasonably protecting the data" even under the previous DPA) but that just weren't codified in an actual law.
About the biggest pain in GDPR? Gathering all the GDPR compliance statements from everyone else we deal with. (Hey, Apple! Are you done yet?!).
Business.
Either Amazon, or Whole Foods, are losing money on this. Maybe it's market-capture. Maybe it's just bringing the profit margin ever-lower.
But if you want to compete you can only do two things: Do the same, or ignore it and let them suffer losses that you're not willing to suffer.
If you're only paying for Prime what you save every time you go into Whole Foods, they are making a loss of 11 times that, even if you only go once a month.
In the same way that Prime Video (despite "giving away" movies to all their Prime subscribers) isn't knocking Netflix out of the market, a deal like this won't knock everyone else out of the market. Netflix still has three times as many users as Amazon Prime Video does. And they are paying SPECIFICALLY for Netflix, not getting it bundled as a freebie on something they already buy for other reasons.
Amazon is certainly a big player and scary competition to have, but they have tried to penetrate all kinds of markets and failed in many of them. Same with Spotify vs iTunes vs Prime Music.
Hell, I used Amazon's "Fresh" delivery service precisely once and wouldn't touch it again. In the UK, it's basically one of the cheapest supermarkets (Morrison's) and the service is pathetic compared to any other supermarket's online service.
Personally, I'd keep doing business. Hell, maybe even approach other providers and try to do a similar deal. But I wouldn't panic about anything other than staying competitive and providing a better service.
Nintendo made their first loss in 30 years at one point.
And they've been going for 129 years.
There's no suggestion that they would have lost money consistently for their first X years of business at all. In fact, when they started to look like they were losing money, they changed tack and moved on.
Apple - made $118m in its second year, and the first year wasn't a loss.
Shell - I can't find enough documented with just a quick search by they certainly aren't an Amazon-like "let's fritter money away in huge chunks now to own the market in 20 years time" kind of business. Hell, they use to import sea-shells. Again, when it's not profitable, they moved to other things.
Facebook - granted. But there's nothing to suggest that Facebook or any of those companies are worth what they say. WhatsApp was priced at something ridiculous and they just bought it up and have done nothing with it.
Sorry, but I can't see that it would be a huge loss to demand "Hey, companies, you need to make money as stated in the primary aim of running a company" (in the UK at least, last time I formed one, they have to be run with the aim to produce profit).
So you already have bikes which are at the top acceleration they can be, and at a speed that all the others are at.
So... erm... what's the point, exactly? It's surely then NOT all about the speed again.
It's a bit like Formula One - everything could be very much faster but you're just not allowed to do that and each year the restrictions get tighter, so you're into people winning by 0.001 of a second (i.e. small enough that a light breeze could make all the difference).
This is the bit that, to me, makes such things very boring indeed. Now, if you said "you can only spend so much" or "you only get a litre of fuel", then it gets interesting again... people can innovate.
But if an electric bike can go no faster than a petrol bike, because of safety and convention, and all the enthusiasts buy that kind of bike, then what does it matter what bike you have at all?
I could sort of it see working, if say all the tickets available were for the quiet-periods, unsold seats, etc., if they were buying up the seats en-masse for a cheap price (saving the theatre having to advertise them, etc. as they're already "sold" as far as they are concerned), or something but... MoviePass don't seem to be doing that.
The Wiki article literally says that they load the cost of the movie onto a pre-paid debit card, which you use to buy the movie. So you pay $X a month. And they give you $Y each day. Unless X > 30Y, I can't see how they ever could make money. And when it is, nobody would bother with it.
But just... giving away a movie ticket every day for the price of a movie ticket once a month isn't that appealing a business model, especially once you stick a middleman into it, apps, pre-paid card numbers, etc.
At what point were they going to morph that business model into something that actually makes profit?
"Moviepass - the subscription cinema ticket service -"
It really wouldn't be that hard to include that in the summary.
P.S. most of the time when I have to Google something, I end up going off Slashdot to read about it instead. Especially if it's some US-specific thing.
If you want to provide content, rather than just parrot articles from elsewhere, days later, with not even any moderation, editing, clarification, or summary, then Slashdot really needs to pull its finger out.
Burn $5.6m and the rest of the money in the market just became more valuable.
"MoviePass has been burning through $21.7 million per month."
"MoviePass for the three months... lost $107 million, earning just over $1 million from marketing deals and $47 million from subscriptions."
It doesn't take a genius to work out that 48/3 = 16 and thus you're earning $5.7m less every month than you're actually taking in.
Your business model is literally worse than "Let's burn $5.6m in cash each month on a big bonfire".
We honestly need - as a planet - to put in laws that stop this shite. I know the shareholders are the main ones being burned, but if after a year of operation you can't show overall profit (or at the very least, contracts GUARANTEEING that the profit will be repaid and progress towards fulfilling those contracts), you should just be shut down. And then again each year. Hell, you shouldn't even be allowed to get to things like IPOs etc. at that point.
Yes, I have owned a business. It wasn't large, I don't claim that. But I never made a loss, not once.
I know that with shares, it's a case of "who's left holding the hot potato" and up until that point people can profit from nothing more than hype, but we need a way of stopping people just taking investments, burning through them, without a single profitable year.
Yes, that would mean no Amazon (or at least, an Amazon starting up in a very different way). But it would also mean that there wouldn't be a thousand Amazon wannabes all doing the same and losing people's money, because it's not JUST the shareholders who lose out. Everyone from employees to suppliers to the taxpayer loses out from such things.
Electric cars (and bikes) just confirm what I always knew.
Nobody's actually interested in the speed, or the acceleration. What they want to do is make a noise and be loud and "sound" cool.
Seriously, Harley Davidson could make an electric motorbike that out-accelerates all their other models (they are actually doing that). All the bike enthusiasts I talk to laugh at that idea - they don't even mention "range" or "battery life" (I think they have a hard time saying "battery" to be honest, if it's not full of some environment-destroying fuel, they can't play Mad Max). They just think the idea of something near-silent is counter to why they buy the bike.
They don't really admit that, but that's all it's ever been about. Not "Hey, I have the faster car". Not "I love the speed". Nope.. .it's how I can get everyone's attention and who's looking at me?
The cars are the same. Line up a dozen sports cars and nobody will look at the electric model. Even though it will out-accelerate the $200,000 supercar (and let's be honest, any race where you're just at top-speed all the time is boring... an electric car would win in a drag-race, in a rally, on a track, etc.).
Look at the motor-sports and electric cars don't really figure. Even the "electric formula one" kind of things get zero attention. But hey, put regenerative braking into something and use that to boost the performance, that's okay because it makes a lot of noise still.
Racing is literally about "who can be noisiest, messiest, cause the most disruption, and nearly trash their car" not "who wins".
For years, consumer cars have gotten faster and faster, but nobody really notices or thinks it matters, because they've also become quieter and quieter. Everyone drives what would have been a Formula One car back in the sixties, but now those kinds of cars are "granny cars", because they don't make lots of noise.
Sorry, but all those "car enthusiasts" that I know spend more time polishing and waxing, and bolting on ridiculous addons to their car than they ever do tweaking performance. Hey, unless you get a modchip that makes the car noisier and smokier...
A child missing warning has almost NOTHING to do with public safety on the scale you're talking about.
That child's safety, yes. But I don't want a text every time he crosses the road not at a crossing, walks along the top of a wall.
It's an ENTIRELY different thing to flood alerts, tornado alerts, etc.
We just don't have missing child alerts like that in my country. They are on police-force websites, missing-child sites, people copy/paste them to Facebook if they're local, but unless it's something incredibly drastic then they are certainly not forced down anyone's throat. There has not been a time when *everyone* was informed indiscriminately about a missing child in such a manner, even with a few that made the news.
Waking up an entire city at 3am because of a reported-missing child is a ridiculous solution and abuse of the service - and just as the article states, it makes people ignore ALL the alerts.
So, yes, if someone starts pressing the panic button and forcing through junk alerts to my phone where *my* life isn't in danger if I don't receive them, then there's going to be a problem. And, note, this only applies if I can't turn them off. If I don't consent to getting such alerts, then I don't consent. Let me burn to death in the forest fire, drown in the flood or be taken away by the tornado.
But if you class "little Johnny 200 miles away has been hiding in the garden and we can't find him" in the same manner, then I'm going to turn them off and charge you the cost of waking me up for absolute nonsense.
I'm more surprised that you're all familiar with such alerts.
The day I recieve one of these, I will not only disable it (or ditch the phone for telephony entirely and go full IP and mute everything else), I will file a formal complaint and initiate a lawsuit if they just say 'tough'.
My country surely has the same facility but unless it's literally 'London is radioactive, stay the fuck away' it shouldn't be used. And people say *we're* a police state. Honestly, you could trigger outright riots and vigilanteism with that kind of crap.
The transaction signing algorithm can be swapped out in an ordinary update. That's been done before and will be done again.
But Bitcoining mining relies on proof-of-work, using hashing-to-create-a-hash-of-zeroes, and it's pretty fundamental. I'm sure that other proofs-of-work are allowed but they would need pretty drastic changes to the way that all Bitcoin miners operate or are optimised, whereas transaction signing wouldn't as they are much rarer calculations only necessary for verification of the ledger, not billions-upon-billions of proof-of-work calculations.
Hashes are actually one of the best ways to stay QC-safe.
At the moment, we use our existing encryption algorithms to generate hashes. Instead, most of the quantum-safe encryption algorithms use hashes to build themselves.
The reason is quite simple if I can use an analogy. It's not 100% accurate, but good enough to make most people understand.
First - a hash.
You take an input, you generate a "mini-mash" of it - you jumble it up and cut bits out in a predictable manner until you get something that is absolutely tiny but built from that original input.
The same input will give the same hash every time, because you do the same thing every time. Yet millions of different inputs might give you that same mini-mash (because they are much fewer hashes than there are data-sets, so by chance they overlap sometimes - a hash collision) but that hardly matters in real life because the chances of those other inputs being valid Microsoft Word files, or containing the same secrets as your files are infinitesimally small.
Quantum-computers attacking conventional encryption works like this:
- you "build a circuit" that performs the same encryption that was used (e.g. AES, ECC, etc.).
- you plug in the ANSWER (the encrypted text) into the end of it.
- by some magic of physics, it instantaneously determines the only possible inputs that could have ever formed that answer. Thus, it works out the SECRET INPUT (i.e. the keys) that was originally used to encrypt it - all in one "tick".
As such, QC defeats traditional encryption entirely. Every encrypted text/web session is one tick away from compromise with zero effort required and only tiny amounts of time expended.
But when you apply that technique to hashes, there may not be only one possible input. In fact there may be an infinity of inputs that give the same hash (because the input can be any size, right? So the mini-mash of a entire novel could the same as the mini-mash of "123" or the same as the mini-mash of a dataset as large as the universe).
As such, the QC can't determine the answer - it gets all the answers and doesn't know which one's right. To know which one was right, you'd have to check them all... and you're now back from "working out the answer instantaneously" to "checking all the possible combinations one at a time".
So instead you can build QC-safe encryption by using hashes upon hashes upon hashes upon hashes. Now any possible inputs a quantum computer may determine is lost in an infinity of other inputs... and it's no longer as simple as "just give us the only input that looks like a Word file" - you have to check them all.
As such, hashes are the basis of much more security, based on their "unknown but potentially infinite amount of data" turned into "a small set of characters" property.
Crypto-currencies use hashes a lot (Bitcoin is/was basically built upon "keep hashing different things on the end of this string until you get a hash of 0 out of it") and so may be the last thing to fall to QC.
In the same way that QC turns cryptanalysis on its head, to solve the problem of QC we turn hashes and encryption on their heads.
Cool.
So all your web browsers and disk encryption programs have got a quantum-safe algorithm in them already, then, and you're using it, right? So that your data is safe for the changeover they're talking about.
I think you'll find this is IBM warning that they - as a company trying to build quantum computers at the moment - see them coming in the next five years, which means we should have moved 5 years ago.
It's a warning that is going unheeded.
No. Elliptic curves are not quantum-safe.
What we have already, you can take and massively increase the key size but that doesn't make the TIME spent any less. It just makes the QC that cracks them "larger" and thus harder to build. Increasing AES / etc. keysize will give us a couple of years past someone making a viable QC. After that we have... what? Nothing in place, certainly nothing commodity, certainly nothing that an ordinary user can use.
It's think it's quite well-known that things just aren't this simple.
1) The common cold is really several hundred different types of the same thing. You can't cure them all, and if you try, another few will pop up.
2) Though the summary mention asthmatics, etc., the common cold is pretty much not worth treating. Go away, take some tissue, it'll be gone in a day or so. Thus treating it is really a waste of time.
3) Almost all cold remedies are "over the counter", not "prescription". It's just not worth treating unless it gets more serious. This means people are WILLING TO PAY and the things they are paying for are pretty much USELESS. They relieve symptoms, not cure it. And that's if you're lucky. Many are no better than placebo.
4) Because of this, they are a huge source of income - cheap and easy to make and regulate, people buy them all the time and use them for every single cold, and do so VOLUNTARILY. No doctor is prescribing them, nor an insurance company covering their purchase (at least in a sensible country).
5) By "curing" the common cold, you are then taking away a LOT of easy money for the pharmaceutical industry, which people choose to pay VOLUNTARILY, for no real good reason, every time they have a cold, which helps - but also hurts - nobody. When that revenue stream dries up, guess what? They'll move to something else or up their prices on other things.
Paying for cold remedies funds your cancer drugs, in part. But you don't have to pay that. You choose to. For placebos and the equivalent of some honey in hot water. It's a stupid thing to do, the remedies literally do nothing worth doing, and they are perpetually present in the population - by the time you are symptomatic, you've already got it from and given it to everyone else.
In that kind of environment, a cure is a nonsense. You'll just force something that's almost completely benign to mutate, and it'll do so very quickly given the number of strains, the transmission rates and the tiny amount of contact required to catch them. And you couldn't cure them all.
Nobody of normal health loses even a day of work to a cold over their lifetime. (Flu, etc., are VERY different beasts).
I speak as someone who DOES NOT medicate with anything non-prescription. I don't take headache tablets, indigestion tablets, cold remedies or anything else. I have literally bought one of each to keep in my bathroom cabinet for guests... they will likely be thrown out unopened when people comment that they look like they are out of the Ark.
Those things cure themselves, or they have simple remedies that don't even need anything beyond what's already in your kitchen.
But people willingly paying for those placebos are paying profits to a pharmaceutical company that would make up for them elsewhere. On things you DON'T want to make people pay extra for. Like asthma inhalers, and insulin testing kits, and so on. Things that are very common, quite cheap to make and not very dangerous so don't require a ton of testing.
Curing the common cold will just give you worse colds that are harder to cure, and more expensive medication. That's pretty much why none of the big pharmaceutical firms have bothered to do it.
Correct. It can't.
"AI" that we have is no-fucking-where near actual intelligence at all. They are large statistical systems, often blurring expert systems, human-fed tuned heuristics, statistical analysis and genetic algorithms in one huge mass of junk.
Notice how AI peaks early, and then plateaus forever. It's easy to train it how to tell an image has a banana in it, but then further training - even to billions of images - doesn't improve it much. And retraining or further refining its training (e.g. find bananas AND apples) leads to utter failure.
Because it completely loses such "inference". The way we build them and teach them makes them form one of two things - human-fed rules and follow them unthinkingly, or arbitrary rules that we can't inspect.
Is it thinking it's a banana because it's yellow? Because it's bendy? Because the image is mostly yellow? Because the center pixels of the image are yellow? Because of the presence of a crescent shape?
And thus we only ever end up fixing it in one way: Telling it what ISN'T a banana enough that those rules are forced to evolve. That's the point at which is falls down and can't "untrain" an assumption that its initial generations were formed entirely upon, and which every subsequent training has reinforced a billion times.
We don't have AI. It doesn't learn. It doesn't infer. It cannot determine the patterns or rules for itself. It just blindly and statistically forms arbitrary "superstitions" about what it's been told which sometimes can be convincing for a short period on simple examples.
If you feed a pigeon, in a sealed box, at random times it will form such superstitions about the pattern of feeding. If it was scratching its feathers when the last random feeding happens, it'll start to scratch its feathers when it was food. If, by chance, that random feeding happens again, you'll see the pigeon scratch its feathers whenever its hungry and get confused when it's then not fed.
AI is not only doing this, it's even dumber than this. The pigeon will eventually unlearn and form a new superstition, quite quickly. Eventually the pigeon will give up trying to predict the system having established that it's random. AI can do neither.
Everything we refer to as AI currently isn't. Nothing is actually thinking for itself. Inferring. Reasoning. Detecting patterns. It's not even a short-term, hand-fed pigeon.
Now, how we go about making machines infer is another matter entirely, but our current approach is drastically wrong and we've been convinced by our own superstitions. Watch someone demonstrating Siri etc. to their friends for the first time.
Privately, they've tried "What is the weather tomorrow" and that pretty much works every time. So they demonstrate that. Then they believe that it's actually understanding those words so they either try themselves, or lead others, to improvise around it with a freeform query. And before long it all falls apart. They don't notice that they've TRAINED THEMSELVES to talk to Siri, not that Siri has learned to understand them. But they convince themselves that it's somehow intelligent.
I honestly believe that if we want to stand a good chance of AI happening, we have to stop small, short-run trainings of something in a limited scope and just go for an artificial life-form. Rather than train a box for a year to understand text and then give up, restrict the box to X amount of nodes, etc. and constantly try to feed it towards the answers we want it to give, we have to look at our own and other species children.
We need long-term, large-scale projects that just sit and listen to the world. We don't interrogate them for the weather, but we watch for signs of patterns. Does attention get focused on a piece of new data that's unusual? Does it start to discard data of its own accord? It takes a baby several years of intensive, personal, focused training after MILLIONS of years of evolution to get close to becoming an actual human (leave
You think it's analogue instead?
Digital radio surely encompasses DAB, Radio over IP, satellite radio, and all kinds of things.
The true distinguishing feature here is "is it just a waveform over the airwaves?". Digital technology allows error detection and correction, compression, efficient use of spectrum, etc.
To be honest, it's only a matter of time before AM/FM disappear for anything but amateur use, but the problem I have is this: we shut off analogue TV broadcasts a long time ago now (2012). The line given to us was that those frequencies would then be used for things like WiFi, 3G, digital TV, etc. thus freeing up a lot of space that we could use for more stuff in narrower channels.
I'm not entirely sure that's happened at all. There are token services running in that band but nothing of any real significance.
I'm therefore not convinced that EVEN LOWER frequencies (thus inherently limited in their maximum bandwidth) are going to be at all worth cleaning up.
Hey Facebook.
Make one app. That has messenger in it. With a bunch of options of what I want it to do (run all the time for messenger, read my photos, etc.).
Try and not make it an app that literally sucks up all my storage just browsing (my gf filled her phone up twice to the brim, when we looked it was all data stored in the Facebook app - removed the app, reinstalled, all was fine again)
Then, maybe, just maybe, I'll consider installing it. But JUST that. Nothing else. No other apps to do the same thing. And, no, you really don't require (or will ever get) one percent of the permissions your current apps demand.
To be honest, the fact that you DELIBERATELY break the Facebook mobile website to remove messenger (when "View as Desktop Site" shows it perfectly well but in a not-nice format) pisses me off more than anything. You are literally trying to force me to use the apps and I have no interest in that.
You know what happens when you try to force people to use products/services they have no interest in? They go elsewhere.
Another 5 years and Facebook will be like MySpace is now.
It is.
But there is no way on Earth that everyone who claims they can hear the difference I stated actually does. It's a standard "Oh, yeah, I can see the 8K from here" bollocks line.
There are people who can remember every single detail of every day of their lives. There are people who can recognise a face in a crowd of millions given only a five second glance at a photo. These things are all possible.
But 99.9% of people do not exhibit such skills, whether they claim to or not. And MP3 vs raw audio is one of those things that's been double-blind tested a thousand times over and people can't tell.
I pity you if you can. It means your music will NEVER sound right. It's as simple as that. But similarly, why should I have to enter into the argument every time it's mentioned when I - and the other 99.9% of the population - can't tell the difference.
And then I get people insist they can tell the difference between high-encoding-bitrate MP3s and raw audio.
Not being active is bad for your health.
But nobody has ever been able to prove that being MORE active than normal is somehow magically BETTER than just being active.
Same for everything - not having enough vitamins is bad (malnutrition). But eating more of them than a normal person requires doesn't turn you into Superman.
And it works in reverse. Eating too many fatty foods is bad. But it doesn't mean that cutting out fatty foods entirely is any better than just eating sensibly.
Moderation in everything.
Any force you like.
Pressure has nothing to do with atoms bouncing, per se. In fact, at high pressure, in solids, etc. they can't do that at all. They exert a force against other atoms.
In a gas, sure, the "atmospheric pressure" is the result of atoms moving around each other, but even there - gravity makes the pressure on the bottom of the container greater than the lid. You might not care, it might only be a small difference, but it's still contributing to the pressure on the floor.
If you have a force exerted on an area, you have a pressure.
In actuality... nope.
And I didn't have a TV for about 5 years.
You just write them a polite letter that says "go away" and then tend to stop bothering you until someone else moves into that house again. Or you buy a TV (the shops have to dob you in by law).