On a busy highway, absolutely - there are any number of instances of people being injured or killed while attempting to get to the shoulder, or even after having safely done so, because they were struck by another vehicle. Expecting an autonomous vehicle to handle that safely in any situation is clearly a non-starter with the current state of play, but hopefully there is at least *some* code to try and automatically handle emergency situations in use by now, e.g. if an emergency vehicle is detected approaching from the rear on a clear stretch of road, to at least slow down and pull over as far as possible towards the shoulder. My interest was where on the spectrum from simply trying to revert to human control through to successfully handling the necessary maneouvers current automomous vehicles might be, although I do suspect it is indeed much closer to the former.
Accordining to TFS there was (as is usually the case with most autonomous vehicles being tested) a human driver present who was able to intervene and bring the car to a halt; The company claims the human test driver did everything right but is now responsible for the citation, so presumably sucks to be him/her. Having made that statement I would hope that Cruise will now cover any costs arising from the almost inevitable legal challenge to the citation though.
That said, I am curious what the current state of the art is for self-driving vehicles with regards to an emergency services vehicle approaching from the rear, or in the direction of travel. While it's possible that the code might detect an emergency vehicle and take specific action like slowing down or pulling over to allow it to pass, I'd be very surprised if we're at the point where it's capable of detecting that it is actually being requested to halt for a ticket. Also, what kind of accuracy there is for discriminating between actual emergency vehicles and things like maintenance vehicles that just have regular hazard lights but might not warrant any additional actions.
American cloud providers lose access to EU markets since we cannot provide the privacy protections they require
Do they, or do we need to wait and see how it works out in practice - most probably with the Microsoft case with the Dublin data centre? I've not seen any detailed breakdowns of the act yet, but it seems that in order to get access to data US agencies still have to ask for it via the government of the nation where the data centre is located to approve the request. There's two ways that can go; "Sure, here you go!", and "Nope, sorry, that data is protected by our privacy laws". That means it's going to come down to whether or not the approving body/individual is more concerned about breaching EU privacy laws, avoiding a potential diplomatic issue, or just keeping their US colleagues happy, and until such time as there's some clear EU legal precedent I don't think it's possible to say which it's going to be. It also might vary from country to country as well, since the EU's laws are not entirely homogenised as many seem to think.
Similar era here; cycles counted and the innermost of a deep nest of loops was usually a good place to start as even one less processor cycle could be removed you could often improve things considerably when you multiplied it all out, although I don't think that's it since there are a few competing DNS engines, which all hopefully pretty well optimised in their core code already at this point (feature bloat aside), and presumably Cloudflare is only going to be running one of them. Realistically, I'm only seeing two options here - Cloudflare coming to some kind of "arrangement" with APNIC for some memorable IP addresses for DNS (one of the few times you do need to memorise them, so that counts) to compete with Google, Quad9, etc. in the alternative DNS service provider space as a new revenue stream, hence people getting upset about the sudden repurposing of what they saw as IPs reserved for reseach use. The second, and I suspect more likely scenario given Cloudflare's DDoS protection services, is that they were fully aware they were going to be getting tens of Gb/s of junk traffic and wanted to do what could be some genuinely useful work on protected DNS services from DDoS attacks in a live environment, since the traffic patterns are likely "good enough". That does still make sense to be done in the IPv4 space since that's where the IRL garbage traffic is and, as you say, any lessons learnt should hopefully translate over to the IPv6 stack easily enough.
There's also a lot coming from captive portals that use 1.1.1.1 as a login/logout gateway IP, including some turnkey solutions provided by the likes of Cisco that are heavily deployed in providing free WiFi services to things like the hospitality trade. Yeah, they could (and should!) have used RFC1918 IPs as the default configuration for this, just like your home router tends to default to 192.168.1.1, but for whatever reason decided to default to 1.1.1.1 instead. Since that (fairly obviously) is highly unlikely to conflict with anything already on the network, guess what got deployed in the live environment?
The IPs are assigned to APNIC, a RIR, and they are free to assign them to whoever they want that meets their assignment policies, including entitities that are not Headquartered in the APNIC region. There is some debate in high-level networking groups like NANOG about whether those procedures were correctly followed, but that ultimately hinges on whether this is a joint APNIC-Cloudflare research project or a permanent assignment. The former is arguably within APNIC's currently agreed scope for the IPs in question, the latter may have circumvented a few procedures or opportunities for debate.
Ultimately though, the last time these IPs were routed - a partnership between APNIC and Google, they got 50Gb/s of garbage, mostly from things that were designed to use unassigned IPs rather than suitable RFC1918 IPs. There are not exactly very many companies that have the necessary infrastructure to filter out 50Gb/s of crap and still provide a useable service with what remains so, research or not, I can't see many people wanting these IPs anyway and if Cloudflare can make some use of them, good on them. Besides all that, there is also the question of why are people still doing "research" on IPv4 space; wouldn't it be better to be focussing on the brave new world of IPv6 - where's my 2001:2001:: resolver, or some such equivalent?
I was replying to someone who specifically mentioned Facebook, and also mentioned Facebook specifically, in a thread discussing Facebook so I think the context should be a pretty big clue as to who "they" meant. Even so, sure you can easily apply that to any of the companies that you can identify who track you and store your data, which probably covers most of the largest data hoarders out of the thousands, so it is a start at least and covers the worst offenders. A SAR can also include a request for information on any third parties your data has been shared with, and that's going to include information a LOT of those thousands of other companies that are more below the radar - and remember the only cost to an individual to create and issue a SAR is some time, so expect a lot of EU privacy activists to be firing off a lot of SARs until they get bored with it.
For those companies that are not on the ball with this, it's potentially death by a thousand admin-cost cuts dealing with the SARs, death by EU fines for demonstrable non-compliance for failing to comply with SARs, or hoping to $deity they can either continue to fly below the radar indefinitely, or at least avoid paying any penalties by virtue of having no EU presence. Quite how the EU courts expect to deal with that last scenario, or more shady operators that have faked company info for that matter, I have no idea (although "unsucessfully" does seem like the most probable answer) but it's going to be interesting finding out, and probably quite expensive for the unlucky ones since it's pretty obvious that this is also a good way for the EU to plug some budget deficits if it can get some major prosecutions under its belt.
Or, more interestingly, what kinds and quantity of data they have on people who have never had an account to start with, but have been tagged by family/friends/colleagues that do. Really curious to see what kinds of data people in that situation who hit Facebook up with an SAR once the GDPR comes into force are going to get back, although they may need to create an account and make a few posts to it in order to link their "offline profile" data to their personal ID.
I suspect you're not the only one planning on doing that. You might find this template useful for maximising your return, because the more points they have to fsck up the better, right?
It's not a "bit" overblown; it's pure hyperbole/clickbait. The monitors are fine, and that includes with MacOS, the issue is that the monitors will no longer work when they are connected through a third party peripheral with drivers that do some encapsulation to send display data (amongst other things) over USB.
That's because, like the OP, you're trying to cram multiple metrics into a single value - whether it's CPUs, GPUs, shoes, or whatever else. Clearly that's not going to work. Perhaps something akin to the "Performance Charts" that many industries have adopted might work though - those colourful charts that typically give a series of ratings for a number of key metrics, e.g. in the case of car tyres stuff like Road Noise, Durability, Fuel Economy, and so on, and and indication of how well they perform in that specific area. That way you could have an agreed industry benchmark for different aspects of the CPU/GPU that relate to stuff like your examples of spreadsheets, gaming, CAD/graphics, etc., and let people chose the chips that provide the most appropiate combination for them. The complication is that CPU/GPU performance is a steadily moving goalpost where as things like tyres don't evolve all that much, so you'd either need to adjust the scale periodically and/or use an exponential scoring system (also potentially confusing for the layperson).
"Users" in the context of "users of the Goo.gl service", which obviously includes both companies and individuals. Sure, many of the things Google has killed have been moribund and the singular form of user isn't probably too wide of the mark, but they've also killed products that have had a small but quite active community, and at least for now goo.gl seems to be closer to the latter category. By March next year though, or whenever they might eventually shut it down for good, I guess it'll be a lot smaller.
Of course it's a good thing. If you're Google. Here's the key feature for them from the FBL info page: "Dynamic Links can help migrate users from your website to your mobile app. Give them an easy way to send themselves a deep link that, when clicked on a mobile device, automatically opens in the right context within your app (even if they need to install the app first)." (emphasis mine).
This is all about getting more people locked into Google's app and advertising ecosystem where they can be more easily tracked and monetized, both through App sales and (of course) selling their data to marketers.
For now. This is Google we're talking about, you know, that company that has a habit of killing products that don't meet its internal targets, users be damned? If the links continue to work for any significant period of time after March 2019 I'm going to be very surprised. Also, as food for thought, Google is also also in a position to expand all the "goo.gl" entries in their own search databases to link directly to their intended targets while breaking them for everyone else in the search biz. Not that a company that does no evil would do that, of course./s
Oh, I'm sure many of them are well aware of that. The telling statement is this one: "But the fallout at the Silicon Valley company has been wide. According to two Facebook employees, workers have been calling on internal message boards for a hunt to find those who leak to the media " Yep, because when you're caught on the wrong side of a line (moral, legal, whatever) the best thing to do is to double down, hunt down those responsible for the leak and throw them against a wall, then carry on as normal, right? Some of the senior execs might to realise that they need to be seen taking action (seeing is one thing, actually doing, however...), but at least some of Facebook's employees are clearly completely on board with the "you have no privacy, we're going to sell ALL the data no matter what" business model.
Quite why it took this long to get to #DeleteFacebook I have no idea, but this really shouldn't be a surprise to anyone at this point.
Of course it's posturing, that's what a lot of diplomacy and politics is, when you get right down to it - same thing with the on-going expulsions of diplomats following the poisoning of the Skripals in the UK; tit-for-tat genitalia size comparisons. Still, as with.EU domain ownership rules, the rules of membership of the EEA are pretty clear as well, and Theresa May has stated repeatedly she's not going to accept any deal that includes some of those rules so, unless that changes, the UK is almost certainly going to be exiting the EEA, either on March 29 2019 or at the end of the "transition period" on 31st Dec 2020. I suppose you could make a case that they're jumping the gun a bit with this since the EEA exit date isn't yet set (unless it's buried in the transition period terms somewhere), but it seems pretty clear that a good deal of the UK's.EU domain owners are not going to meet the criteria and as such need to start making alternative arrangements.
The rules of the.EU domain are pretty clear - registrants are meant to have a connection to the EU, something many of the UK registrants will cease to have come Brexit. The revenue's not the issue here - a few €m per year is nothing in the scheme of things - so this is entirely about sticking the letter of the law and (possibly) an attempt to get another bunch of UK citizens to directly feel some pain from Brexit, of which some will hopefully complain about it and further weaken the UK government's overall position and Conservative MP's chances of retaining their seats.
Also, to nit, the UK has never been in the Eurozone as that specifically refers to countries that have adopted the Euro as their currency. They'll probably be obliged to join if they ever decide that they want to get back in, however, since adopting the Euro (and being part of Schengen) has been a mandatory condition of full membership for some time now. It's obviously going to take a *huge* change in mindset or a lot of people aging out before that even gets floated as a serious possibility in government though.
If you're going to make it a mandatory choice between the available options, then you'd probably need to phrase it as "Which smart speaker would you mistrust the least?", but I do think the comments as people explained their choices would be quite interesting (and in many cases probably quite debatable as well). There have been a few similar polls about "big tech" - one about which you'd be most willing to give up springs to mind - but I don't recall one specifcally about privacy. Given the current climate, now's probably as good a time as any for it.
You're thinking of Hong Kong; Taiwan/Formosa's statehood is complex, but it's essentially been an autonomous nation since 1912. The key point here though is that the ROC is an entirely separate entity to the PRC on the mainland so it's not *quite* the same as the blocked Broadcom takeover of Qualcomm. Going to be interesting to see how Trump handles this given the whole "One China Policy" though; the RPC Chinese are already sensitive about the trade issues, so one mis-tweet could easily exacerbate the political situation further.
Allegedly, in addition to misusing the data Facebook had knowingly provided (albeit under false pretenses of academic research), Cambridge Analytica also went above and beyond the accepted methods of acquiring the profile data using techniques that are skating a thin line between whether they are actually legal or not. Facebook was quite right to ding them. Political partisanship aide (yes, Obama did it too, and FWIW I found that usage rather disturbing as well, although the techniques used now seem to be on a whole other level), it's absolutely unethical and if not squashed now then you can bet that $party_you_dont_approve_of will using it come the next election.
Psychologically, that angle is also rather interesting - people don't like admitting they were scammed / maniupulated, and often get overly defensive as a method of coping with the subconcious knowledge that it has probably happened to them - it's one of the classic stages of acceptance. The reality is that advertisers, politicians, and other shills do this to us Every. Single. Day. and if you step back from the political partisanship it's pretty clear that a lot of voters on all sides got manipulated and had their well-targetted buttons pushed in the US election, the Brexit referendum, and several other elections Cambridge Analytica was supposedly involved in. Do we *really* want to leave that tool in their box unchecked, and continuing to become more and more effective?
You can, but the GDPR takes that even further by letting you request information on all the policies and procedures surrounding that data, as well as those of the partners that the recipient is sharing your information with. Most current data access requests essentially just let you ask "What?", the GDPR gives you the right to ask "How?", "When?", "Who?", and "Why?" as well. I think most of us here are probably well aware of what kinds of information Facebook etc. have and a vague idea of how they use it, but beyond that it all gets very opaque, very fast. That opaqueness is really the key point of the Facebook/Cambridge Analytica story, and something that I suspect the GDPR's SARs in the hands of privacy focussed activists and journalists are going to blow wide open. I'm fully expecting that the rabbit hole is going to go a *lot* deeper than all but the most suspicious/paranoid suspect.
While I agree with you main point that it's a bad comparison, The Onion does technically put a satirical and humorous slant on actual news, and although they tend to use parody more than satire in the premises for their stories the connection to real world events is usually fairly apparent. Unlike Private Eye, The Onion also blatantly makes a lot of stuff up too, of course.
I would be more interested in hearing about what data they actually got and what they paid for it, I want to know more about that market value.
Can't help you on what they paid, but it seems pretty clear that one way or another Cambridge Analytica got hold of pretty much the entire contents of all those 50m Facebook profiles, including stuff that their owners (or as Zuck once supposedly called them, the "dumb fucks") thought was actually "private". If you're in the EU and have a FB profile then you can find out all about want profile contains - and much, much, more! - come May 28th when the GDPR comes into force by hitting them up with a Subject Access Request, or "SAR". Here's a template to get you started.
In terms of illegal activity that seems to be all that can be claimed, and it's a civil issue at that. So far, anyway. Facebook appears to have gone into full-on panic mode all of a sudden which makes me think there's a lot more to this than has been made public yet. Or they just really, really, fear the regulation that seems like it's almost inevitable at this point, at least in the EU, and I dare say Trump will tweet out the US' position soon enough. IIRC, Zuck's a Democrat and Trump's not that fond of Democrats *or* Silicon Valley execs, so place your bets...
Supposedly Facebook's CSO, Alex Stamos, who actually wanted Facebook to look into the Russian misinformation campaign during the US elections, is leaving Facebook after clashes with other senior management, most notably Sheryl Sandberg. Even more potentially damning is that according to Carole Cadwalladr Facebook staff were in Cambridge Analytica's London offices "securing data" when agents of the UK's ICO turned up to do the same. Whether this occured before Cambridge Analytica became the subject of a formal request for a seach warrant is a little unclear though. I think this is starting to look like it's might have got some real legs to it, so I'm going to stock up on the popcorn and see how deep the rabbit hole goes.
Facebook and Cambridge Analytica can at least count their blessings on one thing though; they managed to have all this blow up in their faces *before* the GDPR kicked in.
Slashdot Mirror
On a busy highway, absolutely - there are any number of instances of people being injured or killed while attempting to get to the shoulder, or even after having safely done so, because they were struck by another vehicle. Expecting an autonomous vehicle to handle that safely in any situation is clearly a non-starter with the current state of play, but hopefully there is at least *some* code to try and automatically handle emergency situations in use by now, e.g. if an emergency vehicle is detected approaching from the rear on a clear stretch of road, to at least slow down and pull over as far as possible towards the shoulder. My interest was where on the spectrum from simply trying to revert to human control through to successfully handling the necessary maneouvers current automomous vehicles might be, although I do suspect it is indeed much closer to the former.
Accordining to TFS there was (as is usually the case with most autonomous vehicles being tested) a human driver present who was able to intervene and bring the car to a halt; The company claims the human test driver did everything right but is now responsible for the citation, so presumably sucks to be him/her. Having made that statement I would hope that Cruise will now cover any costs arising from the almost inevitable legal challenge to the citation though.
That said, I am curious what the current state of the art is for self-driving vehicles with regards to an emergency services vehicle approaching from the rear, or in the direction of travel. While it's possible that the code might detect an emergency vehicle and take specific action like slowing down or pulling over to allow it to pass, I'd be very surprised if we're at the point where it's capable of detecting that it is actually being requested to halt for a ticket. Also, what kind of accuracy there is for discriminating between actual emergency vehicles and things like maintenance vehicles that just have regular hazard lights but might not warrant any additional actions.
Do they, or do we need to wait and see how it works out in practice - most probably with the Microsoft case with the Dublin data centre? I've not seen any detailed breakdowns of the act yet, but it seems that in order to get access to data US agencies still have to ask for it via the government of the nation where the data centre is located to approve the request. There's two ways that can go; "Sure, here you go!", and "Nope, sorry, that data is protected by our privacy laws". That means it's going to come down to whether or not the approving body/individual is more concerned about breaching EU privacy laws, avoiding a potential diplomatic issue, or just keeping their US colleagues happy, and until such time as there's some clear EU legal precedent I don't think it's possible to say which it's going to be. It also might vary from country to country as well, since the EU's laws are not entirely homogenised as many seem to think.
Similar era here; cycles counted and the innermost of a deep nest of loops was usually a good place to start as even one less processor cycle could be removed you could often improve things considerably when you multiplied it all out, although I don't think that's it since there are a few competing DNS engines, which all hopefully pretty well optimised in their core code already at this point (feature bloat aside), and presumably Cloudflare is only going to be running one of them. Realistically, I'm only seeing two options here - Cloudflare coming to some kind of "arrangement" with APNIC for some memorable IP addresses for DNS (one of the few times you do need to memorise them, so that counts) to compete with Google, Quad9, etc. in the alternative DNS service provider space as a new revenue stream, hence people getting upset about the sudden repurposing of what they saw as IPs reserved for reseach use. The second, and I suspect more likely scenario given Cloudflare's DDoS protection services, is that they were fully aware they were going to be getting tens of Gb/s of junk traffic and wanted to do what could be some genuinely useful work on protected DNS services from DDoS attacks in a live environment, since the traffic patterns are likely "good enough". That does still make sense to be done in the IPv4 space since that's where the IRL garbage traffic is and, as you say, any lessons learnt should hopefully translate over to the IPv6 stack easily enough.
There's also a lot coming from captive portals that use 1.1.1.1 as a login/logout gateway IP, including some turnkey solutions provided by the likes of Cisco that are heavily deployed in providing free WiFi services to things like the hospitality trade. Yeah, they could (and should!) have used RFC1918 IPs as the default configuration for this, just like your home router tends to default to 192.168.1.1, but for whatever reason decided to default to 1.1.1.1 instead. Since that (fairly obviously) is highly unlikely to conflict with anything already on the network, guess what got deployed in the live environment?
The IPs are assigned to APNIC, a RIR, and they are free to assign them to whoever they want that meets their assignment policies, including entitities that are not Headquartered in the APNIC region. There is some debate in high-level networking groups like NANOG about whether those procedures were correctly followed, but that ultimately hinges on whether this is a joint APNIC-Cloudflare research project or a permanent assignment. The former is arguably within APNIC's currently agreed scope for the IPs in question, the latter may have circumvented a few procedures or opportunities for debate.
Ultimately though, the last time these IPs were routed - a partnership between APNIC and Google, they got 50Gb/s of garbage, mostly from things that were designed to use unassigned IPs rather than suitable RFC1918 IPs. There are not exactly very many companies that have the necessary infrastructure to filter out 50Gb/s of crap and still provide a useable service with what remains so, research or not, I can't see many people wanting these IPs anyway and if Cloudflare can make some use of them, good on them. Besides all that, there is also the question of why are people still doing "research" on IPv4 space; wouldn't it be better to be focussing on the brave new world of IPv6 - where's my 2001:2001:: resolver, or some such equivalent?
I was replying to someone who specifically mentioned Facebook, and also mentioned Facebook specifically, in a thread discussing Facebook so I think the context should be a pretty big clue as to who "they" meant. Even so, sure you can easily apply that to any of the companies that you can identify who track you and store your data, which probably covers most of the largest data hoarders out of the thousands, so it is a start at least and covers the worst offenders. A SAR can also include a request for information on any third parties your data has been shared with, and that's going to include information a LOT of those thousands of other companies that are more below the radar - and remember the only cost to an individual to create and issue a SAR is some time, so expect a lot of EU privacy activists to be firing off a lot of SARs until they get bored with it.
For those companies that are not on the ball with this, it's potentially death by a thousand admin-cost cuts dealing with the SARs, death by EU fines for demonstrable non-compliance for failing to comply with SARs, or hoping to $deity they can either continue to fly below the radar indefinitely, or at least avoid paying any penalties by virtue of having no EU presence. Quite how the EU courts expect to deal with that last scenario, or more shady operators that have faked company info for that matter, I have no idea (although "unsucessfully" does seem like the most probable answer) but it's going to be interesting finding out, and probably quite expensive for the unlucky ones since it's pretty obvious that this is also a good way for the EU to plug some budget deficits if it can get some major prosecutions under its belt.
Or, more interestingly, what kinds and quantity of data they have on people who have never had an account to start with, but have been tagged by family/friends/colleagues that do. Really curious to see what kinds of data people in that situation who hit Facebook up with an SAR once the GDPR comes into force are going to get back, although they may need to create an account and make a few posts to it in order to link their "offline profile" data to their personal ID.
I suspect you're not the only one planning on doing that. You might find this template useful for maximising your return, because the more points they have to fsck up the better, right?
It's not a "bit" overblown; it's pure hyperbole/clickbait. The monitors are fine, and that includes with MacOS, the issue is that the monitors will no longer work when they are connected through a third party peripheral with drivers that do some encapsulation to send display data (amongst other things) over USB.
Laden or unladen?
That's because, like the OP, you're trying to cram multiple metrics into a single value - whether it's CPUs, GPUs, shoes, or whatever else. Clearly that's not going to work. Perhaps something akin to the "Performance Charts" that many industries have adopted might work though - those colourful charts that typically give a series of ratings for a number of key metrics, e.g. in the case of car tyres stuff like Road Noise, Durability, Fuel Economy, and so on, and and indication of how well they perform in that specific area. That way you could have an agreed industry benchmark for different aspects of the CPU/GPU that relate to stuff like your examples of spreadsheets, gaming, CAD/graphics, etc., and let people chose the chips that provide the most appropiate combination for them. The complication is that CPU/GPU performance is a steadily moving goalpost where as things like tyres don't evolve all that much, so you'd either need to adjust the scale periodically and/or use an exponential scoring system (also potentially confusing for the layperson).
"Users" in the context of "users of the Goo.gl service", which obviously includes both companies and individuals. Sure, many of the things Google has killed have been moribund and the singular form of user isn't probably too wide of the mark, but they've also killed products that have had a small but quite active community, and at least for now goo.gl seems to be closer to the latter category. By March next year though, or whenever they might eventually shut it down for good, I guess it'll be a lot smaller.
Of course it's a good thing. If you're Google. Here's the key feature for them from the FBL info page: "Dynamic Links can help migrate users from your website to your mobile app. Give them an easy way to send themselves a deep link that, when clicked on a mobile device, automatically opens in the right context within your app (even if they need to install the app first)." (emphasis mine).
This is all about getting more people locked into Google's app and advertising ecosystem where they can be more easily tracked and monetized, both through App sales and (of course) selling their data to marketers.
For now. This is Google we're talking about, you know, that company that has a habit of killing products that don't meet its internal targets, users be damned? If the links continue to work for any significant period of time after March 2019 I'm going to be very surprised. Also, as food for thought, Google is also also in a position to expand all the "goo.gl" entries in their own search databases to link directly to their intended targets while breaking them for everyone else in the search biz. Not that a company that does no evil would do that, of course. /s
Oh, I'm sure many of them are well aware of that. The telling statement is this one: "But the fallout at the Silicon Valley company has been wide. According to two Facebook employees, workers have been calling on internal message boards for a hunt to find those who leak to the media " Yep, because when you're caught on the wrong side of a line (moral, legal, whatever) the best thing to do is to double down, hunt down those responsible for the leak and throw them against a wall, then carry on as normal, right? Some of the senior execs might to realise that they need to be seen taking action (seeing is one thing, actually doing, however...), but at least some of Facebook's employees are clearly completely on board with the "you have no privacy, we're going to sell ALL the data no matter what" business model.
Quite why it took this long to get to #DeleteFacebook I have no idea, but this really shouldn't be a surprise to anyone at this point.
Of course it's posturing, that's what a lot of diplomacy and politics is, when you get right down to it - same thing with the on-going expulsions of diplomats following the poisoning of the Skripals in the UK; tit-for-tat genitalia size comparisons. Still, as with .EU domain ownership rules, the rules of membership of the EEA are pretty clear as well, and Theresa May has stated repeatedly she's not going to accept any deal that includes some of those rules so, unless that changes, the UK is almost certainly going to be exiting the EEA, either on March 29 2019 or at the end of the "transition period" on 31st Dec 2020. I suppose you could make a case that they're jumping the gun a bit with this since the EEA exit date isn't yet set (unless it's buried in the transition period terms somewhere), but it seems pretty clear that a good deal of the UK's .EU domain owners are not going to meet the criteria and as such need to start making alternative arrangements.
The rules of the .EU domain are pretty clear - registrants are meant to have a connection to the EU, something many of the UK registrants will cease to have come Brexit. The revenue's not the issue here - a few €m per year is nothing in the scheme of things - so this is entirely about sticking the letter of the law and (possibly) an attempt to get another bunch of UK citizens to directly feel some pain from Brexit, of which some will hopefully complain about it and further weaken the UK government's overall position and Conservative MP's chances of retaining their seats.
Also, to nit, the UK has never been in the Eurozone as that specifically refers to countries that have adopted the Euro as their currency. They'll probably be obliged to join if they ever decide that they want to get back in, however, since adopting the Euro (and being part of Schengen) has been a mandatory condition of full membership for some time now. It's obviously going to take a *huge* change in mindset or a lot of people aging out before that even gets floated as a serious possibility in government though.
If you're going to make it a mandatory choice between the available options, then you'd probably need to phrase it as "Which smart speaker would you mistrust the least?", but I do think the comments as people explained their choices would be quite interesting (and in many cases probably quite debatable as well). There have been a few similar polls about "big tech" - one about which you'd be most willing to give up springs to mind - but I don't recall one specifcally about privacy. Given the current climate, now's probably as good a time as any for it.
You're thinking of Hong Kong; Taiwan/Formosa's statehood is complex, but it's essentially been an autonomous nation since 1912. The key point here though is that the ROC is an entirely separate entity to the PRC on the mainland so it's not *quite* the same as the blocked Broadcom takeover of Qualcomm. Going to be interesting to see how Trump handles this given the whole "One China Policy" though; the RPC Chinese are already sensitive about the trade issues, so one mis-tweet could easily exacerbate the political situation further.
Allegedly, in addition to misusing the data Facebook had knowingly provided (albeit under false pretenses of academic research), Cambridge Analytica also went above and beyond the accepted methods of acquiring the profile data using techniques that are skating a thin line between whether they are actually legal or not. Facebook was quite right to ding them. Political partisanship aide (yes, Obama did it too, and FWIW I found that usage rather disturbing as well, although the techniques used now seem to be on a whole other level), it's absolutely unethical and if not squashed now then you can bet that $party_you_dont_approve_of will using it come the next election.
Psychologically, that angle is also rather interesting - people don't like admitting they were scammed / maniupulated, and often get overly defensive as a method of coping with the subconcious knowledge that it has probably happened to them - it's one of the classic stages of acceptance. The reality is that advertisers, politicians, and other shills do this to us Every. Single. Day. and if you step back from the political partisanship it's pretty clear that a lot of voters on all sides got manipulated and had their well-targetted buttons pushed in the US election, the Brexit referendum, and several other elections Cambridge Analytica was supposedly involved in. Do we *really* want to leave that tool in their box unchecked, and continuing to become more and more effective?
You can, but the GDPR takes that even further by letting you request information on all the policies and procedures surrounding that data, as well as those of the partners that the recipient is sharing your information with. Most current data access requests essentially just let you ask "What?", the GDPR gives you the right to ask "How?", "When?", "Who?", and "Why?" as well. I think most of us here are probably well aware of what kinds of information Facebook etc. have and a vague idea of how they use it, but beyond that it all gets very opaque, very fast. That opaqueness is really the key point of the Facebook/Cambridge Analytica story, and something that I suspect the GDPR's SARs in the hands of privacy focussed activists and journalists are going to blow wide open. I'm fully expecting that the rabbit hole is going to go a *lot* deeper than all but the most suspicious/paranoid suspect.
While I agree with you main point that it's a bad comparison, The Onion does technically put a satirical and humorous slant on actual news, and although they tend to use parody more than satire in the premises for their stories the connection to real world events is usually fairly apparent. Unlike Private Eye, The Onion also blatantly makes a lot of stuff up too, of course.
Can't help you on what they paid, but it seems pretty clear that one way or another Cambridge Analytica got hold of pretty much the entire contents of all those 50m Facebook profiles, including stuff that their owners (or as Zuck once supposedly called them, the "dumb fucks") thought was actually "private". If you're in the EU and have a FB profile then you can find out all about want profile contains - and much, much, more! - come May 28th when the GDPR comes into force by hitting them up with a Subject Access Request, or "SAR". Here's a template to get you started.
In terms of illegal activity that seems to be all that can be claimed, and it's a civil issue at that. So far, anyway. Facebook appears to have gone into full-on panic mode all of a sudden which makes me think there's a lot more to this than has been made public yet. Or they just really, really, fear the regulation that seems like it's almost inevitable at this point, at least in the EU, and I dare say Trump will tweet out the US' position soon enough. IIRC, Zuck's a Democrat and Trump's not that fond of Democrats *or* Silicon Valley execs, so place your bets...
Supposedly Facebook's CSO, Alex Stamos, who actually wanted Facebook to look into the Russian misinformation campaign during the US elections, is leaving Facebook after clashes with other senior management, most notably Sheryl Sandberg. Even more potentially damning is that according to Carole Cadwalladr Facebook staff were in Cambridge Analytica's London offices "securing data" when agents of the UK's ICO turned up to do the same. Whether this occured before Cambridge Analytica became the subject of a formal request for a seach warrant is a little unclear though. I think this is starting to look like it's might have got some real legs to it, so I'm going to stock up on the popcorn and see how deep the rabbit hole goes.
Facebook and Cambridge Analytica can at least count their blessings on one thing though; they managed to have all this blow up in their faces *before* the GDPR kicked in.