Telemetry used to be about diagnosing issues and improving software.
"Telemetry" as we see it today is rarely about those things. Its mostly about gathering user information for sale to third parties (primarily advertisers.)
So its hardly surprising that trust in "telemetry" has degraded as the term has more come to imply "personalized ads" rather than "improved experience." In fact most people consider the addition of ads to software (personalized or otherwise) to be the exact opposite of an improved user experience.
Until all competing products do the same thing. Then all you're left with is complaints. And make no mistake, if this is determined to be a success (or at least not a big disaster) then its almost certain the rest of the industry will follow suit, sooner or later.
nVidia is risking pissing us all off by this move while their competitors aren't, but AMD would be risking essentially nothing if they do the same thing in a couple months since there's not really any other options for people to move to. Intel's a very distant third place and not really attempting to compete at the cutting edge. Whoever is below Intel isn't even worth discussing at this point.
A low-competition market doesn't have to be an actual oligopoly to screw over their customers. Sometimes it just takes one producer to risk pulling the trigger on something only-kind-of-bad and everyone just follows along if the action shows overall benefit to the bottom line. This scenario might not get away with actions as bad as a true oligopoly but it can still fall well into the "not good" category.
There's a large difference between "can't" and "won't." They probably _could_ explain it but they just don't want to because your personal information is now their trade secret that they can use and/or sell at their discretion and certainly wouldn't want their competitors to get their hands on.
The presidency might be the big flashy one, but all of these discussions apply equally to state governors, city mayors, and any other elected official at any level of government.
Not to mention issue votes. I imagine the pro-lifers wouldn't be too off-put if they were given the ability to tamper with any vote related to abortion laws for example.
there are people in the polling place watching for you to be doing something like that
Except that they don't usually get to see into your voting booth so it would be hard for them to tell if you pulled out your phone after the curtain was drawn.
Then again, its also easy enough to tell your boss "sorry they caught me before I could take a picture sucks to be you." (Then again, a boss who's willing to coerce your vote probably wouldn't be opposed to punishing you for acts outside your control whether you were telling the truth or not.)
There is no receipt. You fill in the Scantron form and feed it through the machine. The form is then left with the electioneers so that they can perform a manual recount later if needed.
I of course can't say that its done that way everywhere, but even if there's a receipt to show _that_ you voted, there's no reason for it to indicate who you voted for. You don't have anything you could verify that against later even if you wanted to.
Yes you have to trust the electioneers if a recount is called, but no more than you would in a fully paper-based vote. Well at least assuming that your local organizer isn't an idiot and does something like "used forms get put in a garbage pail out in an unmonitored hallway." But again the same could happen with full-paper votes so you're still not losing any trust there that you would have had otherwise.
Yep. Trouble is that we don't trust electronic voting machines (with some justification to be sure) so we want to be able to verify our vote was recorded (and recorded correctly.) Unfortunately if we can verify our vote, so could whoever was coercing our vote in the first place.
I'm not sure how that can be solved. We trust paper votes not because they're paper but because we have multiple people counting (or at least monitoring the counts) so that the need for trust can be diffused across multiple actors whereas with electronic voting, we have to trust a single entity (the manufacturer of the machine.)
Someone higher up on the thread suggested having multiple independent voting machines which all would get us back to the trust diffusing (albeit at significant hardware and maintenance costs.) Of course, that leads to a potential UI issue since the user isn't going to want to have to record their vote independently on 2 or 3 separate screens, and if you only have one screen then you're back to a single point of trust.
Perhaps N+1 screens -- one shared input screen that works like normal and feeds all of the N independent voting machines, and then each machine would have a display to confirm how they recorded the vote. Sort of like a vote receipt but a) doesn't waste paper and b) doesn't leave a physical receipt that coercive agents could use to verify your vote. A final confirm/reenter prompt and off you go.
All N machines would have to be compromised (and in exactly the same) way in order for vote count fraud to not be detected -- that is you'd be able to spread your trust across 2 or 3 manufacturers instead of a single one.
1) You have to trust that because the system says it counted your vote, that it actually did. OK so that's not TOO tough -- if a full list of voting IDs is posted, you can just count the rows and compare to the total votes.
2) Here's the hard part though: You need to verify that your vote counted for the candidate you selected. That means you have to have something in your possession that can be matched against the posted table. This could be as simple as just directly listing the candidate chosen in the table or it could also be obscured, but even in the latter case your boss could coerce whatever access/verification code you received just as easily as he initially coerced you.
3) And finally, you have to trust that the posted list reflects the actual recorded votes. If the list plainly states the candidates chosen that this is as trivial as #1, but if they're obscured (and I'm assuming they'd be obscured in some non-trivial way,) then each person could verify their own vote but would have no way of counting up the number of votes per candidate. So the posted list could reflect the true votes of every person, and the total votes would still have to match, but there would be no way to verify exactly how many votes that each candidate got -- X+Y = (X+Z)+(Y-Z) for any Z you choose.
There's probably some cryptographic way of solving that as well, but "easy" is not an adjective I would necessarily select -- and none of that stops your boss from coercing your vote (and verifying it afterward if necessary, at least assuming you yourself have the possibility of verifying it.)
And how do you guarantee that the manufacturer hasn't added the odd extra circuit that wasn't in the base design? I guess if 3D printing gets advanced enough that you can print your own electronics.. but then you still have the compiler backdoor issue https://en.wikipedia.org/wiki/Backdoor_(computing)#Compiler_backdoors if the printer was built to recognize certain designs and modify them as it was printing. So now you have to build your own 3D printer (and it has to be advanced enough to print electronics.) And then whatever you used to build that has to be analyzed and so on until you're basically starting with a warehouse full of raw materials and thousands or even tens of thousands of hours ahead of you that you can't contract out because you can't trust other people to not build in back doors either. And THEN you have to somehow guarantee that you did everything perfectly and didn't leave any accidental flaws or the whole exercise was for naught.
Obviously those attacks are getting rather complex at that point, and its certainly a lot easier to trust a self-printed circuit created from an open design than it is to trust Windows or any other corporately-controlled system. But at some level, even going to all that trouble still leaves theoretical security risks.
And finally you then use your perfectly secured machine to check out Slashdot for the first time in 4 years after building your own browser because you don't really trust Firefox either, and simply sending and receiving the minimally necessary packets is supplying every router along the way with at least a bit metadata about your location and actions.
At some point, you'll eventually have to accept that risk is a thing and manage it rather than trying to eliminate it, because the latter is very close to impossible. The only true way to guarantee that your computer isn't sending out some sort of information is to turn it off, unplug it, smash it, and move its remains somewhere far away from you just in case its got some embedded RF chip or whatever that you didn't manage to smash.
Detailed hardware info could be a potential privacy risk, though a very low one.
A list of games you own is likely very personalized. Everyone has their own preferences not to mention which specific games are installed at any one time out of a possibly large digital library -- even if nVidia can't read the full Steam or GoG or Origin or whatever libraries (and they probably can't -- at least not yet,) the subset of installed games at any one time could theoretically add a temporal component to an already quite personalized data set, even if its a relatively granular measurement, without having to send any additional information such as last login timestamps or whatever.
Its not. I've got an nVidia card and I'd found the tasks before, but I'd assumed they were associated with geForce Experience (which I removed as soon as I discovered that they force a login even to just run the driver update -- which is the only reason I ever cared about that app in the first place.. I don't use shadowplay or game profiles or any of that other crap.)
I ended up having to use the DDU since the driver uninstaller crashed repeatedly, essentially giving me a full clean reinstall (sans Experience) when I was done.
And yeah after seeing the article I checked again and sure enough, the telemetry tasks were back.
Not sure about the people who claim to not see them.. Maybe there's some sort of A/B trial going on, or perhaps it was only in one specific version and they've already backed off on it, or maybe it only gets applied in specific hardware setups or who knows what else. But it definitely is (or hopefully by now.. was) a thing.
Unfortunately, unless there's enough backlash to make nVidia change their mind.. there's probably at most 2 years before AMD and Intel and every other video card manufacturer (are there still others?) hop on the "me too" bandwagon since turning down an essentially free income source is rather un-American. And any sort of user tracking is a potential income source these days.
There's a huge difference between a product like cheap knock-off toys and an information service like Google.
One is personalized, the other isn't. I'm sure the Chinese would have absolutely no problem buying American products like clothing and DVDs.. except they have no reason to do so since they make all that shit themselves at a fraction of the cost of US producers (and then sell it to us since much of it is produced under license from US or at least western-based companies because again.. fraction of the cost.)
Similarly, if we actually cared about Chinese information services, we probably would start seeing limitations put on them. But like the Chinese with physical products, the US has little need for outside information services since we still produce the (subjectively) best services on the planet, with some input from Europe and Japan here and there.
There's another angle here as well -- the rules that China's making apply equally to their national businesses, so its not an explicit barrier to international trade -- its a barrier to business in general. Now the fact that Chinese companies were probably already doing things like that while foreign companies typically wouldn't have been is another question but its not a question of trade policy as such since the rules are the same on both sides in this case and "I'm not used to that rule" isn't a solid argument.
Imagine if we allowed China to ignore lead safety rules in their products just because they aren't used to having to design things that don't kill people. Obviously that sounds ridiculous but it would be a (somewhat) similar "barrier" to Chinese imports.
No, just enough metadata hanging around to make the requirement trivial anyway. At least with an explicit real name requirement, the user knows they're probably being tracked and spied on. Not that that's much solace.
no "prohibited" topics
No, just topics that will get you sent to gitmo, or sued out of existence (though that's more done by shady companies than the government.) Still, not technically prohibited.
data localization is not required
True, though this is by far the most benign issue on the list, and in some ways non-localized data is a risk as well (if national or even just corporate secrets happen to be stored on a non-local server, its entirely possible for the foreign government to just confiscate the server.) Basically, would you want your emails and data stored in China? Then why would you think the Chinese want theirs stored in the US?
America's monitoring is designed to catch bad guys, not to suppress discussion or prevent the flow of information
Except when they decide simply having/sharing information makes you a "bad guy."
Now certainly China is going to be worse than the US and most other western nations in basically all of those categories, but please don't mistake that for the assumption that the US doesn't participate in them at all -- America does many similar if not the same things, just at a smaller (and often more subtle) scale.
And that's just the government. Many of the largest digital companies in the US are well-known for censoring topics they don't like, and several have attempted to enforce real-name policies. Given how much emphasis the US in particular puts on corporations, that's still a fairly large issue even if its not as severe as the government performing those actions (companies at least don't have the power to imprison you, legitimately or otherwise.)
I don't really know anything about Ataturk, but just in general converting to a democracy is actually kind of hard. Telling someone that what they've known their entire life is wrong or bad can be pretty jarring, especially when their politics and religion is all mashed together to the point that democracy could almost appear to be denying God's will since many non-secular dictators tend to suggest that their power is granted to them by God.
And even if you switch to a democracy, you immediately need to guard against is voting in a prime minister (or president, depending on your setup) who just grabs all power for himself and decides that the next election will be "some day," effectively leaving you with a new dictatorship and a coat of paint over it. There's a lot of "democracies" in the world even today that fit this model.
The only real way to prevent this, at least it seems, is to have a benevolent dictator open up the lower ranks of government to open vote but still retain the ultimate power to override anything stupid that gets done, until such time as both the politicians and the people realize what their roles are in the new society and then the dictator can vacate.
That's no easy task either. Benevolent dictators are few and far between at the best of times, and we're talking a scale of a couple generations give or take -- which is only like 5 or 10 election cycles under the US' 4 year cycle length. Even if you have a benevolent dictator at the start, they've got to either survive that long or be able to hand the reigns over to someone who shares the same mentality.
The US itself gets to be a bit of a special case because they sort of had a "benevolent" dictator in the form of the British government (benevolence in that case mostly being due to sheer distance and inability to promptly squash troublemakers.) So as a country they appear to have started as a straight up democracy but historically its mostly just because nobody called them a separate country during the lead-up period.
Well yes that would be majorly detrimental to my boss. I on the other hand would benefit greatly from the resulting payout.
Depends how well that interview went. And whether your boss thought you were worth increasing your pay or just lets you go for being disloyal. Most people don't tell their boss that they're looking for new employment until they're already fairly certain they've landed something for a reason.. or unless they're basically just bluffing in order to get a raise.
And here's a great scenario that is countered by evidence, given how the ability to track your phone location accurately already exists, as does advertising.
Yes, but so far those things aren't linked (unless you explicitly download McDonald's app or something.) I'm talking about a world where your tracking information is just publicly available (for the narrower definition of "public" that means "large corporations," which seems to be the definition that those kind of things tend to be targeted towards.)
Their problem is they didn't act on anything until after it was too late. They knew exactly where he went and when and didn't think much of it until later.
My guess is "too late" was right around the time he decided to start doing something that would cause the NSA to want to pay attention. Who cares how many times he went grocery shopping in the year prior to leaking the documents? They want to know where he is _now_. And they would have especially loved to know where he was prior to his being granted asylum (ie: if anything happened to him today that could be tracked to US interference, there would be a bit of a political stink raised I'd guess, as that would somewhat undermine Russian security.)
wondeful doomsday scenarios that just simply don't pan out
Yet. It only takes apathy to give up your freedoms. It usually takes a revolution to get freedoms back. Are you willing to sit back and do nothing just hoping that these doomsday scenarios will continue to not pan out?
Trump obviously. Hilary does as much as she can to not talk about the things people want to hear -- because most of what we want to hear about her is details of her numerous scandals.
Trump lies to your face. Hilary schemes behind your back. Hard to say which is worse. Great time for US politics.
That's not always as easy as it sounds. If there was data conversions involved for example, the previous stable build may not even run anymore and would require restoring everything from backup, which may well be a many-many-hour project in itself -- and possibly taking time away from fixing the issue if it was a small-to-mid size company that recycles people into multiple roles (and programmer/IT services is a frequent combination at the best of times.) Just in time to turn around and have to re-convert as soon as you're done because the fix has been completed.
Never mind the fun of the programmers telling you "it'll just be another 2 hours" for 18 hours straight because issues in software tend to branch out in ways that nobody thinks about/remembers and can't include in their estimates until their nose is already in the code and its looking them in the face.
That's understandable (to a degree) in the situation where there was a set schedule ahead of time and things ran over. Then management has to make a decision whether its going to be a bigger hit to their reputation to delay vs releasing garbage.
What we often see though, especially in direct B2B-type software where there's a more intimate relationship between vendor and customer, is that the conversation goes more like this:
Manager: "We want something that does X" Engineer: "OK that will take 6 months" Manager: "Alright we'll call it a year to be safe" Sales: "We sold one! Needs to be installed by end of next month" Manager & Engineer: "#@#%$#@"
And that's with a good manager. Bad managers will give you 3 months when you say 6.. which doesn't really matter in these scenarios because sales has still only given you less than 2.
Jamming 2 weeks of work into 1 week is going to result in cut corners no matter what methodology you're using (or even what line of business you're in, for that matter.)
If you switch to a methodology where you're estimating in 6 month blocks and you're off by 100% like that, you're now 6 months off schedule instead of one week off -- that's even worse!
Not to say agile isn't misimplemented regularly, but if you're not schedules are off by that much of a margin, you need to start by looking at how you're generating time estimates before you bother changing your entire methodology.
Heh. The sad thing is, the telemarketer probably has a better time tracking you than the cops do. At least in the US. John Oliver did a segment on that a few months ago (https://www.youtube.com/watch?v=A-XlyB_QQYs.)
You can argue whether or not Uber is a "taxi" and thus subject to taxi regulations but you can't really argue that a flying car isn't an aircraft. And the FAA regulates _all_ air traffic from toy drones right up to giant commercial jet liners and everything in between.
That would be more like if Uber claimed their drivers weren't subject to normal road laws like traffic signs. The taxi thing would be more like if Southwest Airlines started complaining about the new service, which is unlikely as I doubt anything Uber comes up with will be in a competitive situation with traditional commercial airlines.
It would certainly be detrimental if your boss decided to check up on you on your day off and discovered you were at a competitor's office -- kind of suggests you're looking at other employment.
Or something less drastic: If McDonald's notices you're close to a Burger King and suddenly you get 14 text messages with deals for Big Macs. Perhaps not "incredibly" detrimental but certainly annoying as hell, especially if you happened to just be sitting at a stop light and had no intention of going into Burger King in the first place.
And that's not even starting to get into the really serious issues. Imagine if the US Govt had the capability of tracking Snowden back in 2013.. or even today. I'm sure it wouldn't be too hard to sneak an operative into Russia.
Its not just location -- privacy in general is granted as a right in the US constitution (as well as many many other countries to some extent or another) not because every person requires it 100% of the time, but because many people require it _sometimes_ and you never know when you will become that person for whatever reason.
Most constitutional rights are like that. The right to free speech is kind of meaningless for the vast majority of most peoples' lives because very few of us say anything that's super controversial, and even more rarely do we say it to someone who cares enough to persecute us. But those few people who truly need freedom of speech benefit us all by being able to say what they need to say without (as much) fear.
The trouble is always the carrot.. well that and poorly designed interfaces.
Eventually someone will invent something that a significant number of people "must" have. And then your browser will give you a single all-encompassing "allow this site to access your bluetooth devices?"
And even though all you really wanted was to allow FB to upload images to your bluetooth-enabled digital picture frame, suddenly FB (and all of their apps and partners and whoever else) also has access to your mouse and your gamepad and your webcam and every other device you own that happens to have a bluetooth interface.
I mean there's no question that you need "simple" interfaces for the technically illiterate, but unfortunately that seems more often than not to end up being insanely broad and almost never do they even have an "advanced" mode for people who know what they're doing to have more control.
Slashdot Mirror
Telemetry used to be about diagnosing issues and improving software.
"Telemetry" as we see it today is rarely about those things. Its mostly about gathering user information for sale to third parties (primarily advertisers.)
So its hardly surprising that trust in "telemetry" has degraded as the term has more come to imply "personalized ads" rather than "improved experience." In fact most people consider the addition of ads to software (personalized or otherwise) to be the exact opposite of an improved user experience.
Until all competing products do the same thing. Then all you're left with is complaints. And make no mistake, if this is determined to be a success (or at least not a big disaster) then its almost certain the rest of the industry will follow suit, sooner or later.
nVidia is risking pissing us all off by this move while their competitors aren't, but AMD would be risking essentially nothing if they do the same thing in a couple months since there's not really any other options for people to move to. Intel's a very distant third place and not really attempting to compete at the cutting edge. Whoever is below Intel isn't even worth discussing at this point.
A low-competition market doesn't have to be an actual oligopoly to screw over their customers. Sometimes it just takes one producer to risk pulling the trigger on something only-kind-of-bad and everyone just follows along if the action shows overall benefit to the bottom line. This scenario might not get away with actions as bad as a true oligopoly but it can still fall well into the "not good" category.
There's a large difference between "can't" and "won't." They probably _could_ explain it but they just don't want to because your personal information is now their trade secret that they can use and/or sell at their discretion and certainly wouldn't want their competitors to get their hands on.
The presidency might be the big flashy one, but all of these discussions apply equally to state governors, city mayors, and any other elected official at any level of government.
Not to mention issue votes. I imagine the pro-lifers wouldn't be too off-put if they were given the ability to tamper with any vote related to abortion laws for example.
there are people in the polling place watching for you to be doing something like that
Except that they don't usually get to see into your voting booth so it would be hard for them to tell if you pulled out your phone after the curtain was drawn.
Then again, its also easy enough to tell your boss "sorry they caught me before I could take a picture sucks to be you." (Then again, a boss who's willing to coerce your vote probably wouldn't be opposed to punishing you for acts outside your control whether you were telling the truth or not.)
There is no receipt. You fill in the Scantron form and feed it through the machine. The form is then left with the electioneers so that they can perform a manual recount later if needed.
I of course can't say that its done that way everywhere, but even if there's a receipt to show _that_ you voted, there's no reason for it to indicate who you voted for. You don't have anything you could verify that against later even if you wanted to.
Yes you have to trust the electioneers if a recount is called, but no more than you would in a fully paper-based vote. Well at least assuming that your local organizer isn't an idiot and does something like "used forms get put in a garbage pail out in an unmonitored hallway." But again the same could happen with full-paper votes so you're still not losing any trust there that you would have had otherwise.
Yep. Trouble is that we don't trust electronic voting machines (with some justification to be sure) so we want to be able to verify our vote was recorded (and recorded correctly.) Unfortunately if we can verify our vote, so could whoever was coercing our vote in the first place.
I'm not sure how that can be solved. We trust paper votes not because they're paper but because we have multiple people counting (or at least monitoring the counts) so that the need for trust can be diffused across multiple actors whereas with electronic voting, we have to trust a single entity (the manufacturer of the machine.)
Someone higher up on the thread suggested having multiple independent voting machines which all would get us back to the trust diffusing (albeit at significant hardware and maintenance costs.) Of course, that leads to a potential UI issue since the user isn't going to want to have to record their vote independently on 2 or 3 separate screens, and if you only have one screen then you're back to a single point of trust.
Perhaps N+1 screens -- one shared input screen that works like normal and feeds all of the N independent voting machines, and then each machine would have a display to confirm how they recorded the vote. Sort of like a vote receipt but a) doesn't waste paper and b) doesn't leave a physical receipt that coercive agents could use to verify your vote. A final confirm/reenter prompt and off you go.
All N machines would have to be compromised (and in exactly the same) way in order for vote count fraud to not be detected -- that is you'd be able to spread your trust across 2 or 3 manufacturers instead of a single one.
Its definitely not easy at all:
1) You have to trust that because the system says it counted your vote, that it actually did. OK so that's not TOO tough -- if a full list of voting IDs is posted, you can just count the rows and compare to the total votes.
2) Here's the hard part though: You need to verify that your vote counted for the candidate you selected. That means you have to have something in your possession that can be matched against the posted table. This could be as simple as just directly listing the candidate chosen in the table or it could also be obscured, but even in the latter case your boss could coerce whatever access/verification code you received just as easily as he initially coerced you.
3) And finally, you have to trust that the posted list reflects the actual recorded votes. If the list plainly states the candidates chosen that this is as trivial as #1, but if they're obscured (and I'm assuming they'd be obscured in some non-trivial way,) then each person could verify their own vote but would have no way of counting up the number of votes per candidate. So the posted list could reflect the true votes of every person, and the total votes would still have to match, but there would be no way to verify exactly how many votes that each candidate got -- X+Y = (X+Z)+(Y-Z) for any Z you choose.
There's probably some cryptographic way of solving that as well, but "easy" is not an adjective I would necessarily select -- and none of that stops your boss from coercing your vote (and verifying it afterward if necessary, at least assuming you yourself have the possibility of verifying it.)
And how do you guarantee that the manufacturer hasn't added the odd extra circuit that wasn't in the base design? I guess if 3D printing gets advanced enough that you can print your own electronics.. but then you still have the compiler backdoor issue https://en.wikipedia.org/wiki/Backdoor_(computing)#Compiler_backdoors if the printer was built to recognize certain designs and modify them as it was printing. So now you have to build your own 3D printer (and it has to be advanced enough to print electronics.) And then whatever you used to build that has to be analyzed and so on until you're basically starting with a warehouse full of raw materials and thousands or even tens of thousands of hours ahead of you that you can't contract out because you can't trust other people to not build in back doors either. And THEN you have to somehow guarantee that you did everything perfectly and didn't leave any accidental flaws or the whole exercise was for naught.
Obviously those attacks are getting rather complex at that point, and its certainly a lot easier to trust a self-printed circuit created from an open design than it is to trust Windows or any other corporately-controlled system. But at some level, even going to all that trouble still leaves theoretical security risks.
And finally you then use your perfectly secured machine to check out Slashdot for the first time in 4 years after building your own browser because you don't really trust Firefox either, and simply sending and receiving the minimally necessary packets is supplying every router along the way with at least a bit metadata about your location and actions.
At some point, you'll eventually have to accept that risk is a thing and manage it rather than trying to eliminate it, because the latter is very close to impossible. The only true way to guarantee that your computer isn't sending out some sort of information is to turn it off, unplug it, smash it, and move its remains somewhere far away from you just in case its got some embedded RF chip or whatever that you didn't manage to smash.
It happily loaded the telemetry crap on my system with Experience not installed. As the article (and even TFS) say.
Detailed hardware info could be a potential privacy risk, though a very low one.
A list of games you own is likely very personalized. Everyone has their own preferences not to mention which specific games are installed at any one time out of a possibly large digital library -- even if nVidia can't read the full Steam or GoG or Origin or whatever libraries (and they probably can't -- at least not yet,) the subset of installed games at any one time could theoretically add a temporal component to an already quite personalized data set, even if its a relatively granular measurement, without having to send any additional information such as last login timestamps or whatever.
Its not. I've got an nVidia card and I'd found the tasks before, but I'd assumed they were associated with geForce Experience (which I removed as soon as I discovered that they force a login even to just run the driver update -- which is the only reason I ever cared about that app in the first place.. I don't use shadowplay or game profiles or any of that other crap.)
I ended up having to use the DDU since the driver uninstaller crashed repeatedly, essentially giving me a full clean reinstall (sans Experience) when I was done.
And yeah after seeing the article I checked again and sure enough, the telemetry tasks were back.
Not sure about the people who claim to not see them.. Maybe there's some sort of A/B trial going on, or perhaps it was only in one specific version and they've already backed off on it, or maybe it only gets applied in specific hardware setups or who knows what else. But it definitely is (or hopefully by now.. was) a thing.
Unfortunately, unless there's enough backlash to make nVidia change their mind.. there's probably at most 2 years before AMD and Intel and every other video card manufacturer (are there still others?) hop on the "me too" bandwagon since turning down an essentially free income source is rather un-American. And any sort of user tracking is a potential income source these days.
There's a huge difference between a product like cheap knock-off toys and an information service like Google.
One is personalized, the other isn't. I'm sure the Chinese would have absolutely no problem buying American products like clothing and DVDs.. except they have no reason to do so since they make all that shit themselves at a fraction of the cost of US producers (and then sell it to us since much of it is produced under license from US or at least western-based companies because again.. fraction of the cost.)
Similarly, if we actually cared about Chinese information services, we probably would start seeing limitations put on them. But like the Chinese with physical products, the US has little need for outside information services since we still produce the (subjectively) best services on the planet, with some input from Europe and Japan here and there.
There's another angle here as well -- the rules that China's making apply equally to their national businesses, so its not an explicit barrier to international trade -- its a barrier to business in general. Now the fact that Chinese companies were probably already doing things like that while foreign companies typically wouldn't have been is another question but its not a question of trade policy as such since the rules are the same on both sides in this case and "I'm not used to that rule" isn't a solid argument.
Imagine if we allowed China to ignore lead safety rules in their products just because they aren't used to having to design things that don't kill people. Obviously that sounds ridiculous but it would be a (somewhat) similar "barrier" to Chinese imports.
no "real name" requirement
No, just enough metadata hanging around to make the requirement trivial anyway. At least with an explicit real name requirement, the user knows they're probably being tracked and spied on. Not that that's much solace.
no "prohibited" topics
No, just topics that will get you sent to gitmo, or sued out of existence (though that's more done by shady companies than the government.) Still, not technically prohibited.
data localization is not required
True, though this is by far the most benign issue on the list, and in some ways non-localized data is a risk as well (if national or even just corporate secrets happen to be stored on a non-local server, its entirely possible for the foreign government to just confiscate the server.) Basically, would you want your emails and data stored in China? Then why would you think the Chinese want theirs stored in the US?
America's monitoring is designed to catch bad guys, not to suppress discussion or prevent the flow of information
Except when they decide simply having/sharing information makes you a "bad guy."
Now certainly China is going to be worse than the US and most other western nations in basically all of those categories, but please don't mistake that for the assumption that the US doesn't participate in them at all -- America does many similar if not the same things, just at a smaller (and often more subtle) scale.
And that's just the government. Many of the largest digital companies in the US are well-known for censoring topics they don't like, and several have attempted to enforce real-name policies. Given how much emphasis the US in particular puts on corporations, that's still a fairly large issue even if its not as severe as the government performing those actions (companies at least don't have the power to imprison you, legitimately or otherwise.)
I don't really know anything about Ataturk, but just in general converting to a democracy is actually kind of hard. Telling someone that what they've known their entire life is wrong or bad can be pretty jarring, especially when their politics and religion is all mashed together to the point that democracy could almost appear to be denying God's will since many non-secular dictators tend to suggest that their power is granted to them by God.
And even if you switch to a democracy, you immediately need to guard against is voting in a prime minister (or president, depending on your setup) who just grabs all power for himself and decides that the next election will be "some day," effectively leaving you with a new dictatorship and a coat of paint over it. There's a lot of "democracies" in the world even today that fit this model.
The only real way to prevent this, at least it seems, is to have a benevolent dictator open up the lower ranks of government to open vote but still retain the ultimate power to override anything stupid that gets done, until such time as both the politicians and the people realize what their roles are in the new society and then the dictator can vacate.
That's no easy task either. Benevolent dictators are few and far between at the best of times, and we're talking a scale of a couple generations give or take -- which is only like 5 or 10 election cycles under the US' 4 year cycle length. Even if you have a benevolent dictator at the start, they've got to either survive that long or be able to hand the reigns over to someone who shares the same mentality.
The US itself gets to be a bit of a special case because they sort of had a "benevolent" dictator in the form of the British government (benevolence in that case mostly being due to sheer distance and inability to promptly squash troublemakers.) So as a country they appear to have started as a straight up democracy but historically its mostly just because nobody called them a separate country during the lead-up period.
Well yes that would be majorly detrimental to my boss. I on the other hand would benefit greatly from the resulting payout.
Depends how well that interview went. And whether your boss thought you were worth increasing your pay or just lets you go for being disloyal. Most people don't tell their boss that they're looking for new employment until they're already fairly certain they've landed something for a reason.. or unless they're basically just bluffing in order to get a raise.
And here's a great scenario that is countered by evidence, given how the ability to track your phone location accurately already exists, as does advertising.
Yes, but so far those things aren't linked (unless you explicitly download McDonald's app or something.) I'm talking about a world where your tracking information is just publicly available (for the narrower definition of "public" that means "large corporations," which seems to be the definition that those kind of things tend to be targeted towards.)
Their problem is they didn't act on anything until after it was too late. They knew exactly where he went and when and didn't think much of it until later.
My guess is "too late" was right around the time he decided to start doing something that would cause the NSA to want to pay attention. Who cares how many times he went grocery shopping in the year prior to leaking the documents? They want to know where he is _now_. And they would have especially loved to know where he was prior to his being granted asylum (ie: if anything happened to him today that could be tracked to US interference, there would be a bit of a political stink raised I'd guess, as that would somewhat undermine Russian security.)
wondeful doomsday scenarios that just simply don't pan out
Yet. It only takes apathy to give up your freedoms. It usually takes a revolution to get freedoms back. Are you willing to sit back and do nothing just hoping that these doomsday scenarios will continue to not pan out?
Trump obviously. Hilary does as much as she can to not talk about the things people want to hear -- because most of what we want to hear about her is details of her numerous scandals.
Trump lies to your face. Hilary schemes behind your back. Hard to say which is worse. Great time for US politics.
That's not always as easy as it sounds. If there was data conversions involved for example, the previous stable build may not even run anymore and would require restoring everything from backup, which may well be a many-many-hour project in itself -- and possibly taking time away from fixing the issue if it was a small-to-mid size company that recycles people into multiple roles (and programmer/IT services is a frequent combination at the best of times.) Just in time to turn around and have to re-convert as soon as you're done because the fix has been completed.
Never mind the fun of the programmers telling you "it'll just be another 2 hours" for 18 hours straight because issues in software tend to branch out in ways that nobody thinks about/remembers and can't include in their estimates until their nose is already in the code and its looking them in the face.
That's understandable (to a degree) in the situation where there was a set schedule ahead of time and things ran over. Then management has to make a decision whether its going to be a bigger hit to their reputation to delay vs releasing garbage.
What we often see though, especially in direct B2B-type software where there's a more intimate relationship between vendor and customer, is that the conversation goes more like this:
Manager: "We want something that does X"
Engineer: "OK that will take 6 months"
Manager: "Alright we'll call it a year to be safe"
Sales: "We sold one! Needs to be installed by end of next month"
Manager & Engineer: "#@#%$#@"
And that's with a good manager. Bad managers will give you 3 months when you say 6.. which doesn't really matter in these scenarios because sales has still only given you less than 2.
Jamming 2 weeks of work into 1 week is going to result in cut corners no matter what methodology you're using (or even what line of business you're in, for that matter.)
If you switch to a methodology where you're estimating in 6 month blocks and you're off by 100% like that, you're now 6 months off schedule instead of one week off -- that's even worse!
Not to say agile isn't misimplemented regularly, but if you're not schedules are off by that much of a margin, you need to start by looking at how you're generating time estimates before you bother changing your entire methodology.
Heh. The sad thing is, the telemarketer probably has a better time tracking you than the cops do. At least in the US. John Oliver did a segment on that a few months ago (https://www.youtube.com/watch?v=A-XlyB_QQYs.)
You can argue whether or not Uber is a "taxi" and thus subject to taxi regulations but you can't really argue that a flying car isn't an aircraft. And the FAA regulates _all_ air traffic from toy drones right up to giant commercial jet liners and everything in between.
That would be more like if Uber claimed their drivers weren't subject to normal road laws like traffic signs. The taxi thing would be more like if Southwest Airlines started complaining about the new service, which is unlikely as I doubt anything Uber comes up with will be in a competitive situation with traditional commercial airlines.
It would certainly be detrimental if your boss decided to check up on you on your day off and discovered you were at a competitor's office -- kind of suggests you're looking at other employment.
Or something less drastic: If McDonald's notices you're close to a Burger King and suddenly you get 14 text messages with deals for Big Macs. Perhaps not "incredibly" detrimental but certainly annoying as hell, especially if you happened to just be sitting at a stop light and had no intention of going into Burger King in the first place.
And that's not even starting to get into the really serious issues. Imagine if the US Govt had the capability of tracking Snowden back in 2013.. or even today. I'm sure it wouldn't be too hard to sneak an operative into Russia.
Its not just location -- privacy in general is granted as a right in the US constitution (as well as many many other countries to some extent or another) not because every person requires it 100% of the time, but because many people require it _sometimes_ and you never know when you will become that person for whatever reason.
Most constitutional rights are like that. The right to free speech is kind of meaningless for the vast majority of most peoples' lives because very few of us say anything that's super controversial, and even more rarely do we say it to someone who cares enough to persecute us. But those few people who truly need freedom of speech benefit us all by being able to say what they need to say without (as much) fear.
The trouble is always the carrot.. well that and poorly designed interfaces.
Eventually someone will invent something that a significant number of people "must" have. And then your browser will give you a single all-encompassing "allow this site to access your bluetooth devices?"
And even though all you really wanted was to allow FB to upload images to your bluetooth-enabled digital picture frame, suddenly FB (and all of their apps and partners and whoever else) also has access to your mouse and your gamepad and your webcam and every other device you own that happens to have a bluetooth interface.
I mean there's no question that you need "simple" interfaces for the technically illiterate, but unfortunately that seems more often than not to end up being insanely broad and almost never do they even have an "advanced" mode for people who know what they're doing to have more control.