I'm impressed by the HGST drives, less than 1% failure rate. I haven't touched the Deskstar line of drives since the IBM Deathstar debacle, but I think it's time to take a second look. Hopefully they have not switched over to Western Digital's technology.
Well, HGST drives are still more expensive than Seagate or WD drives of similar capacity.
Remember a hard drive is a very high precision mechanical device that has traditional economic pressures applied to them - everyone wants more for less dollars. So the high precision equipment gets compromised in the name of lower costs.
HGST drives cost more, so presumably they didn't try to eeke out every penny of savings out of it and can still use higher quality parts and manufacturing techniques.
Of course, the best indicator I've found is warranty length - avoid drives that have warranties of under 5 years and you'll probably hit the ones with the lowest failure rates. Manufacturers don't like to do warranty replacements - it costs a lot of time and effort to do, so if they're confident enough to give it a 5 year warranty, chances are it will last all 5 years without a problem. But economic pressures have made cheaper drives with 2-3 year warranties and those generally mean they're not only saving on the warranty, but they compromised the mechanism to achieve lower costs and the cheaper mechanism really is only good for the warranty length.
I still don't see why they couldn't remove the NAND chips and dump their contents, then do an offline brute force attack. I figure that if this is a terrorism case, then NSA could throw their most powerful compute clusters at it.
Since the iPhone 4, the NAND memory has been encrypted. With a key unavailable to software.
It's why a complete phone wipe on iPhone 3GS and prior took several hours, while only taking seconds on an iPhone 4 and up.
So dumping the NAND does absolutely nothing - the key used to encrypt it is hidden inside the SoC itself an inaccessible to software. So you can't pop the NAND off one iPhone and put it in another iPhone.
Android's started encrypting the flash as well, but it's still an optional feature.
Heck, you can have main memory encryption as well - so the data in main memory can't be accessed as well. In this case, it's usually a per-startup key - so every bootup uses a completely different key.
And the iPhone 5c is the last phone where the authentication is done in software. Since the A7 SoC upwards, the secure enclave is what authenticates the PIN code, and forces a wipe of memory if you fail to authenticate after 10 tries.
The problem for Apple is not creating the special firmware - that's easy. The hard part is how to install it without disturbing the data. Right now, to install a software update, you have to have an unlocked phone. Even a DFU update wipes out the user data.
The fundamental parts of the engine are all mechanical. They work without a battery.
Resilience to electrical failure is important.
The sparkplug of a gas engine requires... electricity.
A modern car engine uses an ECU which regulates spark timing and the transmission (usually called a PCU or Powertrain Control Unit nowadays) - it adjusts the spark timing and spark power based on the load of the engine. Lose battery power and the ECU goes dead. Depending on the vehicle, if you drop the battery, it may or may not continue running - the alternator will produce more than sufficient power to keep the engine running, but the battery provides voltage regulation of the entire system.
And there are still completely mechanically driven engines - small aircraft use them, and they're a PITA to manage because you have to manually adjust the mixture (fuel-air ratio) for optimum power as you change altitudes. Experimental avgas aircraft, and production diesels (running on Jet-A) use a FADEC (Full Authority Digital Engine Control) which runs off the ship battery and a backup battery that fully controls the engine. The pilot only has a lever that tells the FADECs (there are two of them for redundancy) how much power to develop - the FADEC figures out the optimum settings to achieve that. You get an increase in efficiency, a decrease in pilot workload and all around increases in efficiency.
Heck, Electronic Fuel Injection isn't on aircraft engines yet - yes, they've had fuel injection for around 25 years or so but it's generally of the continuous spray type.
As for this, it does have some advantages like extreme variable valve timing. Hybrid cars, for example, often use a modified Atkinson cycle engine (modified because it's really an Otto cycle engine, with the intake valve kept open well into the compression stroke to reduce the fuel charge). Atkinson engines are extremely efficient - they have a small intake and compression stroke but a large power stroke (the goal is at the end of the power stroke to have 0 differential between cylinder pressure and air pressure, thus ensuring you have extracted all the energy).. But at the same time, Atkinson engines don't develop as much power. Being able to switch operating modes on the fly can be useful in pure gas-only cars - switch to Atkinson during low power for maximum efficiency (idling, highway), while being able to switch to Otto when you need power (accelerating, for example)
I'm fairly sure that I read somewhere quite a while back that Tor was already broken by one or other of the organs of the US government, and some people doing something illegal via Tor got caught and prosecuted. No?
Well, it's not Tor itself that's the problem it's poor OpSec that was the issue causing the identity of the site owner to leak out. And there's another one involving an Apache module that is configured to listen to requests from localhost by default, except that Tor dark sites do exactly that so it went from a proper configuration to a vulnerability.
And reportedly, the NSA owns a LOT of exit nodes to which they use to monitor traffic. But that's not unexpected - Tor exit node traffic is easily monitored (you can't even trust SSL) by the exit node and poor OpSec again will lead to Tor users being identified.
In short, get everyone to use Tor and they'll be easily identifiable as they start using Facebook, social networking, as well as e-commerce and everything else.
You can only be anonymous by also being anonymous.
Isn't that the case anyway? 99% of games that come out nowadays seem to be based on things like Unreal, Unity, or the big publishers own in house engines like Frostbite.
The amount of people actually doing low level stuff seems to have diminished rapidly over the last decade as engines have become more flexible and it's really just turned into a battle over who has the best toolset and content pipeline now.
So even the big engineering teams don't seem to be expending much effort into engine development - publishers like Ubisoft and EA seem to have many tens of development teams and yet only seem to be using a few different engines across all those teams - certainly the days of every team building their game up from scratch engine and all are long long gone.
Well, using an engine means you get a lot of stuff out of it - firstly, not having to worry about the low-level underlying technologies. Your artists shouldn't need to worry about whether it's OpenGL or DirectX or Metal or Vulkan - that stuff should be abstracted away from them and the toolset should take care of it. Likewise, your modellers don't really care either.
Most of the game is really in asset generation - the textures, models, sounds, music, videos, artwork, scripts, etc.
And using a prebuilt engine has other benefits - porting gets to be much simpler - use Unity and you get iOS, Android, PC, Mac, PS3/PS4, Xbox360/Xbone and other platforms with it. Sure you have to do some work, but most of the hard part has been done for you.
Heck, modern engines can even enable/disable features as necessary - if you have DirectX 12 supported video, it will use those enhanced features, else it will downgrade to DirectX 11 or other API with lowered quality.
Well, the "Allow non-market apps" checkbox is probably checked if the user uses Amazon or Humble Bundle apps, which require sideloading.
And rooting may be done by a user who finds they need to do it in order to install some APK they found on the web. Perhaps to avoid paying the 99 cents on the Play Store so they downloaded it elsewhere for free.
As for clicking the link to the APK and downloading/installing, it's trivial to do. There are categories of apps you can say the APK does that will get people to install it: * Watch/view free porn * Download apps for free * Watch TV and movies for free * Watch premium TV (HBO, etc) for free * Download and watch YouTube videos offline * Watch premium content for free * [Latest hot game] download and play it for free etc. etc. etc.
As a general lot, Android users are cheapskates - this from many developers who realize the only way to make money off Android is to put ads in apps because users will not pay for stuff. So telling them they can get the latest paid stuff for free gets you probably a good 30-50% of the population to install it.
It shouldn't even matter! The CPU should be doing the authentication anyway, with the sensor simply sending the bitmap (or whatever) to it. Having the sensor be a "trusted" part of the authentication system is just as stupid as requiring a "trusted" keyboard for putting in passwords would be.*
That IS what is happening.
But the CPU and sensor are paired up because you don't want to send the sensor data unencrypted across the bus where it's then subject to spoofing attacks. It may seem silly, but it's already been proven on Android phones where a good majority of the sensors do NOT protect the sensor data they send the CPU.
The CPU gets this data and decrypts it. However, to prevent access from user-level software or even kernel level (via privilege escalation techniques - the kernel is just an untrustworthy), the CPU enters a special trusted secure mode which is completely inaccessible to the kernel and userland software. Here your image data is processed, analyzed and a final determination done when the data is compared against the secure memory storage area (secure enclave - which because it is only accessible in secure mode is completely inaccessible to normal software).
The problem happens when you replace the sensor which breaks the pairing and encryption keys. Now you have to decide what to do.
A basic software engineer will say "we'll just re-pair the sensors". Which is great, until you realize you just created a security hole - what if what you just attached wasn't a sensor, but something more sophisticated? Perhaps it's something that pretends it's a sensor, but is really an attack device.
Said attack device can try to feed specially doctored bitmaps to the secure enclave and do power monitoring and other things to try to divulge secret encryption keys used to access main storage or other things. Or perhaps feed in invalid images meant to crash the CPU in secure mode in such a way as to be able to run arbitrary code.
Since this mode is superior to kernel mode, it will be completely invisible to the main OS and can spy on everything (think Intel Management Engine, or System Management Mode (SMM) on x86 - the software runs independently of the OS).
So re-pairing the sensor is a bad idea unless you're in a controlled situation.
Instead, Apple aborts the complete OS with error 53 - the sensor pairing data is mismatched, and the system is no longer trustable. To protect user data, it would be preferable to simply erase the encryption keys so user data cannot be compromised (think of it this way - the people who can carry out the attack would likely be state actors). Because while 99.999% of the time, the sensor will just be another sensor, who's to tell it isn't a sensor designed to hack the system and spy on its user with the ultimate spyware?
This is one of those security balances that has to be worked out - do you try to protect user data against state sponsored attacks that have been proven to occur, or do you try to give the user the ability to fix it, at the risk of completely compromising your security?
Apple chose the former - if the sensor isn't trustable, then the secure enclave is no longer trustable - malware could easily be running and private user data could be sniffed and uploaded for later analysis. So instead, when Apple detects the phone's software may have been compromised, they shut down with error 53.
Once the secure enclave is compromised, all bets are off. And Apple cannot tell if the TouchID sensor was replaced because the user changed it, or if was changed because the NSA needed to spy.
Modern cars will last as long as these did. I say "did" rather than "have" because even as TFA notes, if you were to look under the hood you probably wouldn't even find original parts in them anymore. That is, other than the body, these practically are no longer the same cars that they once were in the 50's.
The reason why modern cars don't seem to last as long in first world countries is because once they break down to a certain point, the labor cost is so expensive that it's cheaper to just get a new car. However in Cuba, the labor price is typically lower while at the same time it's harder to get a hold of new stuff, which means reusing stuff becomes more practical than just making a new one. Like for example, TFA mentions repurposing old dryer motors for key cutting machines.
Even body parts are pretty much fabricated from rare sheet steel.
Discovery already has Cuban Chrome which showcases how Cubans have been keeping these old cars running - in fact, there is an auto club where if you meet certain requirements (mostly aesthetic and emissions), you can act as a taxi for foreigners. Else, you can only drive locals - and of course, tourists pay more.
And any engine will do - they actually show a car running a boat engine because that's all that was available.
When you see the cars from the 50s in Cuba, they're not like the cars from the 50s in North America - the Cuban cars are pretty much body only - everything else has been replaced or cobbled together with what parts were available. While North American restorations typically keep everything together with original motor and everything. The exceptions would be for the "hot rod" modifications which put in a modern engine and it becomes a new powertrain on an existing chassis.
Fashion is fickle. If augmented reality takes off, future fashion might dictate that eyewear is now the new cool/sexy, especially if it's built into nice-looking headgear that also functions as sunglasses when outdoors - lots of people don't mind wearing those. Agreed that it's a very big if, though.
Fashion is too fickle, to be honest, because only a few years ago, people were getting faux-glasses because it gave them the "geek chic" look that was desirable at the time. And by faux-glasses, I mean glasses where the lenses were basically panes of glass - there was absolutely no refractive power in them (because the people had normal vision to begin with).
And it's being adopted in safety equipment - if you're already wearing a hard hat or safety glasses/goggles as part of the jobs requirement, there's your display gear.
Well, yeah. That's a given. The question is: how did that became a factor in skewing the industry so bad as to squeeze the female workforce out?
I think this is greatly cultural. I see a higher proportion of women going into STEM (including software and CS) in countries like India and China than in the West. So there is a cultural factor at play, and it is one worth discussing (hopefully without devolving into misogyny and faux man-rights.)
Probably a late 80s thing, to be sure, because even at Atari, there were significant female population creating video games, and there were many females in the history of computer science as well.
I say 80s because that's when Nintendo came out, after the crash. They did one clever thing to get their NES on store shelves, and it may have had unintended consequences.
First, you have to remember the video game crash of the early 80s - it got to the point where retailers were shying away from anything videogame-related. So how does a company like Nintendo get their videogame machine in stores where retailers refuse to stock videogames?
Easy - you sell it as a toy that kids play with. But here's the rub - toy stores were (and generally still are) separated by gender - you have boy's toys on one set of aisles, and girl's toys on another set, and they will not mix. Nintendo now had a problem - is it a boy's toy or a girl's toy - it can only be one.
They chose boy.
This has very interesting ramifications - first, the Atari and other early console ads featured a whole family playing videogames - father, mother, daughter, son - all gathered around the TV and playing together. After this, Nintendo ads primarily featured boys - since that's how they decided to sell them. No more parents nor daughters - just boys gathering around playing.
Which may explain the perchant for people to regard videogames as what kids do, but not adults (because it was sold as a toy for boys, not the entire family), as well as regarding it as a male endeavour - again, Nintendo marketing as a boy's toy.
Other cultures didn't have this. Japan didn't have a videogame crash, and other countries didn't have to market exclusively to boys, so the whole videogame/computer association with boys never got made through marketing.
The ones that do are running the engine at optimal RPM to product power which will later be used to propel the car.
And the engines are usually running in the Atkinson cycle (on an Otto cycle engine), which gives less power, but greater efficiency. Basically in the Atkinson cycle, the intake stroke is shorter than the power stroke - the goal is that when you're at the end of the power stroke, the cylinder pressure is same as atmospheric. Otto cycle engines have the same intake stroke length compared to power stroke length, so energy is wasted since the gas could still do work.
These are generally modified Otto cycle engines rather than true Atkinson engines - what happens is the intake valve timing is adjusted so to remain open during part of the compression stroke so less fuel-air mix is actually in the cylinder.
As somebody who is very heavily opposed to communism and socialism, I'm interested in seeing the results of UBI (which is neither communism nor socialism, rather just a form of welfare) however I wouldn't want it anywhere I plan to live anytime soon, because it's one of those things where once you have it, it's practically impossible to take away, no matter what kinds of problems it creates or doesn't actually solve.
You do realize you're paying for other people's laziness/unemployability all the time, right?
Anytime any politician says "We will cut back benefits to the lazy and all that" it's really code for "We, the 1%, want to pay less taxes supporting your sorry asses". Because the homeless won't suddenly decide that because their government benefits are cut back, they will suddenly work. A lot of homeless ARE working - they don't make enough to pay for rent, food and other necessities. The ones that aren't, are generally unemployable - mental illness, drugs, alcohol or other problems keep them from holding even the most basic of jobs.
These people will not magically become productive members of society by cutting benefits. The working poor won't magically get better paying jobs (in fact, they'd likely lose their existing jobs), the mentally ill will not become employable (they need medical treatment, but they can't pay for that), etc.
Instead, those people will just become what the desperate do - steal/rob for the money and basic necessities they need. So instead of paying for their support through taxes, you're paying for them through increased crime, increased prices as stores have to cope with more shrinkage, increased security, etc. And yes, you can jail them - pay for more police officers, more courts, more jails. And medical bills - medical treatment inside the ER is the most expensive treatment option available - and the only one available, so you pay through increased insurance premiums helping people who can't pay at all for some of the most expensive medical treatment available.
Yes, it's all well and good to "be responsible" and "take care of yourself" and all that, but there will always be a segment of society that can't or won't. And it's either pay through taxes to take care of them, or abandon them and pay through other societal costs.
The only real question is - will basic income be done right? Because the basic income gets you basically very little - accomodations inside a large barracks-style room (you get a locker for your stuff, but that's it - you sleep with 8/16 other people at night), shared washroom facilities and entertainment, and 3 basic nutritionally complete meals a day.
That's what it really pays for - the absolute basics with very little discretionary money left over. Enough to live on, and for a few people, positively all they really need. If you want more spending money so you can live in more private accommodations, eat better food, have cash to pursue a hobby or anything else, then you need to do something to make extra money.
Sure, a few people will be happy to lounge around with the basics - that's fine. But a lot more people will want to improve their lot in life - I mean barracks style living only appeals so much.
Really, what happens is basic income transforms work from a necessity of survival (you can count having to steal or mug people as an occupation those without jobs have to do to survive), to a means to improve yourself. Eventually people want to have a private bathroom in their living arrangements, or private laundry, or watch more than just OTA TV, etc. So you work as much as you need to feel satisfied.
And it's sort of essential as robots and everything takes over - because those people whose jobs have been displaced aren't going to be able to find new replacement jobs no matter how much extra training they do.
I don't know exactly why Google wants them. Presumably, as a corpus to improve their image processing technologies.
And as a way to improve their facial recognition software, because you'll tag people and then Google will be able to identify the same individuals in other photos. Heck, one feature of Google Glass was to have it upload the photos and Google recognizes everyone on the street. The only way this can happen is if Google has a large corpus of faces so they can identify people in every photo.
FWIW, forgot to mention, the typical reaction time we have is around one minute. Usually too little to have everybody evacuated from premises... But we have history on this regard, and it has proven a very important development.
One minute is not a lot of time, but if you get a warning, you can do a lot of things in that minute that will save lives and reduce the disaster.
For example - gas pipelines can be depressurized and isolation valves closed, so any fires that start from the gas is limited to a very small amount of gas.
Elevators can be set to stop at the next available floor so passengers can get out. (one minute is more than enough for an elevator to go to a floor and evacuate its passengers).
Electricity can be shut off as well to limit the amount of fires that get sparked. Emergency power off functions may be activated in data centers to save critical data and perform quick power off to reduce fire risk.
Trains and public transit can be slowed down or stopped - depending on the system, it may not be possible to evacuate the trains in time, but once stopped, there's a much lower chance of derailing
And humans can go and hide in the strongest part of their building for safety.
One minute is not a lot of time, but a lot can be done to minimize the amount of risk and even reduce post-earthquake fires.
"Passed" isn't quite the right word. These prices are something Saudi Arabia is doing on purpose to try to run all the US shale oil producers out of business. If their plan works (Mwaahahahah! Good kitty), presumably they will then be able to go right back up to the higher prices they were selling Oil at 5 years ago.
You missed right after they buy up all the bankrupt shale gas producers.
The middle east is running out of oil. Countries like Dubai are investing heavily in alternative markets to deal with the impending end of black gold - Dubai is basically trying to be a high end tourist resort, for example.
Saudi Arabia is trying to do it another way - buy up the next set of oil producers so the profits can be taken that way - bankrupt the existing oil producers in North America, buy them all up, and then jack up the prices and let the profits flow into the country. So even when they run out of oil, they still own all the companies producing oil in other countries, thus ensuring prosperity of their own.
Plus, I think they also want to bankrupt Iran who because of the nuclear treaty can sell oil on the market again.
would like to suggest one better: If your phone is one that allows you to remove the batter (i.e. not an Apple or a OnePlus or a few others), just get a spare battery of the type that the phone takes. When your phone dies, reach into your pocket, pull out the spare battery, and switch it for the one that is in the phone. It's instant, efficient, and doesn't require you to juggle your phone plus another box for whatever length of time it takes your phone to charge./blockquote>
And how do you charge the spare battery? Do you expect people to plug in the phone, wait two hours for it to charge, then swap the batteries and charge it again?
And that's the problem with removable batteries. In the early days of AMPS phones, yes, the charging dock had spare slots for extra batteries so you can charge the phone's battery and one or two spares at the same time without having to swap them.
These days, the vast majority of phones only have the phone itself to charge the battery (some phones do have dock accessories that let you charge a battery outside the phone), so to charge spare batteries entails remembering to periodically check the phone, see the battery is full, then swap the battery and put the phone on charge again. Forget to do this and you'll find you have been carrying a spare battery that's dead.
With an external pack, they always can charge at the same time as the phone - just get another charger, and put both on charge together. No worrying about waiting for the phone to finish charging then swapping batteries - both will charge simultaneously and it'll be ready in the morning.
Plus, if you're carrying batteries in your pocket, you worry about them shorting out. External batteries are protected against that.at least by random keys and such you find in pockets.
I can understand why one would need to use that date as test data for an application, but why would anyone set their system date to that in the first place? (Not that I'm apologizing for Apple, that's a pretty stupid bug...)
Even better - why even let the user set the date to a time that far back? If you're going to ask for the current date and time, there should be no reason to be able to set it before the software release date. Maybe set it long in the future, but if you release the software in 2016, there will be no way the current date will be before 2016.
A bigger question is - why didn't they count frames?
We know there are two HD resolutions for OTA - 1080i60 (60 fields per second, or approximately 30 frames/sec), and 720p60 (60 frames per second). These two resolutions and framerates have the nice property that they have the same pixel clock and data rate, so you can choose between resolution and framerate - for traditional TV programming you often just want resolution, so you use 1080i, but for fast action, you want framerate, so you use 720p.
Now, it's easy to see what framerate the camera runs at - take a spot in the game, and count frames while an onscreen timer (like say, the game clock!) ticks away. Then you move to the part of the video in question, then count frames. Since it's less than a second, it would be less than 60 frames in total, so it should be possible to manually hand count how many frames elapsed between the disputed times and then you can compute how much time has elapsed.
This way can also be used to verify that the camera timer is ticking away at the proper rate.
Gigabit LTE means that you'll be able to use up your entire high speed data quota in less than a minute, unless the carriers finally update their data pricing models.
How is it that we've ended up with $10 for 10Gb or less of data now for about ten years? In the meantime, we've gone from inefficient EDGE to unbelievably efficient LTE, with HSPA+ available now for, what, the last five years on most GSM family networks?
Yet the data prices haven't budged. The carriers have more bandwidth than ever, more efficient ways of using it than ever, but they still think they're running ancient EDGE or cdma2000 networks.
Easy - profits.
Remember just a few years ago when people paid 25 cents per text? And some even paid another 25 cents to RECEIVE a text? Same reason - it was a massive profit center
Then texting stopped being a thing - with many ways to avoid it been iMessages and IM apps and Hangouts etc which used much cheaper data instead of SMS. Plus competition made it such that carriers started offering unlimited text plans for $20 extra. And of course, they realized they had a new profit center - data. Even better, they charge by the kilo and not kibi, and for good measure, they toss in the OTA headers as well in the byte count.
So yeah, they're charging because they can because it makes them massive amounts of money. On the bright side, they do adopt the new technologies quickly in an attempt to make you overuse your data plan and pay even more outrageous overage charges.
Stupid truckers routinely follow their GPS up Tail of the Dragon. They blindly drive right by the BIG YELLOW signs that basically say
"If you take your semi past this sign, you are an idiot, you will get stuck, please don't kill any motorcyclists with your stupidity."
They do get stuck way more often than that.
The solution is a truck specific GPS, which they do make What makes them special is they contain height information - before you start route planning, you enter in the height of your rig - the GPS will actually route with that information in mind - avoiding tunnels and routes where overpasses are too low to make it. (This may even entail taking an exit just to get back on the onramp).
The problem is, truck-specific GPSes are expensive and their map data even more so, so truckers often buy much cheaper car GPS units, or just use their phone's GPS system. None of which take height into account.
Of course, getting stuck and the subsequent tow, damage repair and other stuff suddenly makes the extra cost of a truck specific GPS a relative bargain.
Or Nissan Motor Company Ltd invokes the right for "Nissan" to be forgotten for just long enough to affect Nissan Computer's business to the extent that it goes out of business and/or agrees to sell the domain name to the automotive company.
That's not how Right to be Forgotten works.
Right to be Forgotten is an application of traditional meatspace record-keeping. Remember how in many places, after you've been convicted, after so many years that conviction is no longer on the books?
Or take your credit history - it only covers the past 7 years with older records, including things like bankruptcies and all that simply "forgotten" and expunged.
Right to be Forgotten is just like that.
First, it does not delete anything - it cannot. It merely breaks the link between a search term and the link it would point to.
Let's say you went to jail a decade ago and served your time and are completely free. You've lived a virtuous life since then, and a criminal record check would basically show you to be clean because your crime was wiped off the books. But a site archiving imprisonment records still lists you as being in prison. Right to be forgotten means you can break a link between your name and that site - you've done your time, and the state considers you to be clean and you can pass a criminal records check. But then an employer Googles you and sees that you were in jail. Is that fair? By law you did pass and such ancient history should be wiped. But the site showing the information has done no wrong either, so it would be bad to demand that they remove the information.
Or say you declared bankruptcy a decade ago. Since employers are doing credit history checks now, your bankruptcy is no longer shown to them for several years. But if they Google you, because some site archived news like that, you show up.
That's what right to be forgotten is all about. It only applies to individuals and only when the legal limit for such news has expired and is no longer relevant. So if you have a bankruptcy in the past 7 years, you can't invoke right to be forgotten to remove it off the internet - it's still relevant after all.
Or think of it another way - without right to be forgotten, a bunch of children are going to learn the hard way about the repercussions of their indiscretions. After all, in most jurisdictions, once you turn 18, your record is wiped and you start afresh - this includes any run-ins with the law (unless you were tried as an adult). Well, right to be forgotten lets you break all the links between the bad stuff you did as a adolescent teen and also start afresh
For example Doubleclick and those kinds of networks track me across the web even if I've never signed up for an account with them or otherwise accepted their ToS.
Are you sure? I mean, you probably did, probably for GMail or YouTube or some other Google thing. Or an Android phone.
I know Google loves to hide the fact that they own the majority of ad networks out there so everyone THINKS they only do the text ads, but no, Google owns the major ad networks like DoubleClick (they acquired them so many years ago it may even be when/. was "better").
You probably did accept DoubleClick and many other ToS by simply having and using a Google account (and didn't Google a few years ago unify their ToS across everything?)/.
There are a dozen use cases for not full headless and not full desktop. I'll name you one: a laboratory workstation that you both physically sit at and occasionally check up on from your desk or your home by sshing in and running a graphical thingie to monitor to test equipment it's plugged into.
Which works fine if your equipment supports multiple sharing sessions. If not, starting the new monitor may disrupt the existing process, screwing you over. Which is why X and remote desktop are NOT mutually exclusive - sometimes a view-only session is all you need to quickly view a setting without running something that could disrupt your long-running process.
The other reason is if you're on a flaky connection. Do this and X becomes a poor solution because the moment your connection burps, your applications are force-quit and you lose your work.
There are situations where one solution is better than the other. X forwarding is great, but it's not the be-all-end-all solution, especially if something you're doing is single-session only or you're not on a reliable connection (e.g., mobile) where you don't want all your programs to abort because you lost your cell signal.
It seems like advertising is backing away a bit, with the notable exception of the web. Ad-supported cable is dying but the no-ads premiums channels like HBO are doing well, and zero-ad subscription services like Netflix are cleaning up. The tech industry does seem to have more than it's share of advertising companies masquerading as something else. And the number of multi-billion dollar acquisitions for things like chat platforms, many that have subsequently been sold at a fraction of their purchase price, is suggestive of a bubble.
Ad supported cable is dying because a la carte is forcing what was once subscription based revenue (and thus concentrating on programming for a niche) is turning into ad-based revenue (and thus programming must attract eyeballs). So programming that could count on steady subscription revenue and concentrate on the topic at hand must now switch models and alter their presentation to go after what attracts eyeballs. This is a complete change and it's why ad supported cable channels added a bunch of "drama" and other things to formerly fact-based documentary programming. That drama (faked or scripted) attracts eyeballs. The more eyeballs, the higher the ad revenue.
Subscription based services like HBO and Netflix only care about growing subscription revenue, which means they don't care about eyeball quantity - they care about attracting subscribing eyeballs only. Their programming will be directed at what their subscribing public wants and what kind of subscribing public they want to get the dollars from.
So Netflix and HBO will be making programming aimed towards that demographic. You and I feel they're "winning" because we currently are in that demographic - the people who will likely see that programming and subscribe to the service.
But the market is still ripe for ads - the Superbowl for example, gets so many eyeballs its ratings are stratospheric. Which is why it costs over $100K per SECOND of ad time during it (that's $3M for an ad spot). For comparison, a prime-time 30 second ad spot generally commands $80-150K.
Sports, in general, are the highest rated programming on TV which is why they generally pre-empt other programming - the ad dollars spent is immense.
Slashdot Mirror
Well, HGST drives are still more expensive than Seagate or WD drives of similar capacity.
Remember a hard drive is a very high precision mechanical device that has traditional economic pressures applied to them - everyone wants more for less dollars. So the high precision equipment gets compromised in the name of lower costs.
HGST drives cost more, so presumably they didn't try to eeke out every penny of savings out of it and can still use higher quality parts and manufacturing techniques.
Of course, the best indicator I've found is warranty length - avoid drives that have warranties of under 5 years and you'll probably hit the ones with the lowest failure rates. Manufacturers don't like to do warranty replacements - it costs a lot of time and effort to do, so if they're confident enough to give it a 5 year warranty, chances are it will last all 5 years without a problem. But economic pressures have made cheaper drives with 2-3 year warranties and those generally mean they're not only saving on the warranty, but they compromised the mechanism to achieve lower costs and the cheaper mechanism really is only good for the warranty length.
Since the iPhone 4, the NAND memory has been encrypted. With a key unavailable to software.
It's why a complete phone wipe on iPhone 3GS and prior took several hours, while only taking seconds on an iPhone 4 and up.
So dumping the NAND does absolutely nothing - the key used to encrypt it is hidden inside the SoC itself an inaccessible to software. So you can't pop the NAND off one iPhone and put it in another iPhone.
Android's started encrypting the flash as well, but it's still an optional feature.
Heck, you can have main memory encryption as well - so the data in main memory can't be accessed as well. In this case, it's usually a per-startup key - so every bootup uses a completely different key.
And the iPhone 5c is the last phone where the authentication is done in software. Since the A7 SoC upwards, the secure enclave is what authenticates the PIN code, and forces a wipe of memory if you fail to authenticate after 10 tries.
The problem for Apple is not creating the special firmware - that's easy. The hard part is how to install it without disturbing the data. Right now, to install a software update, you have to have an unlocked phone. Even a DFU update wipes out the user data.
The sparkplug of a gas engine requires... electricity.
A modern car engine uses an ECU which regulates spark timing and the transmission (usually called a PCU or Powertrain Control Unit nowadays) - it adjusts the spark timing and spark power based on the load of the engine. Lose battery power and the ECU goes dead. Depending on the vehicle, if you drop the battery, it may or may not continue running - the alternator will produce more than sufficient power to keep the engine running, but the battery provides voltage regulation of the entire system.
And there are still completely mechanically driven engines - small aircraft use them, and they're a PITA to manage because you have to manually adjust the mixture (fuel-air ratio) for optimum power as you change altitudes. Experimental avgas aircraft, and production diesels (running on Jet-A) use a FADEC (Full Authority Digital Engine Control) which runs off the ship battery and a backup battery that fully controls the engine. The pilot only has a lever that tells the FADECs (there are two of them for redundancy) how much power to develop - the FADEC figures out the optimum settings to achieve that. You get an increase in efficiency, a decrease in pilot workload and all around increases in efficiency.
Heck, Electronic Fuel Injection isn't on aircraft engines yet - yes, they've had fuel injection for around 25 years or so but it's generally of the continuous spray type.
As for this, it does have some advantages like extreme variable valve timing. Hybrid cars, for example, often use a modified Atkinson cycle engine (modified because it's really an Otto cycle engine, with the intake valve kept open well into the compression stroke to reduce the fuel charge). Atkinson engines are extremely efficient - they have a small intake and compression stroke but a large power stroke (the goal is at the end of the power stroke to have 0 differential between cylinder pressure and air pressure, thus ensuring you have extracted all the energy).. But at the same time, Atkinson engines don't develop as much power. Being able to switch operating modes on the fly can be useful in pure gas-only cars - switch to Atkinson during low power for maximum efficiency (idling, highway), while being able to switch to Otto when you need power (accelerating, for example)
Well, it's not Tor itself that's the problem it's poor OpSec that was the issue causing the identity of the site owner to leak out. And there's another one involving an Apache module that is configured to listen to requests from localhost by default, except that Tor dark sites do exactly that so it went from a proper configuration to a vulnerability.
And reportedly, the NSA owns a LOT of exit nodes to which they use to monitor traffic. But that's not unexpected - Tor exit node traffic is easily monitored (you can't even trust SSL) by the exit node and poor OpSec again will lead to Tor users being identified.
In short, get everyone to use Tor and they'll be easily identifiable as they start using Facebook, social networking, as well as e-commerce and everything else.
You can only be anonymous by also being anonymous.
Well, using an engine means you get a lot of stuff out of it - firstly, not having to worry about the low-level underlying technologies. Your artists shouldn't need to worry about whether it's OpenGL or DirectX or Metal or Vulkan - that stuff should be abstracted away from them and the toolset should take care of it. Likewise, your modellers don't really care either.
Most of the game is really in asset generation - the textures, models, sounds, music, videos, artwork, scripts, etc.
And using a prebuilt engine has other benefits - porting gets to be much simpler - use Unity and you get iOS, Android, PC, Mac, PS3/PS4, Xbox360/Xbone and other platforms with it. Sure you have to do some work, but most of the hard part has been done for you.
Heck, modern engines can even enable/disable features as necessary - if you have DirectX 12 supported video, it will use those enhanced features, else it will downgrade to DirectX 11 or other API with lowered quality.
Well, the "Allow non-market apps" checkbox is probably checked if the user uses Amazon or Humble Bundle apps, which require sideloading.
And rooting may be done by a user who finds they need to do it in order to install some APK they found on the web. Perhaps to avoid paying the 99 cents on the Play Store so they downloaded it elsewhere for free.
As for clicking the link to the APK and downloading/installing, it's trivial to do. There are categories of apps you can say the APK does that will get people to install it:
* Watch/view free porn
* Download apps for free
* Watch TV and movies for free
* Watch premium TV (HBO, etc) for free
* Download and watch YouTube videos offline
* Watch premium content for free
* [Latest hot game] download and play it for free
etc. etc. etc.
As a general lot, Android users are cheapskates - this from many developers who realize the only way to make money off Android is to put ads in apps because users will not pay for stuff. So telling them they can get the latest paid stuff for free gets you probably a good 30-50% of the population to install it.
That IS what is happening.
But the CPU and sensor are paired up because you don't want to send the sensor data unencrypted across the bus where it's then subject to spoofing attacks. It may seem silly, but it's already been proven on Android phones where a good majority of the sensors do NOT protect the sensor data they send the CPU.
The CPU gets this data and decrypts it. However, to prevent access from user-level software or even kernel level (via privilege escalation techniques - the kernel is just an untrustworthy), the CPU enters a special trusted secure mode which is completely inaccessible to the kernel and userland software. Here your image data is processed, analyzed and a final determination done when the data is compared against the secure memory storage area (secure enclave - which because it is only accessible in secure mode is completely inaccessible to normal software).
The problem happens when you replace the sensor which breaks the pairing and encryption keys. Now you have to decide what to do.
A basic software engineer will say "we'll just re-pair the sensors". Which is great, until you realize you just created a security hole - what if what you just attached wasn't a sensor, but something more sophisticated? Perhaps it's something that pretends it's a sensor, but is really an attack device.
Said attack device can try to feed specially doctored bitmaps to the secure enclave and do power monitoring and other things to try to divulge secret encryption keys used to access main storage or other things. Or perhaps feed in invalid images meant to crash the CPU in secure mode in such a way as to be able to run arbitrary code.
Since this mode is superior to kernel mode, it will be completely invisible to the main OS and can spy on everything (think Intel Management Engine, or System Management Mode (SMM) on x86 - the software runs independently of the OS).
So re-pairing the sensor is a bad idea unless you're in a controlled situation.
Instead, Apple aborts the complete OS with error 53 - the sensor pairing data is mismatched, and the system is no longer trustable. To protect user data, it would be preferable to simply erase the encryption keys so user data cannot be compromised (think of it this way - the people who can carry out the attack would likely be state actors). Because while 99.999% of the time, the sensor will just be another sensor, who's to tell it isn't a sensor designed to hack the system and spy on its user with the ultimate spyware?
This is one of those security balances that has to be worked out - do you try to protect user data against state sponsored attacks that have been proven to occur, or do you try to give the user the ability to fix it, at the risk of completely compromising your security?
Apple chose the former - if the sensor isn't trustable, then the secure enclave is no longer trustable - malware could easily be running and private user data could be sniffed and uploaded for later analysis. So instead, when Apple detects the phone's software may have been compromised, they shut down with error 53.
Once the secure enclave is compromised, all bets are off. And Apple cannot tell if the TouchID sensor was replaced because the user changed it, or if was changed because the NSA needed to spy.
Even body parts are pretty much fabricated from rare sheet steel.
Discovery already has Cuban Chrome which showcases how Cubans have been keeping these old cars running - in fact, there is an auto club where if you meet certain requirements (mostly aesthetic and emissions), you can act as a taxi for foreigners. Else, you can only drive locals - and of course, tourists pay more.
And any engine will do - they actually show a car running a boat engine because that's all that was available.
When you see the cars from the 50s in Cuba, they're not like the cars from the 50s in North America - the Cuban cars are pretty much body only - everything else has been replaced or cobbled together with what parts were available. While North American restorations typically keep everything together with original motor and everything. The exceptions would be for the "hot rod" modifications which put in a modern engine and it becomes a new powertrain on an existing chassis.
Fashion is too fickle, to be honest, because only a few years ago, people were getting faux-glasses because it gave them the "geek chic" look that was desirable at the time. And by faux-glasses, I mean glasses where the lenses were basically panes of glass - there was absolutely no refractive power in them (because the people had normal vision to begin with).
And it's being adopted in safety equipment - if you're already wearing a hard hat or safety glasses/goggles as part of the jobs requirement, there's your display gear.
Probably a late 80s thing, to be sure, because even at Atari, there were significant female population creating video games, and there were many females in the history of computer science as well.
I say 80s because that's when Nintendo came out, after the crash. They did one clever thing to get their NES on store shelves, and it may have had unintended consequences.
First, you have to remember the video game crash of the early 80s - it got to the point where retailers were shying away from anything videogame-related. So how does a company like Nintendo get their videogame machine in stores where retailers refuse to stock videogames?
Easy - you sell it as a toy that kids play with. But here's the rub - toy stores were (and generally still are) separated by gender - you have boy's toys on one set of aisles, and girl's toys on another set, and they will not mix. Nintendo now had a problem - is it a boy's toy or a girl's toy - it can only be one.
They chose boy.
This has very interesting ramifications - first, the Atari and other early console ads featured a whole family playing videogames - father, mother, daughter, son - all gathered around the TV and playing together. After this, Nintendo ads primarily featured boys - since that's how they decided to sell them. No more parents nor daughters - just boys gathering around playing.
Which may explain the perchant for people to regard videogames as what kids do, but not adults (because it was sold as a toy for boys, not the entire family), as well as regarding it as a male endeavour - again, Nintendo marketing as a boy's toy.
Other cultures didn't have this. Japan didn't have a videogame crash, and other countries didn't have to market exclusively to boys, so the whole videogame/computer association with boys never got made through marketing.
And the engines are usually running in the Atkinson cycle (on an Otto cycle engine), which gives less power, but greater efficiency. Basically in the Atkinson cycle, the intake stroke is shorter than the power stroke - the goal is that when you're at the end of the power stroke, the cylinder pressure is same as atmospheric. Otto cycle engines have the same intake stroke length compared to power stroke length, so energy is wasted since the gas could still do work.
These are generally modified Otto cycle engines rather than true Atkinson engines - what happens is the intake valve timing is adjusted so to remain open during part of the compression stroke so less fuel-air mix is actually in the cylinder.
You do realize you're paying for other people's laziness/unemployability all the time, right?
Anytime any politician says "We will cut back benefits to the lazy and all that" it's really code for "We, the 1%, want to pay less taxes supporting your sorry asses". Because the homeless won't suddenly decide that because their government benefits are cut back, they will suddenly work. A lot of homeless ARE working - they don't make enough to pay for rent, food and other necessities. The ones that aren't, are generally unemployable - mental illness, drugs, alcohol or other problems keep them from holding even the most basic of jobs.
These people will not magically become productive members of society by cutting benefits. The working poor won't magically get better paying jobs (in fact, they'd likely lose their existing jobs), the mentally ill will not become employable (they need medical treatment, but they can't pay for that), etc.
Instead, those people will just become what the desperate do - steal/rob for the money and basic necessities they need. So instead of paying for their support through taxes, you're paying for them through increased crime, increased prices as stores have to cope with more shrinkage, increased security, etc. And yes, you can jail them - pay for more police officers, more courts, more jails. And medical bills - medical treatment inside the ER is the most expensive treatment option available - and the only one available, so you pay through increased insurance premiums helping people who can't pay at all for some of the most expensive medical treatment available.
Yes, it's all well and good to "be responsible" and "take care of yourself" and all that, but there will always be a segment of society that can't or won't. And it's either pay through taxes to take care of them, or abandon them and pay through other societal costs.
The only real question is - will basic income be done right? Because the basic income gets you basically very little - accomodations inside a large barracks-style room (you get a locker for your stuff, but that's it - you sleep with 8/16 other people at night), shared washroom facilities and entertainment, and 3 basic nutritionally complete meals a day.
That's what it really pays for - the absolute basics with very little discretionary money left over. Enough to live on, and for a few people, positively all they really need. If you want more spending money so you can live in more private accommodations, eat better food, have cash to pursue a hobby or anything else, then you need to do something to make extra money.
Sure, a few people will be happy to lounge around with the basics - that's fine. But a lot more people will want to improve their lot in life - I mean barracks style living only appeals so much.
Really, what happens is basic income transforms work from a necessity of survival (you can count having to steal or mug people as an occupation those without jobs have to do to survive), to a means to improve yourself. Eventually people want to have a private bathroom in their living arrangements, or private laundry, or watch more than just OTA TV, etc. So you work as much as you need to feel satisfied.
And it's sort of essential as robots and everything takes over - because those people whose jobs have been displaced aren't going to be able to find new replacement jobs no matter how much extra training they do.
And as a way to improve their facial recognition software, because you'll tag people and then Google will be able to identify the same individuals in other photos. Heck, one feature of Google Glass was to have it upload the photos and Google recognizes everyone on the street. The only way this can happen is if Google has a large corpus of faces so they can identify people in every photo.
One minute is not a lot of time, but if you get a warning, you can do a lot of things in that minute that will save lives and reduce the disaster.
For example - gas pipelines can be depressurized and isolation valves closed, so any fires that start from the gas is limited to a very small amount of gas.
Elevators can be set to stop at the next available floor so passengers can get out. (one minute is more than enough for an elevator to go to a floor and evacuate its passengers).
Electricity can be shut off as well to limit the amount of fires that get sparked. Emergency power off functions may be activated in data centers to save critical data and perform quick power off to reduce fire risk.
Trains and public transit can be slowed down or stopped - depending on the system, it may not be possible to evacuate the trains in time, but once stopped, there's a much lower chance of derailing
And humans can go and hide in the strongest part of their building for safety.
One minute is not a lot of time, but a lot can be done to minimize the amount of risk and even reduce post-earthquake fires.
You missed right after they buy up all the bankrupt shale gas producers.
The middle east is running out of oil. Countries like Dubai are investing heavily in alternative markets to deal with the impending end of black gold - Dubai is basically trying to be a high end tourist resort, for example.
Saudi Arabia is trying to do it another way - buy up the next set of oil producers so the profits can be taken that way - bankrupt the existing oil producers in North America, buy them all up, and then jack up the prices and let the profits flow into the country. So even when they run out of oil, they still own all the companies producing oil in other countries, thus ensuring prosperity of their own.
Plus, I think they also want to bankrupt Iran who because of the nuclear treaty can sell oil on the market again.
Even better - why even let the user set the date to a time that far back? If you're going to ask for the current date and time, there should be no reason to be able to set it before the software release date. Maybe set it long in the future, but if you release the software in 2016, there will be no way the current date will be before 2016.
A bigger question is - why didn't they count frames?
We know there are two HD resolutions for OTA - 1080i60 (60 fields per second, or approximately 30 frames/sec), and 720p60 (60 frames per second). These two resolutions and framerates have the nice property that they have the same pixel clock and data rate, so you can choose between resolution and framerate - for traditional TV programming you often just want resolution, so you use 1080i, but for fast action, you want framerate, so you use 720p.
Now, it's easy to see what framerate the camera runs at - take a spot in the game, and count frames while an onscreen timer (like say, the game clock!) ticks away. Then you move to the part of the video in question, then count frames. Since it's less than a second, it would be less than 60 frames in total, so it should be possible to manually hand count how many frames elapsed between the disputed times and then you can compute how much time has elapsed.
This way can also be used to verify that the camera timer is ticking away at the proper rate.
Easy - profits.
Remember just a few years ago when people paid 25 cents per text? And some even paid another 25 cents to RECEIVE a text? Same reason - it was a massive profit center
Then texting stopped being a thing - with many ways to avoid it been iMessages and IM apps and Hangouts etc which used much cheaper data instead of SMS. Plus competition made it such that carriers started offering unlimited text plans for $20 extra. And of course, they realized they had a new profit center - data. Even better, they charge by the kilo and not kibi, and for good measure, they toss in the OTA headers as well in the byte count.
So yeah, they're charging because they can because it makes them massive amounts of money. On the bright side, they do adopt the new technologies quickly in an attempt to make you overuse your data plan and pay even more outrageous overage charges.
They do get stuck way more often than that.
The solution is a truck specific GPS, which they do make What makes them special is they contain height information - before you start route planning, you enter in the height of your rig - the GPS will actually route with that information in mind - avoiding tunnels and routes where overpasses are too low to make it. (This may even entail taking an exit just to get back on the onramp).
The problem is, truck-specific GPSes are expensive and their map data even more so, so truckers often buy much cheaper car GPS units, or just use their phone's GPS system. None of which take height into account.
Of course, getting stuck and the subsequent tow, damage repair and other stuff suddenly makes the extra cost of a truck specific GPS a relative bargain.
That's not how Right to be Forgotten works.
Right to be Forgotten is an application of traditional meatspace record-keeping. Remember how in many places, after you've been convicted, after so many years that conviction is no longer on the books?
Or take your credit history - it only covers the past 7 years with older records, including things like bankruptcies and all that simply "forgotten" and expunged.
Right to be Forgotten is just like that.
First, it does not delete anything - it cannot. It merely breaks the link between a search term and the link it would point to.
Let's say you went to jail a decade ago and served your time and are completely free. You've lived a virtuous life since then, and a criminal record check would basically show you to be clean because your crime was wiped off the books. But a site archiving imprisonment records still lists you as being in prison. Right to be forgotten means you can break a link between your name and that site - you've done your time, and the state considers you to be clean and you can pass a criminal records check. But then an employer Googles you and sees that you were in jail. Is that fair? By law you did pass and such ancient history should be wiped. But the site showing the information has done no wrong either, so it would be bad to demand that they remove the information.
Or say you declared bankruptcy a decade ago. Since employers are doing credit history checks now, your bankruptcy is no longer shown to them for several years. But if they Google you, because some site archived news like that, you show up.
That's what right to be forgotten is all about. It only applies to individuals and only when the legal limit for such news has expired and is no longer relevant. So if you have a bankruptcy in the past 7 years, you can't invoke right to be forgotten to remove it off the internet - it's still relevant after all.
Or think of it another way - without right to be forgotten, a bunch of children are going to learn the hard way about the repercussions of their indiscretions. After all, in most jurisdictions, once you turn 18, your record is wiped and you start afresh - this includes any run-ins with the law (unless you were tried as an adult). Well, right to be forgotten lets you break all the links between the bad stuff you did as a adolescent teen and also start afresh
Actually, white-on-black makes things worse - it makes skinny fonts skinnier and the black "creeps into" the white and makes fonts appear smaller.
So much so if you're doing it, you must increase the size and weight of the font you're using to make it look "normal" again.
Are you sure? I mean, you probably did, probably for GMail or YouTube or some other Google thing. Or an Android phone.
I know Google loves to hide the fact that they own the majority of ad networks out there so everyone THINKS they only do the text ads, but no, Google owns the major ad networks like DoubleClick (they acquired them so many years ago it may even be when /. was "better").
You probably did accept DoubleClick and many other ToS by simply having and using a Google account (and didn't Google a few years ago unify their ToS across everything?)/.
Which works fine if your equipment supports multiple sharing sessions. If not, starting the new monitor may disrupt the existing process, screwing you over. Which is why X and remote desktop are NOT mutually exclusive - sometimes a view-only session is all you need to quickly view a setting without running something that could disrupt your long-running process.
The other reason is if you're on a flaky connection. Do this and X becomes a poor solution because the moment your connection burps, your applications are force-quit and you lose your work.
There are situations where one solution is better than the other. X forwarding is great, but it's not the be-all-end-all solution, especially if something you're doing is single-session only or you're not on a reliable connection (e.g., mobile) where you don't want all your programs to abort because you lost your cell signal.
Ad supported cable is dying because a la carte is forcing what was once subscription based revenue (and thus concentrating on programming for a niche) is turning into ad-based revenue (and thus programming must attract eyeballs). So programming that could count on steady subscription revenue and concentrate on the topic at hand must now switch models and alter their presentation to go after what attracts eyeballs. This is a complete change and it's why ad supported cable channels added a bunch of "drama" and other things to formerly fact-based documentary programming. That drama (faked or scripted) attracts eyeballs. The more eyeballs, the higher the ad revenue.
Subscription based services like HBO and Netflix only care about growing subscription revenue, which means they don't care about eyeball quantity - they care about attracting subscribing eyeballs only. Their programming will be directed at what their subscribing public wants and what kind of subscribing public they want to get the dollars from.
So Netflix and HBO will be making programming aimed towards that demographic. You and I feel they're "winning" because we currently are in that demographic - the people who will likely see that programming and subscribe to the service.
But the market is still ripe for ads - the Superbowl for example, gets so many eyeballs its ratings are stratospheric. Which is why it costs over $100K per SECOND of ad time during it (that's $3M for an ad spot). For comparison, a prime-time 30 second ad spot generally commands $80-150K.
Sports, in general, are the highest rated programming on TV which is why they generally pre-empt other programming - the ad dollars spent is immense.