a) the government can make me unlock it. b) Anyone with a rubber hose can make me unlock it. c) There is a non-zero inducement to cut off my finger. d) If someone figures out a way around it, you'll have a very hard time arguing it wasn't you who unlocked the device.
It's more like risky performance theater art.
And the reason to have a fingerprint sensor is the fact that a majority of smartphone users have no passcode or other security measure set. The reason is simple - a phone is accessed extremely often - I believe Apple quotes easily a thousand times or more a day. When you're using it that often (most of the time it's used for under 5 seconds), a passcode gets in the way - who wants to enter a passcode that many times? End result, they don't secure their phone at all.
The fingerprint sensor lets a user choose a passcode and access their phone without taking significantly longer (it would suck if it took longer to enter your passcode than you were going to use it). So while the OS is resuming from low power state the sensor can be reading the fingerprint and the secure enclave can determine if it's a valid fingerprint, so when the OS is ready the decision is ready.
Of course, Apple also has ways to ensure that the fingerprint is overridable - three bad reads triggers a passcode, a power cycle triggers a passcode and no unlocking for 48 hours also triggers a passcode request.
The law is currently unclear as to what happens when this is triggered.- if you are forced to unlock the phone by fingerprint, but the phone now requires the passcode...
It's not like a faster laptop would make the car run faster. In the end it only becomes an issue when there is a need to replace diagnostics equipment and there aren't enough spares.
I suspect a bit of hardware and software effort could port the interface and stack to an Arduino and then you could access it from a phone or tablet for another 20 years. But the pragmatic part of me wouldn't want to why something unless it's broken.
Well, that is the problem - the Compaq is the only laptop with a special slot that accepts the diagnostics interface card. That's the only reason it's there - the special card you need.
Depending on how proprietary that card is, it may or may not be possible to replace it - I think it was made before every car had an OBD-II port that there's interfaces for via USB, Bluetooth, WiFi and every other thing you have.
In the end, the solution might be to just replace all the ECUs with an OBD-II port. The only reason they haven't is the F1 is a pretty bespoke car, there aren't a lot of them on the streets so there's not that many of them to begin with, so any replacement is basically going to be one off development.
Of course, I don't think there's anything non-standard about the interface - a computer that age is probably using ISA on that port so it should be possible to convert it to a standard ISA bus card, then use a PLX bridge to give you a PCI(e) interface on top of that.
Freedom means the right to privacy. Solid encryption offers that privacy. It also gives criminals a way to hide their data from law enforcement. Long before the digital age, that has been going on in dark alleys and secret underground bunkers.
The government sees a way to gain unprecedented power and will stop at nothing to get it.
I find the trade off acceptable. I'd rather see a few more terrorists escape, than face a government that labels all who choose to encrypt a potential criminal, or worse.
Freedom also means eternal vigilance. There are people who are jealous of people who have freedom and will do anything in their power to destroy it. (We normally call these people "terrorists").
Having freedom means we don't try to oppress our freedoms to get rid of these bad people, but we live with them - it's the price of living in a free world.
We can use encryption, and so can they. Our law enforcement agencies need to take that into account and use more traditional forms of detective work to get at the bad guys. In fact, the NSA's data collection system is completely useless because they collect so much data, not only are they looking for needles in haystacks, but with all the data collection, the haystacks have ballooned in size, while the number of needles stayed the same. More data isn't better. In fact, the other terrorists have been caught by good old traditional detective work.
That's what's really needed - more boots on the ground.
Plus, you should probably mention the importance of encryption to DRM. I think everyone on Slashdot can get behind the idea that without strong, well encrypted, digital rights management systems, the studios would find their movies being quite literally stolen from them by people we can only describe as digital terrorists, their revenues and profits destroyed as unpaying freeloaders enjoy the fruits of the studios hard, expensive, work without paying a cent.
DRM? That's small potatoes.
Encryption is widely used for commerce. You tell people they can't use encryption, and all e-commerce stops hard. People don't want to hand over their details unencrypted ready for the stealing. So they're not going to shop at eBay or Amazon or Alibaba or any other online store anymore. Given how much online commerce is done, this is a rather big problem.
Likewise, no one will want to bank online as well, so they'll either hoard cash, or line up at the bank. What took minutes before will take days, and the slower the cash moves, the slower the economy moves. I'm sure it's a lot easier and quicker to click "Add to Cart" than it is to call up a merchant and order stuff by phone.
Sorry but a CD rip to FLAC is a joke. call me when you have found that rare japan release on SACD and then ripped that to FLAC....
Which is a joke because you can't "rip" SACD to FLAC. You must convert it.
SACDs are in a format called Direct Stream Digital, or DSD. Aka 1-bit because they are 1-bit ADCs and DACs run very fast. (Sampling theorem states if you have an N bit converter, you can oversample it by 2^M times the required sampling rate to get an N+M bit converter. So if you have a bandlimited audio signal and a 14-bit 48kHz ADC, if you oversample it at 192kHz, you turn it into the equivalent of a 16-bit ADC at 48kHz.)
DSD is actually known as "Pulse Density Modulation" (PDM) as opposed to traditional Pulse Code Modulation (PCM) which we normally deal with, and Pulse Width Modulation (PWM).
Of course, the irony of SACDs is PDM is never used directly - if the microphone captures as PDM, it's converted to PCM for processing/mastering/arranging, then converted back to PDM for mastering. (Your cellphone most likely has a PDM "Digital" microphone).
To rip an SACD, you need an old PS3 (fat, running Linux) that can image the SACD. You then run it through a PDM to PCM converter to get audio you can use which is then compressed using FLAC. And the default DSD frequency is about 2.9MHz, which is 44.1kHz * 2^16, turning the 1-bit ADC into a 16-bit ADC..
I wouldn't call 32GB USB drives "cheap" just yet, but you're right that a limit of 2GB "because of USB flash drives" is an extremely low target. I'm not even sure I'd be able to find 4GB or even 8GB drives in stores anymore.
Well, 4GB drives are hard, and I think I saw 8GB ones at the dollar store for around $3 or so. Which is basically cheap and free - I think people give away 4/8GB sticks nowadays. 16GB sticks are maybe $5-10 or so.
The reason is well, Linux users themselves are their own worst enemy.
Desktop use cases are complex. It's something that requires big complex monoliths like systemd and pulseaudio to handle.
Take, for example a simple audio thing. You have a VoIP client running in the background, and you're going about watching YouTube and other thing when someone calls you. Your VoIP add has to play a sound, and if the browser monopolized the sound device, then the ringing would be lost. But no, it's smoothly mixed into the audio and you click answer. The VoIP app opens up the microphone and speakers you were hearing your YouTube videos with and you start the call.
Midway through you realize you're making a disturbance or want a bit more privacy, so you turn on your Bluetooth headset or plug in a USB headset. The audio stack realizes that this is a preferred device for communications when it's present, so it seamlessly transitions the audio playback and recording to the USB or Bluetooth headset, WITHOUT the application skipping a beat or the user doing anything other than setting it up initially.
The application may get a notification it happened, but it doesn't have to care since it was all designed to happen transparently you can't do this if you opened the sounddevice directly.
If a lot of people are wowed by product A and bored silly by product B, it is irrational to argue that the two can't be ranked as to which is the one most worth investing time in to read, given that we have a finite amount of time to spend doing so. Therefore to get people to vote for their 'favourite' seems a rational way forward, despite its subjective foundation.
That depends on the book - some authors are universally bad (L Ron Hubbard) and some are universally good. But those are the exceptions - a book that one person might find bad, another might find really enjoyable. (I don't like fantasy, for example, so LotR would be a negative for me, but that doesn't mean LotR is bad). So now you have to realize that you're building a suggestion engine - if you like a book, then others who liked that book also liked these other books. If you disliked the book, then those other books may or may not be appealing.
Then you get into really complex sets of parameters that determine if you're going to like a book.
It's complex and depending on your tastes, you might find a book that's good is one you hate.
In this case it's the file system hierarchy, not the file system. Personally, I think the argument for longer filenames is bogus. Using longer filenames isn't necessarily going to make their purpose any more clear, and for everything outside of the home folder, the novice user should probably not be touching that stuff, any more than they should be poking around in C:\Windows. Being user friendly is not a feature for things that are not intended for casual use. Autocomplete is an even worse argument: I'm not saving any keystrokes by typing/bi[TAB] versus/bin.
Well, depends. Most people seem to think/sbin/bin are relatively interchangeable, with perhaps/sbin holding "superuser" binaries (hence the 's'). OTOH, the s really meant static, so/sbin held staticly linked binaries so in an emergency, you could try to recover your system using those tools. (Its why it's/sbin/init - the environment isn't set up yet for the dynamic linker). And people think it's superuser stuff because well, those tools are generally what superusers use (because you're using them to fix your computer).
Similary, people think/usr is where the "user" stuff goes - applications and programs users use. Instead, it's Unix System Resources which contains things that make this Unix system useful to users.
Of course, these days it's all a mish-mash and a binary can be somewhere - dynamics in/sbin, statics in/bin, executables in/opt and/var, etc.
Flight software typically doesn't have lots of extra capability: you have to test it over the entire range, so it tends to be "do we have a specific requirement for that? Yes: build it and test it; No, it's nice to have: Don't build it" So your idea of "incorporate lots of flexibility against potential future devices" would be a non-starter: what requirement would you design against for that "potential future device"? How would you justify that particular requirement, as opposed to another? Say your existing software MUST handle 100 byte messages from the reaction wheel controller. and you want to say "why don't we code it for 1000 bytes to make room for expansion?".. that extra space comes at a cost: memory costs money, testing to 1000 costs more than testing to 100. And ultimately, someone will say "well, why not 2000? or 500?" - unless there's some natural "breakpoint" in the cost function, there's no good rationale.
You missed out the extra testing required.
Today your gyro takes 100 bytes. And you build your code for tomorrow's 1000 byte gyro control. Crap, you need to test your code with both kinds to make sure it handles that larger buffer even though it never will occur on the current platform. All that extra testing and all that for essentially useless code drives up costs. For flight-safety software, it already costs a ton of money to provably be correct, and now it has to be provably correct for the hardware that exists now, and correct for hardware that doesn't exist now and will not exist when the software is sent up.
At best, you can write your code so constants and all that are reasonably self-contained so your 100 byte buffer code can be re-used with 1000 bytes later on.
I've always wondered why people would think that fingerprints are a highly secured method of authentication. You leave the things around everywhere you go and you can't change them if they are compromised. Imagine if you dropped little strips of paper with your password (that could never be changed) written on it everywhere you went. How long would your "highly secured" password last if someone decided they wanted into your account? Especially if that person was the government?
And that's why Apple disables the fingerprint reader - after 3 unsuccessful attempts to use the fingerprint reader, 48 hours of no fingerprint, or on a power up.
And people think Apple's method is "asinine" for requiring a passcode. The only reason Apple has a fingerprint reader was to make phones more secure by having more people actually USE a passcode. Because passcodes are a pain when you're having to enter them in 1000 times a day, so a good majority of users don't do that. The fingerprint reader lets you have a passcode but not have to go through the hassle of entering it thousands of times a day.
Yeah, it's hard to take Gawker seriously these days since it's just a big pile of click-bait. That is their business model - writers are paid by the click.
And you also have to remember Gakwer is in a small bit of trouble, what with a lawsuit and some despicable courtroom behavior that surprisingly hasn't got the attention of the judge for openly defying his order.
Oh yeah, and the iPhone 4 thing that while generating a lot of money when it happened... basically locked Gawker out of Apple events for the rest of their existence
And the guy had it on his wrist for 10 months, and took it off two months ago then writes the review? I mean, you had to stop using for two months to figure out its bad parts? Or you overlooked those bad things for 10 whole months?
Seems a lot of bandwagoning to me - Apple post bad results and now everyone's trying to heap on the bad news because of it? (Hint: Apple, dying for 40 years and 1 month). (and by "bad results" it's "we made less money" and less profit, not "we made a loss". They're still raking in money hand over fist, just not as quickly...)
There are billions of things wrong with Apple. Any Android fanboy will give you a good portion of them. Any Windows/Linux fanboy will give you a chunk more.
A bad week for Apple is a week most other companies would kill for.
I think it's more a message to content producers to stop dropping the ball and start releasing your content in all markets, rather than hiding behind geo-blocking as a way to delay releasing it in markets where it's "inconvenient" or "doesn't fit with your profit-maximizing release schedule." If you want to implement different pricing in different markets, then sure go ahead. But don't stupidly withhold content from certain markets while the Internet is abuzz with talk about what happened in the latest episode, then come to the country's government complaining that people in that country are pirating the show.
The reason content producers are doing it the way they are is simple. Money All the distributors are willing to pay $$$$$ for geographic distribution exclusivity. If they weren't granted it (say, to give Netflix the ability to show it too), then they'd only be willing to pay $, and Netflix pays $. So the math is obvious - with geographic limitations, you get $$$$$. With no geographic restrictions, you only get $$.
It's no longer about timing or variable pricing - timing on a lot of shows can be 24 hours worldwide, and pricing is set locally - based again on the exclusivity.
I'm sure Netflix COULD try to pay for worldwide distribution, but then you'd be basically paying $100/month for Netflix - because someone has to max up the difference in the money that the content producers will get going exclusive versus not.
Remember, you're paying for a right to a virtual monopoly, so you're going to pay a lot of money. If you're going to have to compete, you're not going to pay a lot of money. That's where the difference in the money is.
Now, more t hings like this could help reduce the payments - because those distributors will lean on the providers and tell them that unless they enforce the blocks, they're not going to pay so much anymore for the programming. Then it's a back and forth -- will the loss of money from that distributor be made up by offering other companies the right to distribute non-exclusively?
Backwards compatibility is important. Why drop it? 16-bit support is finally gone, but I suspect only because everything anyone still uses (games) has been virtualized already.
16 bit is only gone if you are running the 64 bit version of Windows. if you are running the 32 bit version, which for some bizaare reason still exists, even in Windows 10, then you can still run 16 bit programs.
16 bit is gone because AMD64 does not support it. It was an architectural decision. Win64 can only run Win64 and Win32 apps because that's all the underlying processor supports To run 16 bit requires an emulator - which is what the virtualization products do - they run the 16 bit code in an emulator until it switches to 32 bit mode at which point they run the code on the hardware itself.
Windows 10 supports 32 bit because there are a lot of products where it doesn't make sense - low cost PCs often have 1-2 GB of RAM, so why run 64 bit? I mean, my tablet runs 32-bit Windows 10 because it has 1GB of RAM (and it only cost $100) And of course, I have a couple of Win16 apps that I still can't find equivalents for. Of course, Win16 runs under an emulated Windows 3.1 environment and Windows 10 needs to install the NTVDM emulator for Win16 (complete with classic Win3.1 iconography).
shouldn't every company that gives out private authentication tokens for developer to use should be monitoring sites like github and revoking any tokens found? when you sign up to get an authentication token it says you have to keep it secret or it will be revoked, so why aren't more people doing this?
Why? You assume developers who ask for a key are smart enough to protect the things. I mean, they're developers, not users. They should be smart enough to know how to protect their keys. Especially when they use source code control systems.
And scanning GitHub and others is just a small part of all the places you might find the key, so no company would be able to find them all.
Also, I don't like how we can't do TDD/TTY with smartphones and Internet directly. Why have those old school devices? Argh.
Funny, my company spent (several years ago when we did smartphones) a lot of time and effort getting TDD/TTY devices working with them. Not because we wanted to, but it was a carrier requirement for our customers. Admittedly, it was basically a device that hooked into the headset jack of the phone and transmitted tones over it. Now, the network doesn't transmit the tones directly I'm led to believe - the cellular modem was put into TDD/TTY mode and we fed the audio directly to it where it's decoded and sent over the network as data instead of voice packets.
Of course, these days one wonders why you can't have an app for that, but our smartphones had, by our customers request, TDD/TTY support. Heck, until the customers asked, we didn't even know it was supported.
Apple doesn't have anywhere near enough capital to go private, and I doubt they're going be able to borrow more because they already have a lot of debt. Their stock would have to seriously decline first.
Actually, Apple doesn't have a lot of debt. In fact, the only reason they have debt is because it's cheaper to borrow the money than repatriate it. The interest rate on that money is around 1-2%. If they repatriated that cash, they would lose 40% of it to Uncle Sam.
The big selling point for Atom is that it's almost as efficient as ARM but it runs REAL WINDOWS with all those x86 programs we love. What killed the market for Atom is that people aren't that eager to have Windows on portable devices. Intel went through contortions to implement all the x86 instructions on low-power chip, to support all the legacy software that's written for x86. But with iOS and Android, ARM seems to have all the apps that people want, and they just don't pine for the legacy stuff.
Well, I love my atom-based tablet that came with Windows 8.1 and upgraded to Windows 10. For $100, it does a lot more than a $100 Android tablet (I use an iPad), and being able to just run Win32 stuff (or even Win16 stuff!) is a dream.
What was surprising is how peppy that little chip is - it won't beat my i7 desktop, but for on the go stuff, it works surprisingly well. The only downsides is that 1GB of RAM and limited storage (yes, it has an SD slot, but SD is sloowwww)
Heck, I bought a couple more Atom based PCs - a Lenovo Think Stick and a Kangaroo mobile desktop, both of which were obtained for $100 and $150. Cheap micro desktop PCs.
And yes, I love the ability to run old software on it - I mean, you can play MP3s on any device, but you can't run WinAmp and all the plugins on just any tablet. I wonder how BlueStacks would run on it for the quick Android fix...
Benchmarks are necessary, but not sufficient way to test things.
The reason for benchmarks is simple - you want a scientifically repeatable test that can be used to compare things with each other. This limits the benchmark's utility as a real-world test because it's inherently limited in what it can test. All it gives is how your thing measures up to all the other things out there. And yes, benchmarks will be gamed, doesn't matter the field (see VW, Mitsubishi and everyone else with diesel engines). However, that doesn't mean their utility is null - it's a comparison tool. Just like how your fuel consumption figures are based on somewhat unrealistic test scenarios, they're like that because they have to be repeatable and comparable.
But on devices which are complex, a benchmark will never cover all the use cases and will never cover everything.
(E.g., an audio amplifier is really simple and a benchmark can cover everything because its job is to increase signal strength, so all you need as a benchmark is how far the output waveform deviates from the input waveform. But a preamp that say corrects for room deficiencies cannot be tested by benchmarks alone because its too complex).
So for complex measurements (or things not fully quantifiable, e.g., "image looks better" or "clearer" or "faster", then you need more tests.
Although, for the imaging system a benchmark should be good enough - as all it needs to do is take a photo of a calibration chart and measure the final photo output for errors. Other aspects like lag can be measured as well. In which case if they produce the same results, they should be just as good. (This will be analyzing the image itself internally or via the same screen). If the S7 images are "more vibrant" then perhaps it's the screen itself since OLEDs are known to oversaturate and produce nice looking, but completely color inaccurate photos.
No, this is NOT about ClearChannel's revenue, it's about the fact that the 100w station's signal wanders for a far wider, uncontrolled range, then interfering with the fringes of others that have paid for lots of engineering plans, capital infrastructure, and the operational expenses in hopes of a profit from their station.
If that is true, then the licensed station takes it up with the FCC who goes and finds said pirate radio station.
It's how it's always worked. Most licensed stations use hundreds of kW to transmit (which is why you can pick up a signal pretty much anywhere with even the crappiest of radios), and as long as you're sticking with the broadcast band, you're not going to interfere with non-broadcast licensed services like aircraft (unless you have a terrible transmitter).
FM also has what's called the "capture effect" - if you have two FM stations on the same frequency, the one you hear is the more powerful signal and it acts like the other signal doesn't exist. (It's why aircraft use AM - stepping on another transmission results in a squeal on the receiver).
And pirates intentionally check to make sure their coverage area doesn't interfere - less chance of getting caught if you're not interfering with licensed services. At best, they'd interfere with low-power FM (which are unlicensed, but not pirate stations that transmit with low power for a small area). But with LPFM, the rules are still the same - it's still unlicensed.
Oh Bullshit, the only freedom the GPL limits is the freedom to be a commercial leach and sell other people's work as your own. If you don't want to play by the GPL rules, fine, don't, but quit whining about losing freedoms that you never had.
Said the GPL fanatic who loves to leech BSD code as well.
It's probably worse because GPL locks up the BSD code - any improvements made to the BSD code cannot be contributed back to the original project!
So maybe it should be less about "closed source leeching" and more about "GPL leeching" as well. Because at least Microsoft and other closed-source companies either contribute, or don't. GPL code makes improvements that cannot be integrated back whilst claiming superiority. And parading the modified code as a big F-U to the BSD folks
Always notice how GPL always claims "closed source exploitation" and not "GPL exploitation"? I think this attitude is worse than just Microsoft etc. "leeching"..
Yes iTunes is bloated piece of crap and I never let my iOS device go near it.
Which is a shame because iTunes gives iOS a couple of advantages over Android.
First, encrypted backups through iTunes backup EVERYTHING. Authentication information (which is omitted on non-encrypted backups for obvious reasons) is backed up, as is a bunch of stuff Apple would rather not have on their servers where government can obtain it by warrant.
Having a local backup is good - iCloud backs up the bare minimum - just the stuff Apple won't mind holding for you. iTunes normal backups back up more stuff since they include stuff Apple rather not have on their servers. Encrypted backups are best for they include everything.
The second reason is app backups. Apps disappear from the iTunes store all the time - either the developer fails to renew their iOS development certificate, the app is pulled or the developer pulls it. If you don't have a local backup of the IPA file, then if you delete the app, you can't get it back.
But if you back up the IPA file locally, then even if it's removed from the store, you can reinstall it on your devices at will.
But that's really the entire utility of iTunes nowadays that's actually useful.
It looks pretty good for the money. You can of course run other operating systems on it. Linux, certainly, maybe even Windows. As a Linux machine it's pretty cheap for the spec.
Actually... no. The default Chromebook loader can only boot Linux. There is no BIOS unless the boot firmware includes one (e.g., the Chromebook Pixel included SeaBIOS). This is required if you want to boot... Windows.
Even then, it's a nasty hack - while you can do it, only geeks will be satisfied with the result. Basically, every time you boot it up (including reboots), its going to wait 30 seconds while it says the system is insecure. After 30 seconds, it'll display a screen asking for a recovery USB or SD card. So in those 30 seconds you need to hit a key combo to proceed to boot (Ctrl-D or something).
This will be fine for most/. users, but as a "cheap laptop" solution for the general public... not so much. Needing to hit the key command within a certain period of time is probably going to be the biggest sticking point.
By default, ChromeOS is designed to be locked down so people can use it to browse the web in safety without worrying that something is doing to install something that will destroy their computer, or spy on them, etc. They make great machines to do internet banking with as you can be reasonably assured they have no malware or other thing spying on you. But that also makes life difficult if you want to use it in unsecured open mode.
* they shouldn't be allowed to charge per GB without offering better tools for their users
"they" shouldn't provide the tools at all.
NIST or equivalent body should be generating the standards for measuring traffic use, and if Comcast wants to charge per-GB, they need to provide everyone with a NIST-calibrated and certified meter. Just like your water meter, your gas meter, your electric meter has calibration stamps and seals to indicate that yes, what they measure is accurate.
That meter should have a 0-cost way of reading it - i.e., a display showing how much you've used.
Until that is satisfied, there is no way to do this properly.
Think about it for a moment. There is no standard way to measure a byte. is it 1TB or 1TiB? What bytes are you including? TCP/IP headers and above? Or are you going to toss in DOCSIS headers as well? (your cellphone provider typically includes OTA packet headers and SI prefixes in their plans - the former adds around 10% overhead on average, the latter, well, you know).
Also, will there be a way to disconnect? If you're being DDoS'd, it's going to be annoying to pay for that kind of traffic, so if there is a way to disconnect and incur 0 usage...
Thus, if you're going to sell something on a per-anything, you need to make sure your measurement tools follow a standard.
It makes the ethics of what he did a bit less clear to me. He spent years telling people how to be secure on Tor, then spent a few more unmasking those who didn't listen.
No, I think that's actually perfectly valid.
Because if you want to protect your anonymity, you have to take steps to do so. Tor is not a magic bullet, it has known flaws since the beginning (e.g., exit nodes) and doing stupid things will make you readily identifiable.
In fact, too many people are using Tor as a tool improperly - it's like using encryption improperly. You get a false sense of security when in reality you're making yourself plainly visible. Or using HTTPS and storying your passwords in plain text
No, "just use Tor" will not make you magically anonymous, especially if you immediately go and log into Facebook and Amazon and everywhere else. But too many people believe it will and blithely continue using the 'net as if Tor magically anonymizes them.
So demonstrating that people are stupid isn't a crime - in fact it should be published far and wide so people using it know what people can get at.
And what if the Uber car is on the other side of a busy road? Crossing legally at a crosswalk can easily be more than two minutes to walk to the crosswalk, wait for the traffic light to let you cross, then walk to the car.
Or it can easily be more than two minutes to cross in the middle of the street.
And finding the car is hard enough - perhaps there should be a button saying "Where are you" which tells the driver to honk so you can find them. Because looking for them can easily take a good few minutes as well.
Perhaps Uber needs to have drivers do something. Lyft has those pink car moustaches that are easy to see (from the front, they probably need something for the rear).
Slashdot Mirror
And the reason to have a fingerprint sensor is the fact that a majority of smartphone users have no passcode or other security measure set. The reason is simple - a phone is accessed extremely often - I believe Apple quotes easily a thousand times or more a day. When you're using it that often (most of the time it's used for under 5 seconds), a passcode gets in the way - who wants to enter a passcode that many times? End result, they don't secure their phone at all.
The fingerprint sensor lets a user choose a passcode and access their phone without taking significantly longer (it would suck if it took longer to enter your passcode than you were going to use it). So while the OS is resuming from low power state the sensor can be reading the fingerprint and the secure enclave can determine if it's a valid fingerprint, so when the OS is ready the decision is ready.
Of course, Apple also has ways to ensure that the fingerprint is overridable - three bad reads triggers a passcode, a power cycle triggers a passcode and no unlocking for 48 hours also triggers a passcode request.
The law is currently unclear as to what happens when this is triggered.- if you are forced to unlock the phone by fingerprint, but the phone now requires the passcode...
Well, that is the problem - the Compaq is the only laptop with a special slot that accepts the diagnostics interface card. That's the only reason it's there - the special card you need.
Depending on how proprietary that card is, it may or may not be possible to replace it - I think it was made before every car had an OBD-II port that there's interfaces for via USB, Bluetooth, WiFi and every other thing you have.
In the end, the solution might be to just replace all the ECUs with an OBD-II port. The only reason they haven't is the F1 is a pretty bespoke car, there aren't a lot of them on the streets so there's not that many of them to begin with, so any replacement is basically going to be one off development.
Of course, I don't think there's anything non-standard about the interface - a computer that age is probably using ISA on that port so it should be possible to convert it to a standard ISA bus card, then use a PLX bridge to give you a PCI(e) interface on top of that.
Freedom also means eternal vigilance. There are people who are jealous of people who have freedom and will do anything in their power to destroy it. (We normally call these people "terrorists").
Having freedom means we don't try to oppress our freedoms to get rid of these bad people, but we live with them - it's the price of living in a free world.
We can use encryption, and so can they. Our law enforcement agencies need to take that into account and use more traditional forms of detective work to get at the bad guys. In fact, the NSA's data collection system is completely useless because they collect so much data, not only are they looking for needles in haystacks, but with all the data collection, the haystacks have ballooned in size, while the number of needles stayed the same. More data isn't better. In fact, the other terrorists have been caught by good old traditional detective work.
That's what's really needed - more boots on the ground.
DRM? That's small potatoes.
Encryption is widely used for commerce. You tell people they can't use encryption, and all e-commerce stops hard. People don't want to hand over their details unencrypted ready for the stealing. So they're not going to shop at eBay or Amazon or Alibaba or any other online store anymore. Given how much online commerce is done, this is a rather big problem.
Likewise, no one will want to bank online as well, so they'll either hoard cash, or line up at the bank. What took minutes before will take days, and the slower the cash moves, the slower the economy moves. I'm sure it's a lot easier and quicker to click "Add to Cart" than it is to call up a merchant and order stuff by phone.
Which is a joke because you can't "rip" SACD to FLAC. You must convert it.
SACDs are in a format called Direct Stream Digital, or DSD. Aka 1-bit because they are 1-bit ADCs and DACs run very fast. (Sampling theorem states if you have an N bit converter, you can oversample it by 2^M times the required sampling rate to get an N+M bit converter. So if you have a bandlimited audio signal and a 14-bit 48kHz ADC, if you oversample it at 192kHz, you turn it into the equivalent of a 16-bit ADC at 48kHz.)
DSD is actually known as "Pulse Density Modulation" (PDM) as opposed to traditional Pulse Code Modulation (PCM) which we normally deal with, and Pulse Width Modulation (PWM).
Of course, the irony of SACDs is PDM is never used directly - if the microphone captures as PDM, it's converted to PCM for processing/mastering/arranging, then converted back to PDM for mastering. (Your cellphone most likely has a PDM "Digital" microphone).
To rip an SACD, you need an old PS3 (fat, running Linux) that can image the SACD. You then run it through a PDM to PCM converter to get audio you can use which is then compressed using FLAC. And the default DSD frequency is about 2.9MHz, which is 44.1kHz * 2^16, turning the 1-bit ADC into a 16-bit ADC..
Well, 4GB drives are hard, and I think I saw 8GB ones at the dollar store for around $3 or so. Which is basically cheap and free - I think people give away 4/8GB sticks nowadays. 16GB sticks are maybe $5-10 or so.
The reason is well, Linux users themselves are their own worst enemy.
Desktop use cases are complex. It's something that requires big complex monoliths like systemd and pulseaudio to handle.
Take, for example a simple audio thing. You have a VoIP client running in the background, and you're going about watching YouTube and other thing when someone calls you. Your VoIP add has to play a sound, and if the browser monopolized the sound device, then the ringing would be lost. But no, it's smoothly mixed into the audio and you click answer. The VoIP app opens up the microphone and speakers you were hearing your YouTube videos with and you start the call.
Midway through you realize you're making a disturbance or want a bit more privacy, so you turn on your Bluetooth headset or plug in a USB headset. The audio stack realizes that this is a preferred device for communications when it's present, so it seamlessly transitions the audio playback and recording to the USB or Bluetooth headset, WITHOUT the application skipping a beat or the user doing anything other than setting it up initially.
The application may get a notification it happened, but it doesn't have to care since it was all designed to happen transparently you can't do this if you opened the sounddevice directly.
That depends on the book - some authors are universally bad (L Ron Hubbard) and some are universally good. But those are the exceptions - a book that one person might find bad, another might find really enjoyable. (I don't like fantasy, for example, so LotR would be a negative for me, but that doesn't mean LotR is bad). So now you have to realize that you're building a suggestion engine - if you like a book, then others who liked that book also liked these other books. If you disliked the book, then those other books may or may not be appealing.
Then you get into really complex sets of parameters that determine if you're going to like a book.
It's complex and depending on your tastes, you might find a book that's good is one you hate.
Well, depends. Most people seem to think /sbin /bin are relatively interchangeable, with perhaps /sbin holding "superuser" binaries (hence the 's'). OTOH, the s really meant static, so /sbin held staticly linked binaries so in an emergency, you could try to recover your system using those tools. (Its why it's /sbin/init - the environment isn't set up yet for the dynamic linker). And people think it's superuser stuff because well, those tools are generally what superusers use (because you're using them to fix your computer).
Similary, people think /usr is where the "user" stuff goes - applications and programs users use. Instead, it's Unix System Resources which contains things that make this Unix system useful to users.
Of course, these days it's all a mish-mash and a binary can be somewhere - dynamics in /sbin, statics in /bin, executables in /opt and /var, etc.
You missed out the extra testing required.
Today your gyro takes 100 bytes. And you build your code for tomorrow's 1000 byte gyro control. Crap, you need to test your code with both kinds to make sure it handles that larger buffer even though it never will occur on the current platform. All that extra testing and all that for essentially useless code drives up costs. For flight-safety software, it already costs a ton of money to provably be correct, and now it has to be provably correct for the hardware that exists now, and correct for hardware that doesn't exist now and will not exist when the software is sent up.
At best, you can write your code so constants and all that are reasonably self-contained so your 100 byte buffer code can be re-used with 1000 bytes later on.
And that's why Apple disables the fingerprint reader - after 3 unsuccessful attempts to use the fingerprint reader, 48 hours of no fingerprint, or on a power up.
And people think Apple's method is "asinine" for requiring a passcode. The only reason Apple has a fingerprint reader was to make phones more secure by having more people actually USE a passcode. Because passcodes are a pain when you're having to enter them in 1000 times a day, so a good majority of users don't do that. The fingerprint reader lets you have a passcode but not have to go through the hassle of entering it thousands of times a day.
Yeah, it's hard to take Gawker seriously these days since it's just a big pile of click-bait. That is their business model - writers are paid by the click.
And you also have to remember Gakwer is in a small bit of trouble, what with a lawsuit and some despicable courtroom behavior that surprisingly hasn't got the attention of the judge for openly defying his order.
Oh yeah, and the iPhone 4 thing that while generating a lot of money when it happened... basically locked Gawker out of Apple events for the rest of their existence
And the guy had it on his wrist for 10 months, and took it off two months ago then writes the review? I mean, you had to stop using for two months to figure out its bad parts? Or you overlooked those bad things for 10 whole months?
Seems a lot of bandwagoning to me - Apple post bad results and now everyone's trying to heap on the bad news because of it? (Hint: Apple, dying for 40 years and 1 month). (and by "bad results" it's "we made less money" and less profit, not "we made a loss". They're still raking in money hand over fist, just not as quickly...)
There are billions of things wrong with Apple. Any Android fanboy will give you a good portion of them. Any Windows/Linux fanboy will give you a chunk more.
A bad week for Apple is a week most other companies would kill for.
The reason content producers are doing it the way they are is simple. Money All the distributors are willing to pay $$$$$ for geographic distribution exclusivity. If they weren't granted it (say, to give Netflix the ability to show it too), then they'd only be willing to pay $, and Netflix pays $. So the math is obvious - with geographic limitations, you get $$$$$. With no geographic restrictions, you only get $$.
It's no longer about timing or variable pricing - timing on a lot of shows can be 24 hours worldwide, and pricing is set locally - based again on the exclusivity.
I'm sure Netflix COULD try to pay for worldwide distribution, but then you'd be basically paying $100/month for Netflix - because someone has to max up the difference in the money that the content producers will get going exclusive versus not.
Remember, you're paying for a right to a virtual monopoly, so you're going to pay a lot of money. If you're going to have to compete, you're not going to pay a lot of money. That's where the difference in the money is.
Now, more t hings like this could help reduce the payments - because those distributors will lean on the providers and tell them that unless they enforce the blocks, they're not going to pay so much anymore for the programming. Then it's a back and forth -- will the loss of money from that distributor be made up by offering other companies the right to distribute non-exclusively?
16 bit is gone because AMD64 does not support it. It was an architectural decision. Win64 can only run Win64 and Win32 apps because that's all the underlying processor supports To run 16 bit requires an emulator - which is what the virtualization products do - they run the 16 bit code in an emulator until it switches to 32 bit mode at which point they run the code on the hardware itself.
Windows 10 supports 32 bit because there are a lot of products where it doesn't make sense - low cost PCs often have 1-2 GB of RAM, so why run 64 bit? I mean, my tablet runs 32-bit Windows 10 because it has 1GB of RAM (and it only cost $100) And of course, I have a couple of Win16 apps that I still can't find equivalents for. Of course, Win16 runs under an emulated Windows 3.1 environment and Windows 10 needs to install the NTVDM emulator for Win16 (complete with classic Win3.1 iconography).
Why? You assume developers who ask for a key are smart enough to protect the things. I mean, they're developers, not users. They should be smart enough to know how to protect their keys. Especially when they use source code control systems.
And scanning GitHub and others is just a small part of all the places you might find the key, so no company would be able to find them all.
Funny, my company spent (several years ago when we did smartphones) a lot of time and effort getting TDD/TTY devices working with them. Not because we wanted to, but it was a carrier requirement for our customers. Admittedly, it was basically a device that hooked into the headset jack of the phone and transmitted tones over it. Now, the network doesn't transmit the tones directly I'm led to believe - the cellular modem was put into TDD/TTY mode and we fed the audio directly to it where it's decoded and sent over the network as data instead of voice packets.
Of course, these days one wonders why you can't have an app for that, but our smartphones had, by our customers request, TDD/TTY support. Heck, until the customers asked, we didn't even know it was supported.
Actually, Apple doesn't have a lot of debt. In fact, the only reason they have debt is because it's cheaper to borrow the money than repatriate it. The interest rate on that money is around 1-2%. If they repatriated that cash, they would lose 40% of it to Uncle Sam.
Well, I love my atom-based tablet that came with Windows 8.1 and upgraded to Windows 10. For $100, it does a lot more than a $100 Android tablet (I use an iPad), and being able to just run Win32 stuff (or even Win16 stuff!) is a dream.
What was surprising is how peppy that little chip is - it won't beat my i7 desktop, but for on the go stuff, it works surprisingly well. The only downsides is that 1GB of RAM and limited storage (yes, it has an SD slot, but SD is sloowwww)
Heck, I bought a couple more Atom based PCs - a Lenovo Think Stick and a Kangaroo mobile desktop, both of which were obtained for $100 and $150. Cheap micro desktop PCs.
And yes, I love the ability to run old software on it - I mean, you can play MP3s on any device, but you can't run WinAmp and all the plugins on just any tablet. I wonder how BlueStacks would run on it for the quick Android fix ...
Benchmarks are necessary, but not sufficient way to test things.
The reason for benchmarks is simple - you want a scientifically repeatable test that can be used to compare things with each other. This limits the benchmark's utility as a real-world test because it's inherently limited in what it can test. All it gives is how your thing measures up to all the other things out there. And yes, benchmarks will be gamed, doesn't matter the field (see VW, Mitsubishi and everyone else with diesel engines). However, that doesn't mean their utility is null - it's a comparison tool. Just like how your fuel consumption figures are based on somewhat unrealistic test scenarios, they're like that because they have to be repeatable and comparable.
But on devices which are complex, a benchmark will never cover all the use cases and will never cover everything.
(E.g., an audio amplifier is really simple and a benchmark can cover everything because its job is to increase signal strength, so all you need as a benchmark is how far the output waveform deviates from the input waveform. But a preamp that say corrects for room deficiencies cannot be tested by benchmarks alone because its too complex).
So for complex measurements (or things not fully quantifiable, e.g., "image looks better" or "clearer" or "faster", then you need more tests.
Although, for the imaging system a benchmark should be good enough - as all it needs to do is take a photo of a calibration chart and measure the final photo output for errors. Other aspects like lag can be measured as well. In which case if they produce the same results, they should be just as good. (This will be analyzing the image itself internally or via the same screen). If the S7 images are "more vibrant" then perhaps it's the screen itself since OLEDs are known to oversaturate and produce nice looking, but completely color inaccurate photos.
If that is true, then the licensed station takes it up with the FCC who goes and finds said pirate radio station.
It's how it's always worked. Most licensed stations use hundreds of kW to transmit (which is why you can pick up a signal pretty much anywhere with even the crappiest of radios), and as long as you're sticking with the broadcast band, you're not going to interfere with non-broadcast licensed services like aircraft (unless you have a terrible transmitter).
FM also has what's called the "capture effect" - if you have two FM stations on the same frequency, the one you hear is the more powerful signal and it acts like the other signal doesn't exist. (It's why aircraft use AM - stepping on another transmission results in a squeal on the receiver).
And pirates intentionally check to make sure their coverage area doesn't interfere - less chance of getting caught if you're not interfering with licensed services. At best, they'd interfere with low-power FM (which are unlicensed, but not pirate stations that transmit with low power for a small area). But with LPFM, the rules are still the same - it's still unlicensed.
Said the GPL fanatic who loves to leech BSD code as well.
It's probably worse because GPL locks up the BSD code - any improvements made to the BSD code cannot be contributed back to the original project!
So maybe it should be less about "closed source leeching" and more about "GPL leeching" as well. Because at least Microsoft and other closed-source companies either contribute, or don't. GPL code makes improvements that cannot be integrated back whilst claiming superiority. And parading the modified code as a big F-U to the BSD folks
Always notice how GPL always claims "closed source exploitation" and not "GPL exploitation"? I think this attitude is worse than just Microsoft etc. "leeching"..
Which is a shame because iTunes gives iOS a couple of advantages over Android.
First, encrypted backups through iTunes backup EVERYTHING. Authentication information (which is omitted on non-encrypted backups for obvious reasons) is backed up, as is a bunch of stuff Apple would rather not have on their servers where government can obtain it by warrant.
Having a local backup is good - iCloud backs up the bare minimum - just the stuff Apple won't mind holding for you. iTunes normal backups back up more stuff since they include stuff Apple rather not have on their servers. Encrypted backups are best for they include everything.
The second reason is app backups. Apps disappear from the iTunes store all the time - either the developer fails to renew their iOS development certificate, the app is pulled or the developer pulls it. If you don't have a local backup of the IPA file, then if you delete the app, you can't get it back.
But if you back up the IPA file locally, then even if it's removed from the store, you can reinstall it on your devices at will.
But that's really the entire utility of iTunes nowadays that's actually useful.
Actually... no. The default Chromebook loader can only boot Linux. There is no BIOS unless the boot firmware includes one (e.g., the Chromebook Pixel included SeaBIOS). This is required if you want to boot... Windows.
Even then, it's a nasty hack - while you can do it, only geeks will be satisfied with the result. Basically, every time you boot it up (including reboots), its going to wait 30 seconds while it says the system is insecure. After 30 seconds, it'll display a screen asking for a recovery USB or SD card. So in those 30 seconds you need to hit a key combo to proceed to boot (Ctrl-D or something).
This will be fine for most /. users, but as a "cheap laptop" solution for the general public... not so much. Needing to hit the key command within a certain period of time is probably going to be the biggest sticking point.
By default, ChromeOS is designed to be locked down so people can use it to browse the web in safety without worrying that something is doing to install something that will destroy their computer, or spy on them, etc. They make great machines to do internet banking with as you can be reasonably assured they have no malware or other thing spying on you. But that also makes life difficult if you want to use it in unsecured open mode.
"they" shouldn't provide the tools at all.
NIST or equivalent body should be generating the standards for measuring traffic use, and if Comcast wants to charge per-GB, they need to provide everyone with a NIST-calibrated and certified meter. Just like your water meter, your gas meter, your electric meter has calibration stamps and seals to indicate that yes, what they measure is accurate.
That meter should have a 0-cost way of reading it - i.e., a display showing how much you've used.
Until that is satisfied, there is no way to do this properly.
Think about it for a moment. There is no standard way to measure a byte. is it 1TB or 1TiB? What bytes are you including? TCP/IP headers and above? Or are you going to toss in DOCSIS headers as well? (your cellphone provider typically includes OTA packet headers and SI prefixes in their plans - the former adds around 10% overhead on average, the latter, well, you know).
Also, will there be a way to disconnect? If you're being DDoS'd, it's going to be annoying to pay for that kind of traffic, so if there is a way to disconnect and incur 0 usage...
Thus, if you're going to sell something on a per-anything, you need to make sure your measurement tools follow a standard.
No, I think that's actually perfectly valid.
Because if you want to protect your anonymity, you have to take steps to do so. Tor is not a magic bullet, it has known flaws since the beginning (e.g., exit nodes) and doing stupid things will make you readily identifiable.
In fact, too many people are using Tor as a tool improperly - it's like using encryption improperly. You get a false sense of security when in reality you're making yourself plainly visible. Or using HTTPS and storying your passwords in plain text
No, "just use Tor" will not make you magically anonymous, especially if you immediately go and log into Facebook and Amazon and everywhere else. But too many people believe it will and blithely continue using the 'net as if Tor magically anonymizes them.
So demonstrating that people are stupid isn't a crime - in fact it should be published far and wide so people using it know what people can get at.
And what if the Uber car is on the other side of a busy road? Crossing legally at a crosswalk can easily be more than two minutes to walk to the crosswalk, wait for the traffic light to let you cross, then walk to the car.
Or it can easily be more than two minutes to cross in the middle of the street.
And finding the car is hard enough - perhaps there should be a button saying "Where are you" which tells the driver to honk so you can find them. Because looking for them can easily take a good few minutes as well.
Perhaps Uber needs to have drivers do something. Lyft has those pink car moustaches that are easy to see (from the front, they probably need something for the rear).