Even that's not true. They have a very particular accent that is only used for presenting the news (their other output has much more variation) which is a clarified variation on University English, but it's there if you listen for it. I suppose you could consider it to be at the centre of a cluster of accents that come under the heading of British English.
Similarly, there's another cluster of accents that constitute American English. The spread on the cluster is tighter than on the BE cluster, as far as I can tell. Arguably, you could describe an accent as being an offset from the barycenter of the cluster. You might not think you've got an accent, but from the perspective of someone in another cluster, you most certainly have. (Confusingly, the clusters may overlap somewhat. Yay for complexity!)
No mention of Shock waves, or even a hint of what might cause such shock, or how such shock could be transmitted in the vacuum of space.
Via the interstellar medium, of course. It's pretty tenuous, but most certainly is capable of sustaining phenomena like shock waves. Which isn't to say that that's necessarily the particular process that is dominant in the galactic arms; it could also be something relating to magnetism, as the physics of a flowing magnetically-coupled medium is viciously difficult to work with (i.e., highly non-linear). And I've got no idea what happens at the phase change boundaries between the parts of the ISM which are plasmas and the parts which are conventional (tenuous) gasses; phase changes can do "interesting" things.
As for what's powering it all, you've got some exceptionally powerful energy sources out there. Black holes in particular can pump vast amounts of energy into the surrounding volume of space. The stellar wind from very high mass stars would be another interesting source.
Does ride-sharing really need to be regulated beyond a requirement that the vehicles and drivers have proper insurance?
Is it a taxi service — the taking of a person to a place they nominate in return for a fee — or not? If it is, there is a need for some regulation. An example of the kind of reasonable regulation is to require that nobody with a conviction for a sex crime should be able to be a taxi driver. Having to have a particular level of insurance (or better) is another reasonable regulation. It's all about ensuring that the services that are there are not actively hazardous for people to use.
The problem with services such as the one in the article is that they're trying to offer a taxi service while saying "I promise I'm not a taxi service" and keeping their fingers crossed behind their back. That's basically dishonest. Of course, it's a response to the over-regulation in some locales where the real taxis have actually effectively been organised as a cartel. Which is also wrong. Two wrongs do not make a right; they just layer wrong on wrong.
What's really needed is to pare back the regulation to what is actually needed and to end the medallion system as it has ended up, or at least end the limits on how many are issued and who can have them. That will then also let you ban subcontracting of these things to people who may not have passed background checks, etc. Fix your regulation, you fools!
I must be missing something about this concept. If you're getting paid (with a net profit) to drive people around, why is it called ride sharing? How is it not a taxi service?
They're probably trying to work around some regulations. In the UK it's pretty clear: if it's done for hire, it's one of two types of taxi. A "hackney carriage" is what most people think of as a taxi: it's one you can hail in the street. There's also a "private hire" taxi, where you have to have some sort of prior arrangement before being picked up; it covers a whole range of options (up to a maximum size of vehicle and excluding a few more highly regulated categories) such as limousines and dial-a-rides. If you charge to take someone somewhere, you're covered by the regs. Only genuine ride-sharing or hitching a lift, i.e. stuff which people don't consider hiring, would not be regulated.
Sounds like what you need are a similar sort of thing. There's nothing wrong with having "hire a taxi over the internet" services; they work well. They just shouldn't pretend to be anything other than what they are.
linode.com does month to month, prorates per day. disclaimer - very long time customer, very happy.
And the point of a Cloud is that you can buy as little as an hour, possibly even less with some services. That really changes how you use things, and the sort of business you do with it, as there's a lot of behaviour that varies on that scale much more than it does at the level of months. It's also cheaper if you only need a few hours of processing occasionally; that applies to a lot of things.
Of course, you can combine things. You might put your persistent control systems at fixed IP addresses in linode, but spin up workers in a Cloud as required. Mind you, you can do the same thing inside Clouds using a cheap low-resource instance at a fixed address. Working out the best combination overall is non-trivial, especially in abstract.
I would really like this feature for normal Java applications that use the JVM to get around the firewall.
What? That doesn't make any sense, not unless you're talking about basing whether code can get through the firewall on the path to the executable or something equally silly (given the existence of the JVM, Python, Ruby, Perl,...).
Of course. Admittedly it is by definition, and the part where it says they are is secret so you're not allowed to know about what it exactly involves...
Meanwhile, some 60+ years later and Europe still wasn't making such pencils (well, not commonly enough, anyways).
In my experience (at school, long ago) a separate eraser was better because it did a better job of erasing by virtue of allowing you to use a broader surface for erasing. Convenience is all very well, but I preferred the "convenience" of being able to actually erase pencil markings from the page.
Win8 is an interesting ball of wax. We already in the progress of migration to it -- mostly new Dell laptops. It is universally hated by everyone I talk to. People hate it for two reasons:
a) sake of change for the sake of change when there was nothing "wrong" with the old UI b) Eveyone agrees Metro makes perfect sense on a tablet but screwing over the desktop users pisses off a lot of people because you are forcing them to waste their time and IT's time to relearn how to do the same thing as before. It is a hindrance from us doing our job and we are already overloaded as it is.
Having observed a number of people using Win8, the thing which they reallyreally don't like is the drag-mouse-to-edge to make something invisible pop up that you still need for doing things with the machine. They just don't learn what's going on, and hate it when it happens by accident and doubly hate it when they can't find it when they're looking for it. Setting aside a few pixels so that people can see and be reminded that there is something important there would be by far the best solution, at least on laptops and desktops where there's plenty of screen space.
Windows 3.0 Program Manager supported multiple, overlapping windows.
Strictly, the overlapping windows came in with Windows 2. I had a copy. Win 3.0 was much nicer, if slower and far more bloated. (After all, you didn't need much colour on your screen, did you?)
you're right. what intel does do is an excellent version of frequency scaling, turning off unused execution units, etc. this is better than a dedicated low-power core, because it's finer grained and there's some benefit even without software particilation.
What's it like when you put a real workload on it? Benchmarks are interesting and all, but the proof is always reality. (I mistrust benchmarking because it's so easy to tune things to do well in a benchmark without actually being particularly good on anything else; this has definitely happened in the past with CPU design too.)
Mind you, Intel's main problem is that there's a large and expanding market out there where their CPUs just aren't the things that people choose. People select ARM-based systems because that's what the existing software is written for and what the existing users have, and the benefit of integrating with existing platforms is very strong indeed.
Well, in a sense yes. Just not in a useful sense when it comes to thinking about AI. It's truly trivial to model trinary in binary.
Human minds are holographic. If we made computers that operated on holographic principals we would indeed get closer to reproducing the human mind. However, being like humans is not a requirement for being intelligent.
I really doubt that minds are holographic. (That has a very specific meaning relating the bulk information content to the information presented on the surface, but that seems irrelevant here.) What minds are is very intimately tied up with processing the body and, crucially, implemented on top of a self-modifying substrate. Whether or not a mind is like software (or perhaps more accurately a running program), it's definitely not like any program any sane programmer ever wrote.
You get more than enough complexity with just "self-modifying" without having to invoke the more profound strangenesses of "quantum computing" or "holographic information".
I wonder how much of that money generated by the government, which it doesn't need, as it's obviously not spending more than it gets from taxes, will be distributed to each citizen.
It's the Tories, so what do you bet that it will be divided "fairly" among the "deserving" citizens.
However it is still annoying that we still have no operator overloading.
It's very easy to go horribly off the rails if you start overloading operators (or declaring completely new ones!) so if you're going to propose it, for goodness' sake do it by requiring people to implement a proper mathematical structure (e.g., a group, a ring or a field).
you have no idea what the code does when you look at it.
You've got that problem in C++ too, except that's with thanks to the fun world of custom operators and the world's most confusingly complex template system. The biggest single issue with Java is that it's officiously bureaucratic to the point of insanity. It does make it relatively easy to read someone else's code though, which isn't a charge often levied against C++ in my experience.
I can confirm that Java "uptakes" are often far over 4 years.
We were stuck on 1.4 for development of one of our flagship applications for ages just in case one of our users hadn't upgraded this millennium. Yuck (especially as 1.4's support for things like XML handling was a Special World Of Pain). Mind you, we're on 1.6 now and are transitioning to 1.7 probably this year; the large ZIP support is the killer feature that makes the upgrade pain worthwhile (YMMV; we're working with rather large data archives).
[Clang] is also has much better diagnostics and c++ support so it is a positive change.
It's also a lot more modular, reputedly (I've not verified that myself), and seems pretty fast to build. Mind you, I'd only really be interested if they were to work more on the stability of the ABIs of critical support libraries. Having to recompile for each version of libstdc++ was really annoying, given that maintaining a whole Linux distribution was never a desired personal goal of mine. Being able to ship the same binary to all users is nice, but even just one version per broad platform is better than what happens all too often.
I think it's a shame that open compiler projects can't share more, even if just their techniques and not the code itself. I can see that the difference in license and philosophies behind them would push the camps apart though.
Unless they chose numbers that they could break, but which other country's agencies who also independently discovered differential analysis might not.
That's a very unsafe assumption. It's always wise to assume that if you know an algorithmic weakness, others can find it too (either by independent discovery or by espionage). Even having a master key that can override individual secured systems is unsafe, even though keys are not themselves generally regarded as weaknesses, as it is incredibly valuable to search for what that key is and pretty easy to verify that it is correct once found.
The NSA's principle MO seems to be derived from signals intelligence though, i.e., working out what people are talking about without actually having the text of what they are saying. Collect enough of the graph of datapoints and you can know what is going on even without the knowing all the labels attached to those points...
Ah, so we've looked through the source code and hardware design to verify that's the case?
You want the fingerprint data kept locally so that you can let the authorised user unlock the phone even when the network isn't present. That's a very strong design constraint. (You can't easily verify whether the data goes anywhere else as well but that's going to be hard to do with any approach. Are you 100% sure that your password isn't sometimes sent elsewhere? I know I'm not, even when I'm using Linux; some PAM scripts are very opaque...)
Slashdot Mirror
We're told it's all supposed to go through change control
It does! But they're on different branches and never merge or rebase...
The only accent-free English is BBC English.
Even that's not true. They have a very particular accent that is only used for presenting the news (their other output has much more variation) which is a clarified variation on University English, but it's there if you listen for it. I suppose you could consider it to be at the centre of a cluster of accents that come under the heading of British English.
Similarly, there's another cluster of accents that constitute American English. The spread on the cluster is tighter than on the BE cluster, as far as I can tell. Arguably, you could describe an accent as being an offset from the barycenter of the cluster. You might not think you've got an accent, but from the perspective of someone in another cluster, you most certainly have. (Confusingly, the clusters may overlap somewhat. Yay for complexity!)
No mention of Shock waves, or even a hint of what might cause such shock, or how such shock could be transmitted in the vacuum of space.
Via the interstellar medium, of course. It's pretty tenuous, but most certainly is capable of sustaining phenomena like shock waves. Which isn't to say that that's necessarily the particular process that is dominant in the galactic arms; it could also be something relating to magnetism, as the physics of a flowing magnetically-coupled medium is viciously difficult to work with (i.e., highly non-linear). And I've got no idea what happens at the phase change boundaries between the parts of the ISM which are plasmas and the parts which are conventional (tenuous) gasses; phase changes can do "interesting" things.
As for what's powering it all, you've got some exceptionally powerful energy sources out there. Black holes in particular can pump vast amounts of energy into the surrounding volume of space. The stellar wind from very high mass stars would be another interesting source.
Does ride-sharing really need to be regulated beyond a requirement that the vehicles and drivers have proper insurance?
Is it a taxi service — the taking of a person to a place they nominate in return for a fee — or not? If it is, there is a need for some regulation. An example of the kind of reasonable regulation is to require that nobody with a conviction for a sex crime should be able to be a taxi driver. Having to have a particular level of insurance (or better) is another reasonable regulation. It's all about ensuring that the services that are there are not actively hazardous for people to use.
The problem with services such as the one in the article is that they're trying to offer a taxi service while saying "I promise I'm not a taxi service" and keeping their fingers crossed behind their back. That's basically dishonest. Of course, it's a response to the over-regulation in some locales where the real taxis have actually effectively been organised as a cartel. Which is also wrong. Two wrongs do not make a right; they just layer wrong on wrong.
What's really needed is to pare back the regulation to what is actually needed and to end the medallion system as it has ended up, or at least end the limits on how many are issued and who can have them. That will then also let you ban subcontracting of these things to people who may not have passed background checks, etc. Fix your regulation, you fools!
I must be missing something about this concept. If you're getting paid (with a net profit) to drive people around, why is it called ride sharing? How is it not a taxi service?
They're probably trying to work around some regulations. In the UK it's pretty clear: if it's done for hire, it's one of two types of taxi. A "hackney carriage" is what most people think of as a taxi: it's one you can hail in the street. There's also a "private hire" taxi, where you have to have some sort of prior arrangement before being picked up; it covers a whole range of options (up to a maximum size of vehicle and excluding a few more highly regulated categories) such as limousines and dial-a-rides. If you charge to take someone somewhere, you're covered by the regs. Only genuine ride-sharing or hitching a lift, i.e. stuff which people don't consider hiring, would not be regulated.
Sounds like what you need are a similar sort of thing. There's nothing wrong with having "hire a taxi over the internet" services; they work well. They just shouldn't pretend to be anything other than what they are.
linode.com does month to month, prorates per day. disclaimer - very long time customer, very happy.
And the point of a Cloud is that you can buy as little as an hour, possibly even less with some services. That really changes how you use things, and the sort of business you do with it, as there's a lot of behaviour that varies on that scale much more than it does at the level of months. It's also cheaper if you only need a few hours of processing occasionally; that applies to a lot of things.
Of course, you can combine things. You might put your persistent control systems at fixed IP addresses in linode, but spin up workers in a Cloud as required. Mind you, you can do the same thing inside Clouds using a cheap low-resource instance at a fixed address. Working out the best combination overall is non-trivial, especially in abstract.
I would really like this feature for normal Java applications that use the JVM to get around the firewall.
What? That doesn't make any sense, not unless you're talking about basing whether code can get through the firewall on the path to the executable or something equally silly (given the existence of the JVM, Python, Ruby, Perl, ...).
Does this mean that the NSA is PCI Compliant?
Of course. Admittedly it is by definition, and the part where it says they are is secret so you're not allowed to know about what it exactly involves...
Meanwhile, some 60+ years later and Europe still wasn't making such pencils (well, not commonly enough, anyways).
In my experience (at school, long ago) a separate eraser was better because it did a better job of erasing by virtue of allowing you to use a broader surface for erasing. Convenience is all very well, but I preferred the "convenience" of being able to actually erase pencil markings from the page.
The product page is light on details, but I'd be surprised if that logic wasn't there precisely to negotiate charge rate.
Well something's got to send the data to the NSA...
Please don't.
Just pardon him and stop wasting government time and taxpayer money and frivolous dog and pony shows.
You want a pardon issued to George W. Bush?
Win8 is an interesting ball of wax. We already in the progress of migration to it -- mostly new Dell laptops. It is universally hated by everyone I talk to. People hate it for two reasons:
a) sake of change for the sake of change when there was nothing "wrong" with the old UI
b) Eveyone agrees Metro makes perfect sense on a tablet but screwing over the desktop users pisses off a lot of people because you are forcing them to waste their time and IT's time to relearn how to do the same thing as before. It is a hindrance from us doing our job and we are already overloaded as it is.
Having observed a number of people using Win8, the thing which they really really don't like is the drag-mouse-to-edge to make something invisible pop up that you still need for doing things with the machine. They just don't learn what's going on, and hate it when it happens by accident and doubly hate it when they can't find it when they're looking for it. Setting aside a few pixels so that people can see and be reminded that there is something important there would be by far the best solution, at least on laptops and desktops where there's plenty of screen space.
Windows 3.0 Program Manager supported multiple, overlapping windows.
Strictly, the overlapping windows came in with Windows 2. I had a copy. Win 3.0 was much nicer, if slower and far more bloated. (After all, you didn't need much colour on your screen, did you?)
you're right. what intel does do is an excellent version of frequency scaling, turning off unused execution units, etc. this is better than a dedicated low-power core, because it's finer grained and there's some benefit even without software particilation.
What's it like when you put a real workload on it? Benchmarks are interesting and all, but the proof is always reality. (I mistrust benchmarking because it's so easy to tune things to do well in a benchmark without actually being particularly good on anything else; this has definitely happened in the past with CPU design too.)
Mind you, Intel's main problem is that there's a large and expanding market out there where their CPUs just aren't the things that people choose. People select ARM-based systems because that's what the existing software is written for and what the existing users have, and the benefit of integrating with existing platforms is very strong indeed.
So you can see exactly which politician worked on which law and why.
Because nobody ever told a lie in a commit comment...
Binary++ is not == Trinary.
Well, in a sense yes. Just not in a useful sense when it comes to thinking about AI. It's truly trivial to model trinary in binary.
Human minds are holographic. If we made computers that operated on holographic principals we would indeed get closer to reproducing the human mind. However, being like humans is not a requirement for being intelligent.
I really doubt that minds are holographic. (That has a very specific meaning relating the bulk information content to the information presented on the surface, but that seems irrelevant here.) What minds are is very intimately tied up with processing the body and, crucially, implemented on top of a self-modifying substrate. Whether or not a mind is like software (or perhaps more accurately a running program), it's definitely not like any program any sane programmer ever wrote.
You get more than enough complexity with just "self-modifying" without having to invoke the more profound strangenesses of "quantum computing" or "holographic information".
But you're forgetting - he's been assigned by the PHB to a "more important" part of the program.
But you're forgetting — he's smart so he keeps an eye on what must be done as well as what some failed MBA thinks is required.
I wonder how much of that money generated by the government, which it doesn't need, as it's obviously not spending more than it gets from taxes, will be distributed to each citizen.
It's the Tories, so what do you bet that it will be divided "fairly" among the "deserving" citizens.
very easy to remote wipe iphones
But can you remote wet wipe an iPhone?
However it is still annoying that we still have no operator overloading.
It's very easy to go horribly off the rails if you start overloading operators (or declaring completely new ones!) so if you're going to propose it, for goodness' sake do it by requiring people to implement a proper mathematical structure (e.g., a group, a ring or a field).
you have no idea what the code does when you look at it.
You've got that problem in C++ too, except that's with thanks to the fun world of custom operators and the world's most confusingly complex template system. The biggest single issue with Java is that it's officiously bureaucratic to the point of insanity. It does make it relatively easy to read someone else's code though, which isn't a charge often levied against C++ in my experience.
I can confirm that Java "uptakes" are often far over 4 years.
We were stuck on 1.4 for development of one of our flagship applications for ages just in case one of our users hadn't upgraded this millennium. Yuck (especially as 1.4's support for things like XML handling was a Special World Of Pain). Mind you, we're on 1.6 now and are transitioning to 1.7 probably this year; the large ZIP support is the killer feature that makes the upgrade pain worthwhile (YMMV; we're working with rather large data archives).
[Clang] is also has much better diagnostics and c++ support so it is a positive change.
It's also a lot more modular, reputedly (I've not verified that myself), and seems pretty fast to build. Mind you, I'd only really be interested if they were to work more on the stability of the ABIs of critical support libraries. Having to recompile for each version of libstdc++ was really annoying, given that maintaining a whole Linux distribution was never a desired personal goal of mine. Being able to ship the same binary to all users is nice, but even just one version per broad platform is better than what happens all too often.
I think it's a shame that open compiler projects can't share more, even if just their techniques and not the code itself. I can see that the difference in license and philosophies behind them would push the camps apart though.
Unless they chose numbers that they could break, but which other country's agencies who also independently discovered differential analysis might not.
That's a very unsafe assumption. It's always wise to assume that if you know an algorithmic weakness, others can find it too (either by independent discovery or by espionage). Even having a master key that can override individual secured systems is unsafe, even though keys are not themselves generally regarded as weaknesses, as it is incredibly valuable to search for what that key is and pretty easy to verify that it is correct once found.
The NSA's principle MO seems to be derived from signals intelligence though, i.e., working out what people are talking about without actually having the text of what they are saying. Collect enough of the graph of datapoints and you can know what is going on even without the knowing all the labels attached to those points...
Ah, so we've looked through the source code and hardware design to verify that's the case?
You want the fingerprint data kept locally so that you can let the authorised user unlock the phone even when the network isn't present. That's a very strong design constraint. (You can't easily verify whether the data goes anywhere else as well but that's going to be hard to do with any approach. Are you 100% sure that your password isn't sometimes sent elsewhere? I know I'm not, even when I'm using Linux; some PAM scripts are very opaque...)