It is normal to:
Lock the booze so your kids cannot drink it
Really? My parents left theirs unlocked. I knew where it was, and was allowed to drink small amounts from a fairly early age (in the UK, you can drink alcohol in your own home from the age of 5). I was also taught what would happen if I drank a lot.
Lock the gate, so your kids do not run across the street
There was no gate between my house and the street on the front, and the front door had a yale lock so you just turn the handle to open it. I could do this by about the time I was old enough to run across the street. My parents taught me to always look before running into the street and what would happen if I didn't (i.e. potentially getting hit by a car, not a threat of punishment).
Lock the chemicals (like chlorine) so they do not drink it
Nope, stored in the cupboard under the sink. Easy for me to get to. My parents taught me what the poison and corrosive substance symbols meant by the time I was about 4 or 5.
Lock the medicines
Nope, stored in a bathroom cabinet. Again, I was told what happened if you take medicine for something that you're not suffering from (well, with some exaggerations of the unpleasant effects, but close enough).
I suspect that a lot of the problems in modern society come from children not being taught early on to make informed choices about risk. If I'd wanted to drink a load of booze, bleach, or eat random pills, I was perfectly able to from about the age of 6 or so, but I was taught that the outcome would probably be going to hospital and having my stomach pumped if I was lucky and death if I wasn't. On the Internet, the potential for harm is much lower. No one is actually harmed by seeing some porn (although getting malware installed is a problem for random surfing). The only real danger is if the child agrees to meet random people in person, and that's fairly easy to avoid if you're a moderately attentive parent even if you fail to teach the child that it's a bad idea.
I should add that nothing in the list was particularly unusual for people I knew growing up. Locking up things that might be harmful to children just didn't happen.
If a pedi acting like a kid is sending inappropriate stuff it is nice to know and sadly is quite common if you ask law enforcement agencies
I was a teenager when home Internet access became possible, so I missed out on most of this, but I do remember when I was quite young (5ish) being taught basic safety in that regard: don't go off with people you don't know, if someone claims to be there on behalf of my parents then ask them for a password, and so on. When I got Internet access, I was given similar advice: don't give out your home address to anyone, don't agree to meet anyone in person that you met online, remember that the person you're talking to online might not be the person you think they are, and so on. No need for spyware, just some education.
Close, it's a question of interconnect. We have massively parallel computers already. In the human brain contains about 100 billion neurons, and having that many features on an IC is not hard to imagine although a neuron needs several transistors to emulate because it's triggered by a threshold, not just a binary value.., each neuron is (on average) connected with around 7,000 others. in contrast, efficient connections between transistors are limited by the planar arrangement of the die. This means that it gets really hard if you want them to have connections to more than a small number (on the order of 5-10) of other features.
They're half right. The problem is that most people who use a web browser will never open the preferences panel, so will not know that it's set. If, on first launch, it asked 'do you want to be tracked [yes / no]' then that would make more sense.
The crime rate also dropped during the second world war in the UK. One anthropological explanation for both is that, in general, humans commit (premeditated) crimes against members of a different tribe - that crime is effectively the kind of war that you get when two tribes are living in the same village. Once you identify an external enemy, be it the Germans or the 1%, then the local distinctions become less important and so the number of potential victims goes down.
I suspect that's not very representative of the majority of the Chinese population, who are mostly involved in subsistence farming. I would expect to see atheism concentrated in the urban areas, where there is greater access to education. It's also tricky when it comes to Buddhism, as there are theistic and atheistic variants.
By that definition, a lot of US owned missiles are also designed purely for terrorism, as they include hardware that destroys the control system just before impact to prevent any of the components from being reverse engineered.
That was what I read, although that was third hand with the BBC quoting the Washington Post (I think) quoting an anonymous source, who said that Obama specifically authorised it in his role as Commander in Chief. Generally, attacking a foreign power needs to go quite a long way up the chain of command to get approval. The CIA wouldn't have launched the Bay of Pigs debacle, for example, without approval from the President.
Psion had a really great product with the Series 3 and descendants (3a and 3c), and a pretty good one with the Series 5. The 5 had a really nice keyboard, but relatively poor battery life. Then they more or less lost the plot. The great thing about the Series 3 was that you could keep it in your pocket and use it frequently for a couple of weeks on a pair of AA batteries. The Series 5 needed charging overnight. It also involved a complete rewrite of the OS. EPOC16 was mainly written in 8086 assembly and used segments to give each application its own 64KB address space (most applications ran in under 10KB of RAM). EPOC32 had to run on 32-bit ARM, and was a little bit heavier, partly because the machine had a proper MMU and partially because it could afford to be in a machine with 16 times as much RAM, but it was still a very nice system to use.
Psion then tried to move into larger devices, with the Series 7 and the netBook. Both were 'subnotebooks', fitting squarely into the market where they're as unportable as a laptop and as underpowered as a handheld. Oddly enough, they were not a commercial success.
EPOC32, however, was. Rebranded Symbian, it was licensed to pretty much all mobile phone makers and was the dominant platform for high-end mobile phones for some years. Until the launch of the iPhone, it had about 70% of the Smartphone and Featurephone markets, although both iOS and Android (combined with spectacular mismanagement by Nokia) have eroded it to almost nothing now.
Psion sold off the hardware part to focus on Symbian, because their later hardware sucked - the netBook part of the acquisition was irrelevant to the company that bought them for their modem hardware, so it was largely killed off. Nokia bought the Symbian part later because most of their products were based on it.
It depends on the enforcement and wording. For example, the DMCA only protects 'effective' countermeasures. One argument that has been made is that any countermeasure that has been compromised is no longer effective, which would mean that the act of breaking DRM would demonstrate that it was not effective and therefore not protected by the DMCA. If the Canadian version has similar wording and courts uphold this interpretation, then it's a win...
Your examples are somewhat flawed, because a lot of projects do exactly that. In Clang, for example, which I think gets better parsing performance than anything else around, small objects are allocated with a bump-the-pointer allocator with a per-AST scope, and then are freed by just freeing the whole allocation at once. This is entirely possible to do with manual memory management, because you have complete control over allocation policy. With automatic garbage collection, the collector has to infer lifetime, the developer can't explicitly state it.
Automatic GC, however, deals a lot better with aliasing, cyclic data structures, and especially with concurrency.
Erlang is closer to what Alan Kay meant when he coined the term "object oriented" than pretty much any other language has yet realised
I'm not sure it's still there, but there used to be a thing on the Erlang web page that said that Erlang was 'not an object oriented language like...' and then listed a number of other languages that were all less object oriented than Erlang.
Garbage collection is also important for concurrency. Explicitly managing the lifetime of objects goes from being hard to being almost impossible once you thrown in concurrency.
Really? It was originally written for the PDP-11 and I've run TeX on SPARC, PowerPC, i386 and x86-64. I doubt there's anything sufficiently unportable to fail to run on ARM. You can install a Debian chroot on Android, and there are TeXLive packages for Debian/ARM, so I don't see what the problem is...
The mobile graphics space has been dominated by one player: PowerVR. ARM and nVidia are more recent entrants. AMD doesn't yet have anything in this space, although that will change very soon.
The result I'm guessing is that most compilers are written to use SYSCALL and SYSRET rather than SYSENTER and SYSEXIT
Compilers never emit either instruction, because the source languages don't provide a system call mechanism. This code appears in hand-written assembly stubs in libc. A typical libc will contain multiple versions of the system call function for x86 (SYSCALL, SYSENTER, int 80h) and will install the correct one depending on the CPUID results.
Doesn't it do it automatically? With FreeBSD, there is, by default, a daily security report emailed to the root user - you can forward this to another account if you want. This includes things like login failures and installed software with known vulnerabilities (if you have portaudit installed).
RAM is volatile. No matter how much RAM you have, you need to flush writes to persistent storage periodically and if seeks are expensive then that's quickly going to become a bottleneck. Operating systems generally try to flush files to disk fairly quickly so that power failures won't result in losing much data.
If you want the update to be visible to multiple threads, it still needs to go via the cache coherency protocol. Whether you're doing a register-memory add or a register-register add and a store makes no difference.
Speaking as a compiler writer who works mainly on optimisation: we do the same size optimisations for ARM and x86. All of that stuff happens in the processor-independent part of the compilation pipeline. Anything that benefits x86 in this regard is likely to benefit ARM equally.
ARMv7 means anything with Cortex in its name. That's basically any ARM chip that you're likely to encounter these days, unless you are doing really low-end embedded development.
The Raspberry Pi has been described as "a smartphone minus the screen". It's $25-$35. A smartphone is in the range of $300-$600. Order of magnitude difference, and that's not because of the processor
The Raspberry Pi has a 700MHz ARM11 core. A modern Smartphone has at least a 1GHz+ Cortex A8 core. The ARM11 was introduced in 2002. This is a really old design, the sort that you find in ultra-low-end $100 Android tablets.
Slashdot Mirror
Someone who is able to use the Internet is, by definition, old enough to be taught the dangers...
It is normal to:
Lock the booze so your kids cannot drink it
Really? My parents left theirs unlocked. I knew where it was, and was allowed to drink small amounts from a fairly early age (in the UK, you can drink alcohol in your own home from the age of 5). I was also taught what would happen if I drank a lot.
Lock the gate, so your kids do not run across the street
There was no gate between my house and the street on the front, and the front door had a yale lock so you just turn the handle to open it. I could do this by about the time I was old enough to run across the street. My parents taught me to always look before running into the street and what would happen if I didn't (i.e. potentially getting hit by a car, not a threat of punishment).
Lock the chemicals (like chlorine) so they do not drink it
Nope, stored in the cupboard under the sink. Easy for me to get to. My parents taught me what the poison and corrosive substance symbols meant by the time I was about 4 or 5.
Lock the medicines
Nope, stored in a bathroom cabinet. Again, I was told what happened if you take medicine for something that you're not suffering from (well, with some exaggerations of the unpleasant effects, but close enough).
I suspect that a lot of the problems in modern society come from children not being taught early on to make informed choices about risk. If I'd wanted to drink a load of booze, bleach, or eat random pills, I was perfectly able to from about the age of 6 or so, but I was taught that the outcome would probably be going to hospital and having my stomach pumped if I was lucky and death if I wasn't. On the Internet, the potential for harm is much lower. No one is actually harmed by seeing some porn (although getting malware installed is a problem for random surfing). The only real danger is if the child agrees to meet random people in person, and that's fairly easy to avoid if you're a moderately attentive parent even if you fail to teach the child that it's a bad idea.
I should add that nothing in the list was particularly unusual for people I knew growing up. Locking up things that might be harmful to children just didn't happen.
If a pedi acting like a kid is sending inappropriate stuff it is nice to know and sadly is quite common if you ask law enforcement agencies
I was a teenager when home Internet access became possible, so I missed out on most of this, but I do remember when I was quite young (5ish) being taught basic safety in that regard: don't go off with people you don't know, if someone claims to be there on behalf of my parents then ask them for a password, and so on. When I got Internet access, I was given similar advice: don't give out your home address to anyone, don't agree to meet anyone in person that you met online, remember that the person you're talking to online might not be the person you think they are, and so on. No need for spyware, just some education.
Close, it's a question of interconnect. We have massively parallel computers already. In the human brain contains about 100 billion neurons, and having that many features on an IC is not hard to imagine although a neuron needs several transistors to emulate because it's triggered by a threshold, not just a binary value.., each neuron is (on average) connected with around 7,000 others. in contrast, efficient connections between transistors are limited by the planar arrangement of the die. This means that it gets really hard if you want them to have connections to more than a small number (on the order of 5-10) of other features.
They're half right. The problem is that most people who use a web browser will never open the preferences panel, so will not know that it's set. If, on first launch, it asked 'do you want to be tracked [yes / no]' then that would make more sense.
The crime rate also dropped during the second world war in the UK. One anthropological explanation for both is that, in general, humans commit (premeditated) crimes against members of a different tribe - that crime is effectively the kind of war that you get when two tribes are living in the same village. Once you identify an external enemy, be it the Germans or the 1%, then the local distinctions become less important and so the number of potential victims goes down.
I suspect that's not very representative of the majority of the Chinese population, who are mostly involved in subsistence farming. I would expect to see atheism concentrated in the urban areas, where there is greater access to education. It's also tricky when it comes to Buddhism, as there are theistic and atheistic variants.
By that definition, a lot of US owned missiles are also designed purely for terrorism, as they include hardware that destroys the control system just before impact to prevent any of the components from being reverse engineered.
That was what I read, although that was third hand with the BBC quoting the Washington Post (I think) quoting an anonymous source, who said that Obama specifically authorised it in his role as Commander in Chief. Generally, attacking a foreign power needs to go quite a long way up the chain of command to get approval. The CIA wouldn't have launched the Bay of Pigs debacle, for example, without approval from the President.
Psion had a really great product with the Series 3 and descendants (3a and 3c), and a pretty good one with the Series 5. The 5 had a really nice keyboard, but relatively poor battery life. Then they more or less lost the plot. The great thing about the Series 3 was that you could keep it in your pocket and use it frequently for a couple of weeks on a pair of AA batteries. The Series 5 needed charging overnight. It also involved a complete rewrite of the OS. EPOC16 was mainly written in 8086 assembly and used segments to give each application its own 64KB address space (most applications ran in under 10KB of RAM). EPOC32 had to run on 32-bit ARM, and was a little bit heavier, partly because the machine had a proper MMU and partially because it could afford to be in a machine with 16 times as much RAM, but it was still a very nice system to use.
Psion then tried to move into larger devices, with the Series 7 and the netBook. Both were 'subnotebooks', fitting squarely into the market where they're as unportable as a laptop and as underpowered as a handheld. Oddly enough, they were not a commercial success.
EPOC32, however, was. Rebranded Symbian, it was licensed to pretty much all mobile phone makers and was the dominant platform for high-end mobile phones for some years. Until the launch of the iPhone, it had about 70% of the Smartphone and Featurephone markets, although both iOS and Android (combined with spectacular mismanagement by Nokia) have eroded it to almost nothing now.
Psion sold off the hardware part to focus on Symbian, because their later hardware sucked - the netBook part of the acquisition was irrelevant to the company that bought them for their modem hardware, so it was largely killed off. Nokia bought the Symbian part later because most of their products were based on it.
It depends on the enforcement and wording. For example, the DMCA only protects 'effective' countermeasures. One argument that has been made is that any countermeasure that has been compromised is no longer effective, which would mean that the act of breaking DRM would demonstrate that it was not effective and therefore not protected by the DMCA. If the Canadian version has similar wording and courts uphold this interpretation, then it's a win...
Your examples are somewhat flawed, because a lot of projects do exactly that. In Clang, for example, which I think gets better parsing performance than anything else around, small objects are allocated with a bump-the-pointer allocator with a per-AST scope, and then are freed by just freeing the whole allocation at once. This is entirely possible to do with manual memory management, because you have complete control over allocation policy. With automatic garbage collection, the collector has to infer lifetime, the developer can't explicitly state it.
Automatic GC, however, deals a lot better with aliasing, cyclic data structures, and especially with concurrency.
Erlang is closer to what Alan Kay meant when he coined the term "object oriented" than pretty much any other language has yet realised
I'm not sure it's still there, but there used to be a thing on the Erlang web page that said that Erlang was 'not an object oriented language like...' and then listed a number of other languages that were all less object oriented than Erlang.
Garbage collection is also important for concurrency. Explicitly managing the lifetime of objects goes from being hard to being almost impossible once you thrown in concurrency.
LaTeX won't work on ARM
Really? It was originally written for the PDP-11 and I've run TeX on SPARC, PowerPC, i386 and x86-64. I doubt there's anything sufficiently unportable to fail to run on ARM. You can install a Debian chroot on Android, and there are TeXLive packages for Debian/ARM, so I don't see what the problem is...
The mobile graphics space has been dominated by one player: PowerVR. ARM and nVidia are more recent entrants. AMD doesn't yet have anything in this space, although that will change very soon.
The result I'm guessing is that most compilers are written to use SYSCALL and SYSRET rather than SYSENTER and SYSEXIT
Compilers never emit either instruction, because the source languages don't provide a system call mechanism. This code appears in hand-written assembly stubs in libc. A typical libc will contain multiple versions of the system call function for x86 (SYSCALL, SYSENTER, int 80h) and will install the correct one depending on the CPUID results.
Doesn't it do it automatically? With FreeBSD, there is, by default, a daily security report emailed to the root user - you can forward this to another account if you want. This includes things like login failures and installed software with known vulnerabilities (if you have portaudit installed).
RAM is volatile. No matter how much RAM you have, you need to flush writes to persistent storage periodically and if seeks are expensive then that's quickly going to become a bottleneck. Operating systems generally try to flush files to disk fairly quickly so that power failures won't result in losing much data.
Yes, spending money on something that makes you more productive, what kind of idiot would do that?
-pipe has been the default for, what, 10 years now?
If you want the update to be visible to multiple threads, it still needs to go via the cache coherency protocol. Whether you're doing a register-memory add or a register-register add and a store makes no difference.
Speaking as a compiler writer who works mainly on optimisation: we do the same size optimisations for ARM and x86. All of that stuff happens in the processor-independent part of the compilation pipeline. Anything that benefits x86 in this regard is likely to benefit ARM equally.
ARMv7 means anything with Cortex in its name. That's basically any ARM chip that you're likely to encounter these days, unless you are doing really low-end embedded development.
The Raspberry Pi has been described as "a smartphone minus the screen". It's $25-$35. A smartphone is in the range of $300-$600. Order of magnitude difference, and that's not because of the processor
The Raspberry Pi has a 700MHz ARM11 core. A modern Smartphone has at least a 1GHz+ Cortex A8 core. The ARM11 was introduced in 2002. This is a really old design, the sort that you find in ultra-low-end $100 Android tablets.