You won't think it's such a good idea when you're strapped to a gurney getting wheeled to the Place of Happy Release because you said something bad about Obama.
If you drive along a motorway in the UK there's a nominal 70 mph speed limit and you'll find you have the following lanes.
Inside lane : lorries limited to 70mph, cars that can't do 70mph, etc
Middle Lane: Most people will do about 80 mph on the speedometer. The speedometers in UK are marked down by 5% so this is really 66 mph. It is rumoured that police won't prosecute people they catch doing this speed in a 70mph limit. I do this and pull into the inside lane if it is empty.
Outside Lane: People do 90-100mph. The police will prosecute at these speeds. Still presumably they have speed trap detectors or a GPS app that warns them. Or they've done something illegal like register the car to someone who doesn't exist.
Now this works fine except for something I call Wanker Driver Syndrome or WDS. A driver with WDS will whizz up behind you in the inside lane in a BMW or Mercedes doing 90-100mph if the inside lane is clear and slam his breaks on to get back to 80mph. He will hang on your tail for a while with his lights on main beam to fuck up your dark adjusted vision and then - before you have a chance to change lane - whizz off in the outside lane which is always clear in this situation. His (and it's always a he) reasons for doing this are obscure since he could just stay in the outside lane. Or drive at 80mph like everyone else, Or switch to the outside lane when he wants to overtake.
It's completely different from the US where everyone sets their cruise control to the speed limit and leaves it at that. WDS suffers in the US get jailed or shot or something.
So the benefit to the laser lights is safety for said WDS sufferer. You'll get a blast of light long before he has to slam the breaks on. It's very much not to the benefit of people who don't have WDS. I.e. the rest of us.
If the machine next to your hospital bed displays a laughing skull and starts playing mod tunes whilst demanding you pay by credit card to an account in Russia to avoid being "pwned by l33tgr0up" that is likely not a good sign.
Don't say that. Imagine if the Russians had said that to Gorbachev when announced Glasnost. Our overlords have announced a period of openness and we should encourage it. Otherwise we'll be back in the gulag on meager rations of Jon Katz stories with a page that takes 10 minutes to load and looks like ass.
Though I'll say one for their attempts at a redesign - they may have looked like ass but it looked exactly the same ass whether you use Chrome, Opera, Firefox or IE. That's technically pretty impressive in that their horrible unusable CSS and Javascript monstrosity was browser independent. They must have tested it on each browser very carefully in order to make sure they same - and admittedly tortuous - user experience was had on each one.
Come to think of it slashdot's decision to embrace Glasnost is probably happening for much the same reasons as the USSR's if you look at the way the number of comments have dropped over the last few years.
It's probably because of low wages too. I remember when I was in Malaysia I was with someone who dropped off some wages for Indonesians working in a Malaysia factory. Now I worked out the hourly rate and when I went back to Taiwan I mentioned it to people who know about things and they were surprised it was lower than someone doing the same job in China.
So a lot of Taiwanese companies like Foxconn are adopting a "China+1" strategy for manufacturing, i.e. factories in China plus one other lower wage country in Asia. Indonesia seems like a good bet for this.
Actually another thing that's interesting is that in places like Indonesia most people have a phone. Smartphone penetration is lower but it is growing fast
There are a lot places with more smartphones per capita than the US (56%). And a lot of them are in Asia - Hong Kong(62%), South Korea(73%), Singapore(71%) being the obvious ones. So the odds are that most Indonesians will buy a smartphone sooner or later. Importing stuff in Asia is a nightmare because of duties and bureaucracy. So one way around that is to make things in country instead of importing them. Foxconn are a contract manufacturer so they make things for other people's brands. It may well be that those brands think that manufacturing in Indonesia is a good bet because of a mix of low costs and the fact that getting a "Made in Indonesia" stamp on the device means they can avoid import duty on the devices they sell there. Also most likely they can get some assistance from the government given that they're building a factory which can employee a lot of people.
Another thing is that Indonesia may have its issues but it is probably easier to get your profits back from Indonesia to Taiwan. I've talked to people in Taiwan who've pointed out that doing that from China is non trivial. The Chinese RMB for example isn't convertible but the Indonesian Rupiah is.
Also no ARM processor shipping has the horsepower to do a good job at emulating even a low end x86. I think even a really good JIT engine running on the fastest ARM you can get is going to be slower at running x86 code than even the slowest x86.
Back when Dec was trying to get people to buy Alphas they wrote a really clever piece of software called FX!32. It would take an x86 binary and initially run it slowly via instruction by instruction emulation. However while doing that it would profile the application and use the information it gained from that to work out which bits were worth JITting to native Alpha code. Since at that point the fastest Alpha was faster than x86 it ended up running JITted code faster too. So at one point the fastest way to run x86 code was in FX!32 on Alpha.
Now that was possible because of that speed advantage. However really everything comes down to power consumption. Alpha was a server processor and burnt watts prodigiously even compared to Intel. It was also a very clean design and just at the point where Risc was a very efficient design philosophy - the fastest Alpha run about twice the speed of the fastest x86 and the ISA was so recent that it could be implemented efficiently. As time went by it seemed like Risc seemed less of a net win - e.g. Risc chips with 32 bit fixed length instructions had worse code density than x86 and x86 chips moved to a design where x86 opcodes were decoded into Risc like uops internally.
Now with ARM it's not like that. ARM cores are designed for mobile devices and are brilliant for low power and small core size. However there aren't people selling ARMs that are faster than the x86. In fact in general the fastest ARM has run neck and neck with the slowest x86.
So most of those names were meant to be there, it is right for them to be published, it is right to publish the names of politicians, generals bureaucrats, etc, who are involved in this sort of activity, it is right even to publish the names of corrupt radio stations in Kabul that were taking SYOPS programme content. It is also right to publish the names of those people who have been killed and murdered and who need to be investigated and it is right to publish the names of all incidental characters who themselves are not at serious and probable risk of physical harm. Those incidental characters are someone who owns a company for example is just involved in shipping operations.... So then there is the question were there any sort of villagers or so on who gave information that might lead to reprisals, were there some of those? Um there were some villagers who - who had given information, um so that is a regrettable oversight, but it is not our, not merely our oversight it was the oversight of the United States military who should've never included that material and who falsely classified it, and who then made it available to everyone and it then got out."
Bummer for those "villagers" who opposed the Taliban. Or for anyone who operated a radio station that was pro government and anti Taliban.
So he was in favour of publishing names. And he did too, after a Twitter poll of his followers. Up to that point there was no evidence that Leigh publishing the password had caused the unredacted cables to become generally available.
The Guardian book revealed the diplomatic files were placed by WikiLeaks on a secure online server in July 2010, which it was agreed would only be online for a matter of hours.
This server held a heavily encrypted file containing the unredacted embassy cables database. Assange had given Leigh the password to unlock this file once he had obtained it, and this password was included in the book - seven months after the temporary file was taken offline. No trace could be found through web links or Google's archives of this file ever being visible through this secure server.
However, at a later stage the same encrypted file and at least one other encrypted with the same password was posted on the peer-to-peer file-sharing network BitTorrent. One of these files was first published on 7 December 2010, just hours before Assange's arrest. In the days running up to his arrest, Assange had spoken of "taking precautions" in the event of anything untoward happening to him.
This file, it was later discovered, was the same file that had been shared with the Guardian via the secure server. It shared the same file name and file size, and could be unlocked using the same password as that given to Leigh.
Daniel Domscheit-Berg, a former member of staff at WikiLeaks who is attempting to set up a rival whistleblowing website, discovered this republished file and shared information on WikiLeaks's security breach with a small group of journalists.
Avoiding the re-use of passwords and avoiding republishing temporary files are both considered basic security procedures among online security experts.
However, the file was not discovered or downloaded by the public. By 10am on Thursday it had been accessed once in the previous 31 days, despite mounting speculation about its existence.
Initial news stories did not give details of the location of files or of passwords. Later, WikiLeaks and some of its supporters published a series of hints about the passwords and files.
At about 11pm on Wednesday an anonymous Twitter user discovered the published password and opened a separate file - not the one shared with the Guardian - that had also been circulating on file-sharing networks for several m
This language makes me wonder whether the Guardian was facing an "official direction" for the return or disposal of the material under section 8(5) of the Official Secrets Act 1989.
It would be an offence under section 6(2) of the Act for the Guardian to knowingly make a damaging disclosure of any information, document or other article which (section 6(1)(a))
(i) relates to security or intelligence, defence or international relations; and
(ii) has been communicated in confidence by or on behalf of the United Kingdom to another State...
and (also section 6(1)(a))
has come into a person's possession as a result of having been disclosed (whether to him or another) without the authority of that State...
Documents leaked by Edward Snowden about the work of GCHQ must I think fall within the scope of section 6(1), having presumably been communicated in confidence by the UK intelligence agencies to another state, the US, and having come into the Guardian's possession without US authority.
If that's right, then, as I've said, the Guardian and its editor would risk committing an offence if it published any of that information which was "damaging". By the interaction of section 6(4) and section 1(4)(a), by the way, disclosure of security or intelligence information would be "damaging" if (section 1(4)(a))
it causes damage to the work of, or of any part of, the security and intelligence services
In those circumstances, section 8(5) would apply. It says
Where a person has in his possession or under his control any document or other article which it would be an offence under section 6 above for him to disclose without lawful authority, he is guilty of an offence if he fails to comply with an official direction for its return or disposal.
This all dates back to when secret documents were not digital - e.g. paper or microfilm. If you had them you'd could be directed to "return or dispose[destroy]" them. And if you failed you could be prosecuted.
Incidentally if you or I rather than the Guardian were doing this the consequences would likely be much more drastic. The police would seize all your computers and get a court order to get access to all your offsite backups. In fact this is what happens when people steal data from their employers let alone from the NSA/GCHQ.
In the case of the Guardian it seems like they've gone through this charade as the minimum they can legally do given that the Guardian has told them that copies of the data exist in the US.
Interviewer: "So come on, redactions are going on at the same time, now there is or isn't a row going on about redaction, I haven't the faintest clue whether there is or isn't...?
Mr Assange: No, there's no row going on about redactions at all....There was a group of reports where although they were not really intelligence informants there were sort of hotline tips...something called threat reports comprised one in five of the Afghan War Logs and so we held them back for a line by line redaction...But what we didn't do was redact one in five lines, putting black marker through it, we just removed them, and so it looked like we hadn't redacted everything but in fact we had redacted a fifth of all material, and this permitted an attack, a political attack, to come from The Times of London.... So The Times did a proxy war on The Guardian through us by attacking us.... So most of those names were meant to be there, it is right for them to be published, it is right to publish the names of politicians, generals bureaucrats, etc, who are involved in this sort of activity, it is right even to publish the names of corrupt radio stations in Kabul that were taking SYOPS programme content. It is also right to publish the names of those people who have been killed and murdered and who need to be investigated and it is right to publish the names of all incidental characters who themselves are not at serious and probable risk of physical harm. Those incidental characters are someone who owns a company for example is just involved in shipping operations.... So then there is the question were there any sort of villagers or so on who gave information that might lead to reprisals, were there some of those? Um there were some villagers who - who had given information, um so that is a regrettable oversight, but it is not our, not merely our oversight it was the oversight of the United States military who should've never included that material and who falsely classified it, and who then made it available to everyone and it then got out."
Assange never wanted to redact but was forced his media partners. Then he published the full unredacted cables on wikileaks' website. Which they denounced
In a joint statement, the Guardian, El Pais, New York Times and Der Spiegel said they "deplore the decision of WikiLeaks to publish the unredacted state department cables, which may put sources at risk".
And before you mention the password that appeared in David Leigh's book that was supposed to be for a temporary copy of the archive
WikiLeaks claimed its disclosure was prompted after conflicts between Assange and former WikiLeaks associates led to one highlighting an error made months before. When passing the documents to the Guardian, Assange created a temporary web server and placed an encrypted file containing the documents on it. The Guardian was led to believe this was a temporary file and the server would be taken offline after a period of hours.
However, former WikiLeaks staff member Daniel Domscheit-Berg, who parted acrimoniously with WikiLeaks, said instead of following standard security precautions and creating a temporary folder, Assange instead re-used WikiLeaks's "master password". This password was then unwittingly placed in the Guardian's book on the embassy cables, which was pu
You are also forgetting that a x86 or a amd64 is a RISC cpu with a layer od CISC hidding the RISC. That layer takes cpu space, power and resources (both designing and working).
Look at a die shot of a x86 CPU. Most of the space is cache. The actual CPU is a small fraction. Sure on an embedded system where you care about power and die area ARM makes sense. Switching a desktop or server class machine to ARM suffers from diminishing returns because the CPU is already such a small percentage of die area.
So far, the LZSS routine in my code looks like this:
THUMB2: 76 bytes
THUMB: 76 bytes
ARM64: 96 bytes
ARM32: 116 bytes (for comparison, x86 is 63 bytes)
Still it is quite a bit worse than x86. x64 incidentally is the same code size as x86
. I mean, the instruction set has specialized instructions for handling packed decimal! And then there's the near worthless string search REPNE CMPSB family of instructions. The Boyer-Moore string search algorithm is much faster, and dates back to 1977. Another sad thing is that for some CPUs, the built in DIV instruction was so slow that sometimes it was faster to do integer division with shifts and subtracts. That's a serious knock on Intel that they did such a poor job of implementing DIV. A long time criticism of the x86 architecture has been that it has too few registers, and what it does have is much too specialized. Like, only AX and DX can be used for integer multiplication and division. And BX is for indexing, and CX is for looping (B is for base and C is for count you know-- it's like the designers took their inspiration from Sesame Street's Cookie Monster and the Count!) This forces a lot of juggling to move data in and out of the few registers that can do the desired operation. This particular problem has been much alleviated by the addition of more registers and shadow registers, but that doesn't address the numerous other problems. Yet another feature that is obsolete is the CALL and RET and of course the PUSH and POP instructions, because once again they used a stack. Standard thinking 40 years ago.
It was standard on the 8086 (introduced in 1978). The 80368 (1985) is a general purpose register machine and can use a 0:32 flat memory mode. And modern x64 (2003) has twice as many registers and the ABI specified SSE for floating point, not 8087. Also in 64 bit mode segment bases and limits for code and data (i.e any instruction which does not have a segment override prefix) are ignored.
I.e pretty much all the things you're complaining about have been fixed and if you look at benchmarks x64 chips have been faster than their Risc competitors pretty much since x64 was introduced. Going on about segments, floating point stacks and REPNE MOVSD now is absurd.
And if you look at the way the 8086 took over from the 8080 what they did made a lot of sense. You could machine convert a CPM 8080 program to a MS DOS 8086 one and have it run fine. Meanwhile a native 8086 program had access to 1M of address space, up from 64K on 8080 and Z80. Like CPM MSDOS ran on commodity hardware which was cheaper than the big iron boxes and given most people were running things like Visicalc and Wordstar it was perfectly sufficient.
Rooftop solar is several times more dangerous than nuclear power and wind power. It is still much, much safer than coal and oil, because those have a lot of air pollution deaths.
And a police officer has the technical capacity to walk into my house and shoot me dead. That I can appreciate his likely skill with a service revolver doesn't mean he gets to shoot me dead at a whim.
Right but you accept the fact that the police need to have the capability to shoot people, right? Because if you were an armed robber or something they'd need to be able to do that to stop you.
Similarly the NSA needs to have the capability to spy on people - terrorists, Russian or Chinese spies, or - if WWIII starts - Russian or Chinese soldiers are all people the NSA needs to be able to spy on. In fact it's highly irritating when people who tweet their every thought and bowel movement whine about this. The NSA aren't going to spy on them because a) everything they think is public and b) they're not interesting to anyone, let alone the NSA. Logically given limited resources it's more likely that people like the Boston Bombers are the target of surveillance than people someone memorably referred to as 'twitter cunts'.
If you look at WWII Anglo American SIGINT like breaking the Enigma code was absolutely vital to the war effort and saved the UK from defeat. As China moves towards parity with the West and confronts Japan over the Senkakus it's not impossible the US may find itself in a similar situation. In the long run it's not impossible that Russia will threaten the Ukraine militarily - after all it did more than threaten Georgia.
And in fact having a major SIGINT advantage over Russia and China is likely to act as a deterrent on them doing something like this. Conversely Snowden visiting both and telling them the US's capabilities is likely to make them think they're the ones with the advantage.
The only reason you'd think Snowden did the right thing is if you think the US is the sole source of evil in the world and Russia and China are both governed by people who act robotically in the best interests of humanity eschewing any personal gain. How likely is it really that the people who govern the US are the only ones vulnerable to corruption and the far less open political systems of Russia and China magically produce incorruptible leaders?
I'd say as bad as the US's politicians are the openness of the system means they are likely a lot less bad than those in China or Russia. In which case I'd rather the US has the SIGINT advantage. Snowden did exactly the wrong thing in taking US secrets to Russia and China and the Guardian is wrong to publish US secrets.
If someone wants to spend more to be at the forefront, why not congratulate them and allow the march of progress continue?
Because they're doing enough of that congratulation themselves whilst whizzing past you in the carpool lane or while making passive aggressive comments about how 'gas guzzlers' should be banned by the government.
My favourite Chinese salesman-ism is, when encouraging you to try some food that looks, sounds or tastes a bit horrible is "...very good for you" said it a suitably serious tone.
E.g.
"Black dog. Very good for you"
Or, if you have a nasty rash from eating the food or breathing the air for an extended period
You can bullshit HTTP. Just learn GET and POST and learn a speech about what 'idempotent' means. Once you explain that most people's eyes glaze over, so you can skip the rest.
With the increase of insect resistance to pesticides we'd going to need some novel countermeasures. I propose widespread deployment of Diatomaceous Earth. Sure some people will die but the rest will evolve resistance.
Now you'll say "That's complete madness. Humans are the ultimate K selected species so they've got no chance of out evolving r selected insects" Ordinarily that would be true but by careful use of mutagens and radiation and encouraging r selected traits like promiscuity and abandoning your kids. I think we can win.
And we need too, to ensure that human civilisation, not bed bug, dominates New York. NOW AND FOREVER.
They are not legally required to so why would they?
Because they can put it up on blog and make money out of ads? Because it works as a loss leader for their other paid services. Because it's cheap to publish these days and if you're going to do the research for your own curiosity why not publish?
Makes me like them a whole lot. Now I'm not really in the market for a US centric dating service but if I were I'd use them. Plus they could always write a book full of this sort of stuff.
Does everyone publish all their data? No of course not. Still the trend is that people increasingly do do it for the reasons mentioned.
I don't know your reference for saying that "the models are broken". In my understanding the models used e.g. in IPCC reports, are quite good.
It is completely unreasonable to dismiss them just because they are not perfect. The proper approach is to study the discrepancies, reason about their possible causes and estimate the effect of the errors on the question you are seeking to answer with the model.
What I don't get is why they don't chuck out the models that are bad, keep the ones that are good and invent ones that are better. Right now it almost seems like they do the opposite - the ones that predict OMG! Runaway Global Warming! get loads of press time. And the ones that don't predict anything too drastic get largely ignored and the people that made them called evil deniers.
Actually that's what denier really means. If you think the world is warming slowly but it's no great problem like the satellite temperature measurements say you're an evil denier. In fact unless you support massive CO2 cuts now on the basis of the most alarmist model you're a denier. I.e. it's really an argument about policy, not whether you think the world will be 0.5 degree C or 1.0 degree C warmer in a hundred years time.
You can see that when geoengineering is brought up. The Royal Society did a study that showed that Sulphate aerosols for example could be used to effect "a reduction of solar input by about 2%" to "balance the effect on global mean temperature of a doubling of CO2" for "total annual cost at 10s of billion dollars". So we don't need to rely on the precautionary principle to tell us we need to cut our CO2 emissions to zero now just in case. We can go on as we are, monitor temperature and build ourselves a planetary thermostat quickly and cheaply if it becomes necessary. Of course the 'cut CO2 now' lobby hate this.
They also hate it when you point out that CO2 emissions in Europe and the US are trending down. China's CO2 emissions are increasing massively. If you want a global CO2 cut you'd need to get China to stop industrializing. Which they won't do
Or of course that the actual satellite measurements of temperature are undershooting all the models - you need to use the adjusted temperature measurements from ground stations. And the adjusted temperature measurements only do that because the past has been getting cooler. Those cavemen better watch out, pretty soon it will be below absolute zero when they are.
What this is really about is that you've got people who'd make a load of money if everyone was forced to by CO2 permits. It's pure rent seeking by them. Or people who know deep down we're all sinners for our materialistic lifestyle and want to force everyone into the modern equivalent of sackcloth and ashes to repent.
Slashdot Mirror
You won't think it's such a good idea when you're strapped to a gurney getting wheeled to the Place of Happy Release because you said something bad about Obama.
If you drive along a motorway in the UK there's a nominal 70 mph speed limit and you'll find you have the following lanes.
Inside lane : lorries limited to 70mph, cars that can't do 70mph, etc
Middle Lane: Most people will do about 80 mph on the speedometer. The speedometers in UK are marked down by 5% so this is really 66 mph. It is rumoured that police won't prosecute people they catch doing this speed in a 70mph limit. I do this and pull into the inside lane if it is empty.
Outside Lane: People do 90-100mph. The police will prosecute at these speeds. Still presumably they have speed trap detectors or a GPS app that warns them. Or they've done something illegal like register the car to someone who doesn't exist.
Now this works fine except for something I call Wanker Driver Syndrome or WDS. A driver with WDS will whizz up behind you in the inside lane in a BMW or Mercedes doing 90-100mph if the inside lane is clear and slam his breaks on to get back to 80mph. He will hang on your tail for a while with his lights on main beam to fuck up your dark adjusted vision and then - before you have a chance to change lane - whizz off in the outside lane which is always clear in this situation. His (and it's always a he) reasons for doing this are obscure since he could just stay in the outside lane. Or drive at 80mph like everyone else, Or switch to the outside lane when he wants to overtake.
It's completely different from the US where everyone sets their cruise control to the speed limit and leaves it at that. WDS suffers in the US get jailed or shot or something.
So the benefit to the laser lights is safety for said WDS sufferer. You'll get a blast of light long before he has to slam the breaks on. It's very much not to the benefit of people who don't have WDS. I.e. the rest of us.
Who do you suppose noticed the breaches, and how?
If the machine next to your hospital bed displays a laughing skull and starts playing mod tunes whilst demanding you pay by credit card to an account in Russia to avoid being "pwned by l33tgr0up" that is likely not a good sign.
Don't say that. Imagine if the Russians had said that to Gorbachev when announced Glasnost. Our overlords have announced a period of openness and we should encourage it. Otherwise we'll be back in the gulag on meager rations of Jon Katz stories with a page that takes 10 minutes to load and looks like ass.
Though I'll say one for their attempts at a redesign - they may have looked like ass but it looked exactly the same ass whether you use Chrome, Opera, Firefox or IE. That's technically pretty impressive in that their horrible unusable CSS and Javascript monstrosity was browser independent. They must have tested it on each browser very carefully in order to make sure they same - and admittedly tortuous - user experience was had on each one.
Come to think of it slashdot's decision to embrace Glasnost is probably happening for much the same reasons as the USSR's if you look at the way the number of comments have dropped over the last few years.
It's probably because of low wages too. I remember when I was in Malaysia I was with someone who dropped off some wages for Indonesians working in a Malaysia factory. Now I worked out the hourly rate and when I went back to Taiwan I mentioned it to people who know about things and they were surprised it was lower than someone doing the same job in China.
So a lot of Taiwanese companies like Foxconn are adopting a "China+1" strategy for manufacturing, i.e. factories in China plus one other lower wage country in Asia. Indonesia seems like a good bet for this.
Actually another thing that's interesting is that in places like Indonesia most people have a phone. Smartphone penetration is lower but it is growing fast
http://www.emarketer.com/Artic...
From 12% to 24% in one year. 84% of people have a mobile phone though.
Also if you look here
http://mashable.com/2013/08/27...
There are a lot places with more smartphones per capita than the US (56%). And a lot of them are in Asia - Hong Kong(62%), South Korea(73%), Singapore(71%) being the obvious ones. So the odds are that most Indonesians will buy a smartphone sooner or later. Importing stuff in Asia is a nightmare because of duties and bureaucracy. So one way around that is to make things in country instead of importing them. Foxconn are a contract manufacturer so they make things for other people's brands. It may well be that those brands think that manufacturing in Indonesia is a good bet because of a mix of low costs and the fact that getting a "Made in Indonesia" stamp on the device means they can avoid import duty on the devices they sell there. Also most likely they can get some assistance from the government given that they're building a factory which can employee a lot of people.
Another thing is that Indonesia may have its issues but it is probably easier to get your profits back from Indonesia to Taiwan. I've talked to people in Taiwan who've pointed out that doing that from China is non trivial. The Chinese RMB for example isn't convertible but the Indonesian Rupiah is.
Also no ARM processor shipping has the horsepower to do a good job at emulating even a low end x86. I think even a really good JIT engine running on the fastest ARM you can get is going to be slower at running x86 code than even the slowest x86.
Back when Dec was trying to get people to buy Alphas they wrote a really clever piece of software called FX!32. It would take an x86 binary and initially run it slowly via instruction by instruction emulation. However while doing that it would profile the application and use the information it gained from that to work out which bits were worth JITting to native Alpha code. Since at that point the fastest Alpha was faster than x86 it ended up running JITted code faster too. So at one point the fastest way to run x86 code was in FX!32 on Alpha.
Now that was possible because of that speed advantage. However really everything comes down to power consumption. Alpha was a server processor and burnt watts prodigiously even compared to Intel. It was also a very clean design and just at the point where Risc was a very efficient design philosophy - the fastest Alpha run about twice the speed of the fastest x86 and the ISA was so recent that it could be implemented efficiently. As time went by it seemed like Risc seemed less of a net win - e.g. Risc chips with 32 bit fixed length instructions had worse code density than x86 and x86 chips moved to a design where x86 opcodes were decoded into Risc like uops internally.
Now with ARM it's not like that. ARM cores are designed for mobile devices and are brilliant for low power and small core size. However there aren't people selling ARMs that are faster than the x86. In fact in general the fastest ARM has run neck and neck with the slowest x86.
We heard you liked Metro so we put Metro on you PC so you use your mouse to make touch screen gestures.
What about this
So most of those names were meant to be there, it is right for
them to be published, it is right to publish the names of
politicians, generals bureaucrats, etc, who are involved in this
sort of activity, it is right even to publish the names of corrupt radio
stations in Kabul that were taking SYOPS programme content. It is
also right to publish the names of those people who have been
killed and murdered and who need to be investigated and it is
right to publish the names of all incidental characters who
themselves are not at serious and probable risk of physical harm.
Those incidental characters are someone who owns a company for
example is just involved in shipping operations.... So then there is the
question were there any sort of villagers or so on who gave
information that might lead to reprisals, were there some of those?
Um there were some villagers who - who had given information,
um so that is a regrettable oversight, but it is not our, not merely
our oversight it was the oversight of the United States military
who should've never included that material and who falsely
classified it, and who then made it available to everyone and it
then got out."
Bummer for those "villagers" who opposed the Taliban. Or for anyone who operated a radio station that was pro government and anti Taliban.
So he was in favour of publishing names. And he did too, after a Twitter poll of his followers. Up to that point there was no evidence that Leigh publishing the password had caused the unredacted cables to become generally available.
http://www.theguardian.com/med...
The Guardian book revealed the diplomatic files were placed by WikiLeaks on a secure online server in July 2010, which it was agreed would only be online for a matter of hours.
This server held a heavily encrypted file containing the unredacted embassy cables database. Assange had given Leigh the password to unlock this file once he had obtained it, and this password was included in the book - seven months after the temporary file was taken offline. No trace could be found through web links or Google's archives of this file ever being visible through this secure server.
However, at a later stage the same encrypted file and at least one other encrypted with the same password was posted on the peer-to-peer file-sharing network BitTorrent. One of these files was first published on 7 December 2010, just hours before Assange's arrest. In the days running up to his arrest, Assange had spoken of "taking precautions" in the event of anything untoward happening to him.
This file, it was later discovered, was the same file that had been shared with the Guardian via the secure server. It shared the same file name and file size, and could be unlocked using the same password as that given to Leigh.
Daniel Domscheit-Berg, a former member of staff at WikiLeaks who is attempting to set up a rival whistleblowing website, discovered this republished file and shared information on WikiLeaks's security breach with a small group of journalists.
Avoiding the re-use of passwords and avoiding republishing temporary files are both considered basic security procedures among online security experts.
However, the file was not discovered or downloaded by the public. By 10am on Thursday it had been accessed once in the previous 31 days, despite mounting speculation about its existence.
Initial news stories did not give details of the location of files or of passwords. Later, WikiLeaks and some of its supporters published a series of hints about the passwords and files.
At about 11pm on Wednesday an anonymous Twitter user discovered the published password and opened a separate file - not the one shared with the Guardian - that had also been circulating on file-sharing networks for several m
Err no. It's because of the way the Official Secrets Act 1989 works
http://www.headoflegal.com/201...
This language makes me wonder whether the Guardian was facing an "official direction" for the return or disposal of the material under section 8(5) of the Official Secrets Act 1989.
It would be an offence under section 6(2) of the Act for the Guardian to knowingly make a damaging disclosure of any information, document or other article which (section 6(1)(a))
(i) relates to security or intelligence, defence or international relations; and
(ii) has been communicated in confidence by or on behalf of the United Kingdom to another State ...
and (also section 6(1)(a))
has come into a person's possession as a result of having been disclosed (whether to him or another) without the authority of that State ...
Documents leaked by Edward Snowden about the work of GCHQ must I think fall within the scope of section 6(1), having presumably been communicated in confidence by the UK intelligence agencies to another state, the US, and having come into the Guardian's possession without US authority.
If that's right, then, as I've said, the Guardian and its editor would risk committing an offence if it published any of that information which was "damaging". By the interaction of section 6(4) and section 1(4)(a), by the way, disclosure of security or intelligence information would be "damaging" if (section 1(4)(a))
it causes damage to the work of, or of any part of, the security and intelligence services
In those circumstances, section 8(5) would apply. It says
Where a person has in his possession or under his control any document or other article which it would be an offence under section 6 above for him to disclose without lawful authority, he is guilty of an offence if he fails to comply with an official direction for its return or disposal.
This all dates back to when secret documents were not digital - e.g. paper or microfilm. If you had them you'd could be directed to "return or dispose[destroy]" them. And if you failed you could be prosecuted.
Incidentally if you or I rather than the Guardian were doing this the consequences would likely be much more drastic. The police would seize all your computers and get a court order to get access to all your offsite backups. In fact this is what happens when people steal data from their employers let alone from the NSA/GCHQ.
In the case of the Guardian it seems like they've gone through this charade as the minimum they can legally do given that the Guardian has told them that copies of the data exist in the US.
One of the -false- accusations against wikileaks was their undiscriminate leaking of classified documents.
False?
http://download.cabledrum.net/...
Interviewer: "So come on, redactions are going on at the same time, now there is
or isn't a row going on about redaction, I haven't the faintest clue
whether there is or isn't...?
Mr Assange: No, there's no row going on about redactions at all....There was a
group of reports where although they were not really intelligence
informants there were sort of hotline tips...something called threat
reports comprised one in five of the Afghan War Logs and so we held
them back for a line by line redaction...But what we didn't do was
redact one in five lines, putting black marker through it, we just
removed them, and so it looked like we hadn't redacted everything but
in fact we had redacted a fifth of all material, and this permitted an
attack, a political attack, to come from The Times of London.... So The
Times did a proxy war on The Guardian through us by attacking us....
So most of those names were meant to be there, it is right for
them to be published, it is right to publish the names of
politicians, generals bureaucrats, etc, who are involved in this
sort of activity, it is right even to publish the names of corrupt radio
stations in Kabul that were taking SYOPS programme content. It is
also right to publish the names of those people who have been
killed and murdered and who need to be investigated and it is
right to publish the names of all incidental characters who
themselves are not at serious and probable risk of physical harm.
Those incidental characters are someone who owns a company for
example is just involved in shipping operations.... So then there is the
question were there any sort of villagers or so on who gave
information that might lead to reprisals, were there some of those?
Um there were some villagers who - who had given information,
um so that is a regrettable oversight, but it is not our, not merely
our oversight it was the oversight of the United States military
who should've never included that material and who falsely
classified it, and who then made it available to everyone and it
then got out."
Assange never wanted to redact but was forced his media partners. Then he published the full unredacted cables on wikileaks' website. Which they denounced
http://www.bbc.co.uk/news/worl...
In a joint statement, the Guardian, El Pais, New York Times and Der Spiegel said they "deplore the decision of WikiLeaks to publish the unredacted state department cables, which may put sources at risk".
And before you mention the password that appeared in David Leigh's book that was supposed to be for a temporary copy of the archive
http://www.theguardian.com/med...
WikiLeaks claimed its disclosure was prompted after conflicts between Assange and former WikiLeaks associates led to one highlighting an error made months before. When passing the documents to the Guardian, Assange created a temporary web server and placed an encrypted file containing the documents on it. The Guardian was led to believe this was a temporary file and the server would be taken offline after a period of hours.
However, former WikiLeaks staff member Daniel Domscheit-Berg, who parted acrimoniously with WikiLeaks, said instead of following standard security precautions and creating a temporary folder, Assange instead re-used WikiLeaks's "master password". This password was then unwittingly placed in the Guardian's book on the embassy cables, which was pu
You are also forgetting that a x86 or a amd64 is a RISC cpu with a layer od CISC hidding the RISC. That layer takes cpu space, power and resources (both designing and working).
Look at a die shot of a x86 CPU. Most of the space is cache. The actual CPU is a small fraction. Sure on an embedded system where you care about power and die area ARM makes sense. Switching a desktop or server class machine to ARM suffers from diminishing returns because the CPU is already such a small percentage of die area.
Thumb-2 is a great idea but unfortunately it is not supported for 64 bit mode. 64 bit mode is AArch64 which used 32 bit fixed length instructions
http://www.arm.com/files/downl...
That being said it's not as bad as ARM32
https://groups.google.com/d/to...
So far, the LZSS routine in my code looks like this:
THUMB2: 76 bytes
THUMB: 76 bytes
ARM64: 96 bytes
ARM32: 116 bytes
(for comparison, x86 is 63 bytes)
Still it is quite a bit worse than x86. x64 incidentally is the same code size as x86
http://www.deater.net/weave/vm...
8086 58 bytes
x86_x32 66 bytes
x86_64 66 bytes
arm_thumb2 76 bytes
arm_thumb 76 bytes
m68k 88 bytes
arm64 96 bytes
z80 96 bytes
arm eabi 116 bytes
Of course this benchmark is a bit silly.
. I mean, the instruction set has specialized instructions for handling packed decimal! And then there's the near worthless string search REPNE CMPSB family of instructions. The Boyer-Moore string search algorithm is much faster, and dates back to 1977. Another sad thing is that for some CPUs, the built in DIV instruction was so slow that sometimes it was faster to do integer division with shifts and subtracts. That's a serious knock on Intel that they did such a poor job of implementing DIV. A long time criticism of the x86 architecture has been that it has too few registers, and what it does have is much too specialized. Like, only AX and DX can be used for integer multiplication and division. And BX is for indexing, and CX is for looping (B is for base and C is for count you know-- it's like the designers took their inspiration from Sesame Street's Cookie Monster and the Count!) This forces a lot of juggling to move data in and out of the few registers that can do the desired operation. This particular problem has been much alleviated by the addition of more registers and shadow registers, but that doesn't address the numerous other problems. Yet another feature that is obsolete is the CALL and RET and of course the PUSH and POP instructions, because once again they used a stack. Standard thinking 40 years ago.
It was standard on the 8086 (introduced in 1978). The 80368 (1985) is a general purpose register machine and can use a 0:32 flat memory mode. And modern x64 (2003) has twice as many registers and the ABI specified SSE for floating point, not 8087. Also in 64 bit mode segment bases and limits for code and data (i.e any instruction which does not have a segment override prefix) are ignored.
I.e pretty much all the things you're complaining about have been fixed and if you look at benchmarks x64 chips have been faster than their Risc competitors pretty much since x64 was introduced. Going on about segments, floating point stacks and REPNE MOVSD now is absurd.
And if you look at the way the 8086 took over from the 8080 what they did made a lot of sense. You could machine convert a CPM 8080 program to a MS DOS 8086 one and have it run fine. Meanwhile a native 8086 program had access to 1M of address space, up from 64K on 8080 and Z80. Like CPM MSDOS ran on commodity hardware which was cheaper than the big iron boxes and given most people were running things like Visicalc and Wordstar it was perfectly sufficient.
http://nextbigfuture.com/2011/...
Rooftop solar is several times more dangerous than nuclear power and wind power. It is still much, much safer than coal and oil, because those have a lot of air pollution deaths.
And a police officer has the technical capacity to walk into my house and shoot me dead. That I can appreciate his likely skill with a service revolver doesn't mean he gets to shoot me dead at a whim.
Right but you accept the fact that the police need to have the capability to shoot people, right? Because if you were an armed robber or something they'd need to be able to do that to stop you.
Similarly the NSA needs to have the capability to spy on people - terrorists, Russian or Chinese spies, or - if WWIII starts - Russian or Chinese soldiers are all people the NSA needs to be able to spy on. In fact it's highly irritating when people who tweet their every thought and bowel movement whine about this. The NSA aren't going to spy on them because a) everything they think is public and b) they're not interesting to anyone, let alone the NSA. Logically given limited resources it's more likely that people like the Boston Bombers are the target of surveillance than people someone memorably referred to as 'twitter cunts'.
If you look at WWII Anglo American SIGINT like breaking the Enigma code was absolutely vital to the war effort and saved the UK from defeat. As China moves towards parity with the West and confronts Japan over the Senkakus it's not impossible the US may find itself in a similar situation. In the long run it's not impossible that Russia will threaten the Ukraine militarily - after all it did more than threaten Georgia.
And in fact having a major SIGINT advantage over Russia and China is likely to act as a deterrent on them doing something like this. Conversely Snowden visiting both and telling them the US's capabilities is likely to make them think they're the ones with the advantage.
The only reason you'd think Snowden did the right thing is if you think the US is the sole source of evil in the world and Russia and China are both governed by people who act robotically in the best interests of humanity eschewing any personal gain. How likely is it really that the people who govern the US are the only ones vulnerable to corruption and the far less open political systems of Russia and China magically produce incorruptible leaders?
I'd say as bad as the US's politicians are the openness of the system means they are likely a lot less bad than those in China or Russia. In which case I'd rather the US has the SIGINT advantage. Snowden did exactly the wrong thing in taking US secrets to Russia and China and the Guardian is wrong to publish US secrets.
I guess this means the Angry Birds tie in with The Smurfs and James Bond are not going to happen.
If someone wants to spend more to be at the forefront, why not congratulate them and allow the march of progress continue?
Because they're doing enough of that congratulation themselves whilst whizzing past you in the carpool lane or while making passive aggressive comments about how 'gas guzzlers' should be banned by the government.
My favourite Chinese salesman-ism is, when encouraging you to try some food that looks, sounds or tastes a bit horrible is "...very good for you" said it a suitably serious tone.
E.g.
"Black dog. Very good for you"
Or, if you have a nasty rash from eating the food or breathing the air for an extended period
"Very good for skin".
Such fashion
Hipster Approved
Fair Trade Vegan Offal
Hot Doge
Within Reach
You can bullshit HTTP. Just learn GET and POST and learn a speech about what 'idempotent' means. Once you explain that most people's eyes glaze over, so you can skip the rest.
With the increase of insect resistance to pesticides we'd going to need some novel countermeasures. I propose widespread deployment of Diatomaceous Earth. Sure some people will die but the rest will evolve resistance.
Now you'll say "That's complete madness. Humans are the ultimate K selected species so they've got no chance of out evolving r selected insects" Ordinarily that would be true but by careful use of mutagens and radiation and encouraging r selected traits like promiscuity and abandoning your kids. I think we can win.
And we need too, to ensure that human civilisation, not bed bug, dominates New York. NOW AND FOREVER.
They are not legally required to so why would they?
Because they can put it up on blog and make money out of ads? Because it works as a loss leader for their other paid services. Because it's cheap to publish these days and if you're going to do the research for your own curiosity why not publish?
Look at OkCupid. Publishing things like this
http://blog.okcupid.com/index....
Makes me like them a whole lot. Now I'm not really in the market for a US centric dating service but if I were I'd use them. Plus they could always write a book full of this sort of stuff.
Does everyone publish all their data? No of course not. Still the trend is that people increasingly do do it for the reasons mentioned.
No worries. We'll send the Grim Reefer to get all Falling Down on their asses
http://tech.slashdot.org/comme...
I don't know your reference for saying that "the models are broken". In my understanding the models used e.g. in IPCC reports, are quite good.
It is completely unreasonable to dismiss them just because they are not perfect. The proper approach is to study the discrepancies, reason about their possible causes and estimate the effect of the errors on the question you are seeking to answer with the model.
Look at this
http://www.drroyspencer.com/20...
What I don't get is why they don't chuck out the models that are bad, keep the ones that are good and invent ones that are better. Right now it almost seems like they do the opposite - the ones that predict OMG! Runaway Global Warming! get loads of press time. And the ones that don't predict anything too drastic get largely ignored and the people that made them called evil deniers.
Actually that's what denier really means. If you think the world is warming slowly but it's no great problem like the satellite temperature measurements say you're an evil denier. In fact unless you support massive CO2 cuts now on the basis of the most alarmist model you're a denier. I.e. it's really an argument about policy, not whether you think the world will be 0.5 degree C or 1.0 degree C warmer in a hundred years time.
You can see that when geoengineering is brought up. The Royal Society did a study that showed that Sulphate aerosols for example could be used to effect "a reduction of solar input by about 2%" to "balance the effect on global mean temperature of a doubling of CO2" for "total annual cost at 10s of billion dollars". So we don't need to rely on the precautionary principle to tell us we need to cut our CO2 emissions to zero now just in case. We can go on as we are, monitor temperature and build ourselves a planetary thermostat quickly and cheaply if it becomes necessary. Of course the 'cut CO2 now' lobby hate this.
They also hate it when you point out that CO2 emissions in Europe and the US are trending down. China's CO2 emissions are increasing massively. If you want a global CO2 cut you'd need to get China to stop industrializing. Which they won't do
http://photos.mongabay.com/09/...
Or of course that the actual satellite measurements of temperature are undershooting all the models - you need to use the adjusted temperature measurements from ground stations. And the adjusted temperature measurements only do that because the past has been getting cooler. Those cavemen better watch out, pretty soon it will be below absolute zero when they are.
What this is really about is that you've got people who'd make a load of money if everyone was forced to by CO2 permits. It's pure rent seeking by them. Or people who know deep down we're all sinners for our materialistic lifestyle and want to force everyone into the modern equivalent of sackcloth and ashes to repent.
When people disagree with me about policy I also think they are stupid stupid-heads who don't understand The Science.