That's a bit of a stretch - Android apps are written in Java, compiled to Java bytecode, and then converted to the Dalvik bytecode.
But anyway, yeah, VMs don't necessarily suck, like everything it depends what you do with them. How many times have we heard that Python was the language to use for rapid development, and that speed didn't matter because only 1% of CPU time is spent in the interpreter, whilst the other 99% is spent in custom C libraries. It's the same with Dalvik - the speed constrained Java classes like video decoding are written in C (or maybe even ARM assembler), and then called from the Java code. At the end of the day Flash is all over the web and its performance isn't great, so I doubt this is going to hold Android back. Having said that, I'm still looking forward to native C code running on Android, which I expect will happen once Skia and Dalvik are running on the Neo Freerunner. If you're interested in details there's a lot of info on android-internals.org
Of course it's the network. They did not even check to see if the iphone was on a 2G or 3G network.
retards.
Given that both phones are network locked and provided on contract in the UK, for most people (who don't hack their phones) the phone and network are intrinsically linked, and they will judge the performance of one by the other.
But note the article update:
"we have now run a set of tests and concluded that, indeed, both phones load pages at a similar speed over Wi-Fi. This means there's little difference in processor or browser performance. Clearly the G1 is a superior Web phone to the Omnia, but it seems to be O2's network that is holding the iPhone back."
I have the same setup. The virtual desktop works fine - somehow Xorg figures out that half the virtual desktop is a different resolution and does the right thing. I have taskbar pop up on 1400x1050 screen, can't remember why.
Same virtual screen ('Virtual 2800 1050' in xorg.conf), in/etc/kde3/kdm/Xsetup :
Also, please consider this. One of those Muslims I've been aquainted with was not just a homophobe, but a very open "wish-we-could-stone-them" one. Ironically, I've met him in a Western university.
Funnily enough, I've had this same discussion with a Christian medical student at a well known British university. This guy (who apart from his religious views was actually a decent person) said that he wouldn't treat homosexuals, even if he found some in a car crash and they would otherwise die, because the Bible says it's a crime against God and they should be put to death anyway. Religion, eh?
So, Android needs to be ported, like any OS. Mobile phones, in particular, have very specific hardware. If you tried to put the OS from the G1 onto another phone, you'd need to add drivers for the other chips on it, especially things like the cellular baseband chip, and the hardware for things like audio input/output, LEDs, etc. It's sort of like RockBox in that it requires a large chunk of work to be ported. They initially ran on only one device, but, over time, gained additional compatibility.
Ports shouldn't require a massive amount of work - mobile hardware seems to have evolved to a fairly common platform, ARM CPU, some LCD driver, audio device, wifi chip, etc. Many of those chipsets will already have Linux drivers, or slightly similar variants. When Linux was ported to the Xbox there was some amount of work, but a great deal was already done - the kernel ran on x86 already, the open source NV20 video driver worked with only some minor tweaks, audio was an Intel chipset and the driver worked with only minor changes, USB was a standard chipset and the driver worked immediately, etc. I doubt there are many mobile phones out there that would require the whole kernel porting from scratch, and certainly those that are wouldn't be used as the base for new Android based hardware.
So how open is it? Surely it's easy to disable the "kill switch"?
More lazy questions:
Can you replace the kernel with a custom compiled one? Are all of the drivers open source? Is the desktop open? Which widget toolkit does it use - can you run gtk/qt/x apps on there? Can you compile real apps or just Java? What distribution does it run? What package manager? Can you even update the applications? Is there any chance of a proper distribution like Ubuntu being ported to this thing?
a government that supported a terrorist organization is as illegitimate at the terrorist organization itself
Does this apply to the U.S. government, too? After all, the U.S. supported the Afghan Mujahideen when they were launching "terrorist" attacks against Russian soldiers.
One of my non-geek friends got an Aspire One. It hasn't exactly been plain sailing.
The Aspire One runs Linpus Linux, it's a Fedora variant but they've modified lots of packages, and there's no online repository for their custom RPMs, and the changes weren't fed upstream. There are some major problems: "yum update" (either global, or anything that touches the modified packages) will replace the custom packages and break stuff,mplayer has very few codecs (no xvid), installing vlc requires manually linking directories, the menu system is non-standard so installed software isn't in the menu (in fact, the custom openoffice packages do set up menu entries, but if you update them to the fedora packages they break), and you can't update to firefox 3 without major hassles.
A hacked up Chinese Fedora clone isn't exactly a great example of a fluid Linux experience - stock Ubuntu would have been far preferable, and saved me from all those "this is Linux? I thought you said it was better than Windows. This would've been far easier in Windows." comments.
Actually it looks like that page about the FSA probably is a relic - from Paypal Lux migration FAQ:
What is the change to PayPal and my account?
PayPal was granted a bank license with the Luxembourg bank authority. Under this license, PayPal will be regulated centrally by the Commission de Surveillance du Secteur Financier (CSSF), the Luxembourg bank authority.
On 2 July 2007, your customer relationship will be automatically transferred from PayPal (Europe) Ltd. to PayPal Luxembourg.
So, Paypal (Europe) Ltd. may well have an e-money license, but you aren't a customer of them anymore:-)
According to this article in the Telegraph, PayPal moved to Luxembourg and became a bank.
Whether they're still regulated by the FSA is an interesting question. they claim to be. However, this may just be a relic of the pre-Luxembourg organisation, or it may be a way of avoiding the need to be re-approved as an e-money issuer.
E-Money balances aren't covered by any form of insurance or security - the FSA e-money sourcebook states that companies must invest in high-security bonds etc., but ultimately a company could go bust and their users would have no more claim on their holdings than any other creditor. The £1000 amount is just an upper limit on an account balance before they have to tell you, and you have to agree that you know, that you have no right to compensation. You have no right to compensation anyway (ELM section 1.5) - the notification regulation is just there to limit the losses of those people who would claim that they never knew.
A terrorist organisation - or at least, one that's even remotely successful for any length of time - doesn't have the pyramid-style management structure that you're used to seeing.
Not true - the Provisional IRA managed to successfully launch attacks against the British for 40 years, and they had a hierarchical organisational structure.
There's no simple answer here. As the FSF say in the answer you link to "This is a legal question, which ultimately judges will decide." And you missed the rest of the answer following your quote "But if the semantics of the communication are intimate enough, exchanging complex internal data structures, that too could be a basis to consider the two parts as combined into a larger program." It could certainly be argued in a court that distributing an application that performs video encoding by transferring commands and data frames across an IPC link constitutes a derivative work, especially if there's no way for the encoding application to work if the GPL component were removed (as would be the case here). The mere aggregation clause was meant to apply more to clear cut cases like distributing a Linux distribution, where differently licensed unconnected software like, say, emacs and skype, could be incorporated on the same CD.
Their commercial software pipelines frames into their wrapper ; they are separate processes, not linked, and thus their use does not violate GPL
Well, if they don't distribute, then the GPL indeed doesn't apply. But if it they do, then an argument could definitely be made that the GPL would apply if the GPL code were an essential part of the software as a whole (ie. it couldn't be replaced), and they were distributing both sets of code together as a single software suite. The GPL license doesn't say anything about code running in a different process being automatically excluded from the licensing requirements, it only talks about "derivative work", without specifying exactly what that means. The same argument exists over the question of closed source kernel modules - many prominent kernel developers believe that they're illegal. The GPL FAQ says:
"I'd like to incorporate GPL-covered software in my proprietary system. Can I do this by putting a 'wrapper' module, under a GPL-compatible lax permissive license (such as the X11 license) in between the GPL-covered part and the proprietary part?
No. The X11 license is compatible with the GPL, so you can add a module to the GPL-covered program and put it under the X11 license. But if you were to incorporate them both in a larger program, that whole would include the GPL-covered part, so it would have to be licensed as a whole under the GNU GPL.
The fact that proprietary module A communicates with GPL-covered module C only through X11-licensed module B is legally irrelevant; what matters is the fact that module C is included in the whole."
CO2 causes global warming, apparently on Mars and Jupiter too.
"There have been claims that warming on Mars and Pluto are proof that the recent warming on Earth is caused by an increase in solar activity, and not by greenhouses gases. But we can say with certainty that, even if Mars, Pluto or any other planets have warmed in recent years, it is not due to changes in solar activity.
The Sun's energy output has not increased since direct measurements began in 1978 (see Climate myth special: Global warming is down to the Sun, not humans). If increased solar output really was responsible, we should be seeing warming on all the planets and their moons, not just Mars and Pluto." source
the failure of Global Warming models, which all assume a constant input from the sun
Climate scientists are well aware that the sun contributes to Earth's climate. However, it is only one factor, and greenhouse gases provide a more accurate model of recent and historical temperature trends.
Taxes that apply the real world cost of an externality are common in market based systems. See, for example, the California air pollution tax - for a nice introductory economics analysis see "The Undercover Economist".
Also, agreeing with trading partners to implement similar taxation systems is possible - cross-border taxation with the E.U. is an obvious example, where countries are in effect regulated through general agreement in to not dropping VAT or corporation tax to "harmful" levels (where harmful is something that encourages artificial tax structures within the otherwise free-trade market).
A good question to ask is how is the "mean" earth temperature measured, both current and Historic and whats the variance?
The global absolute mean is difficult to calculate, since it requires all equipment and readings to be calibrated globally. That's why climate scientists tend to calculate the relative changes at whatever geographical point they're sampling. The sequences of relative changes can then be combined to produce statistically accurate portraits of the global change (relative change, not absolute values). The absolute value of the global mean doesn't really tell us anything, as we are concerned with regional variations.
There are some facts out there that do matter (e.g. CO2 increases from industrialization). But thats not a proof of any claimed causal effects.
Irrefutable proofs are for mathematics. In the real world, we know that CO2 is a greenhouse gas, and the model of how we expect increased levels of it to behave with respect to temperature matches observations from the real world. If you have a model that better matches the data, then go ahead and get it published.
Am I missing something here? It appears that the treaty has been ratified, and U.S. citizens can be extradited to the U.K. for any crime without any evidence being presented in a U.S. court.
"We conclude that, given the clear differences in definition, the UK can no longer rely on US assurances that it does not use torture, and we recommend that the Government does not rely on such assurances in the future."
This means that for terrorism crimes, it's very likely that extradition requests to the U.S. will have to be denied, since the U.S. carries out activities that the U.K. considers torture. And a "no-torture" guarantee is worthless, since the U.S. doesn't consider the acts as torture in the first place. At a minimum, expect this issue to be brought up in legal challenges to extradition.
On 30 September 2006 the US Senate unanimously ratified the treaty.source
Of course, I would like to see the UK extradite a U.S. business man (Ian Norris, Morgan Crucible), or even an internet pirate (Hew Raymond Griffiths, DrinkOrDie). I imagine many people would claim such a thing to be unconstitutional - the alternative, that any crime committed in a globalised post-internet world can be prosecuted by any extradition treaty nation, regardless of the laws of the nation in which the defendant actually resides.
Maybe I should have said "Anything that's considered a worse experience than having your limbs pulled from their sockets has got to be pretty bad." Whether psychological torture inflicts a form of pain in itself is another, perhaps more philosophical, question, but at the end of the day what matters is the sum experience of the person experiencing the events.
There are many documented stories from drowning survivors who recall the pain before losing consciousness. A quick search brought up the following quote, I'm sure you can find others.
One myth about drowning that survives to this day is that it is a painless, almost pleasant way to die. Survivors testify otherwise:
"...When the cramp hit me, I sank to the bottom of the lake 12 feet down, in a doubled-up position. Compounding the wracking pain in my trunk was a mounting choking sensation. (Try holding your mouth and nose after taking a deep breath. Hold your breath until it becomes unbearable; then try holding it a few seconds past the unbearable point. It's a horrible sensation and would give you a dim idea of just one aspect of how it feels to drown.) The pressure of the water caused a stabbing pain in my eyes and ears... try to keep your head when water begins to seep into your already tortured lungs and your body is a mass of pain and you know you are dying... I remember that I screamed down there against a solid wall of waterï½I remember that I threshed and bobbed, but only succeeded in burrowing my head into the slime of the lake floor...."
The protester videos are street theater. They use a waterproof covering to ensure that no water is actually inhaled. The truth is that, by all accounts, drowning is a very painful experience. In Algeria, people often died from waterboarding (see Henri Alleg). In the Middle Ages waterboarding was generally used when the rack failed. Anything that's more painful than having your limbs pulled from their sockets has got to be pretty bad.
Slashdot Mirror
That's a bit of a stretch - Android apps are written in Java, compiled to Java bytecode, and then converted to the Dalvik bytecode.
But anyway, yeah, VMs don't necessarily suck, like everything it depends what you do with them. How many times have we heard that Python was the language to use for rapid development, and that speed didn't matter because only 1% of CPU time is spent in the interpreter, whilst the other 99% is spent in custom C libraries. It's the same with Dalvik - the speed constrained Java classes like video decoding are written in C (or maybe even ARM assembler), and then called from the Java code. At the end of the day Flash is all over the web and its performance isn't great, so I doubt this is going to hold Android back. Having said that, I'm still looking forward to native C code running on Android, which I expect will happen once Skia and Dalvik are running on the Neo Freerunner. If you're interested in details there's a lot of info on android-internals.org
Given that both phones are network locked and provided on contract in the UK, for most people (who don't hack their phones) the phone and network are intrinsically linked, and they will judge the performance of one by the other.
But note the article update:
"we have now run a set of tests and concluded that, indeed, both phones load pages at a similar speed over Wi-Fi. This means there's little difference in processor or browser performance. Clearly the G1 is a superior Web phone to the Omnia, but it seems to be O2's network that is holding the iPhone back."
I have the same setup. The virtual desktop works fine - somehow Xorg figures out that half the virtual desktop is a different resolution and does the right thing. I have taskbar pop up on 1400x1050 screen, can't remember why.
Same virtual screen ('Virtual 2800 1050' in xorg.conf), in /etc/kde3/kdm/Xsetup :
xrandr --output VGA-0 --mode 1280x1024
xrandr --output VGA-0 --right-of LVDS
There's probably a better way to automagically do it, but this works great for me.
Kind of like how racially ambiguous videogame characters probably appeal to a wider range of players? Remember Jade is black? and Do You Care About Race in Games?
Also, please consider this. One of those Muslims I've been aquainted with was not just a homophobe, but a very open "wish-we-could-stone-them" one. Ironically, I've met him in a Western university.
Funnily enough, I've had this same discussion with a Christian medical student at a well known British university. This guy (who apart from his religious views was actually a decent person) said that he wouldn't treat homosexuals, even if he found some in a car crash and they would otherwise die, because the Bible says it's a crime against God and they should be put to death anyway. Religion, eh?
So, Android needs to be ported, like any OS. Mobile phones, in particular, have very specific hardware. If you tried to put the OS from the G1 onto another phone, you'd need to add drivers for the other chips on it, especially things like the cellular baseband chip, and the hardware for things like audio input/output, LEDs, etc. It's sort of like RockBox in that it requires a large chunk of work to be ported. They initially ran on only one device, but, over time, gained additional compatibility.
Ports shouldn't require a massive amount of work - mobile hardware seems to have evolved to a fairly common platform, ARM CPU, some LCD driver, audio device, wifi chip, etc. Many of those chipsets will already have Linux drivers, or slightly similar variants. When Linux was ported to the Xbox there was some amount of work, but a great deal was already done - the kernel ran on x86 already, the open source NV20 video driver worked with only some minor tweaks, audio was an Intel chipset and the driver worked with only minor changes, USB was a standard chipset and the driver worked immediately, etc. I doubt there are many mobile phones out there that would require the whole kernel porting from scratch, and certainly those that are wouldn't be used as the base for new Android based hardware.
So how open is it? Surely it's easy to disable the "kill switch"?
More lazy questions:
Can you replace the kernel with a custom compiled one? Are all of the drivers open source?
Is the desktop open? Which widget toolkit does it use - can you run gtk/qt/x apps on there?
Can you compile real apps or just Java?
What distribution does it run? What package manager? Can you even update the applications? Is there any chance of a proper distribution like Ubuntu being ported to this thing?
Does this apply to the U.S. government, too? After all, the U.S. supported the Afghan Mujahideen when they were launching "terrorist" attacks against Russian soldiers.
One of my non-geek friends got an Aspire One. It hasn't exactly been plain sailing.
The Aspire One runs Linpus Linux, it's a Fedora variant but they've modified lots of packages, and there's no online repository for their custom RPMs, and the changes weren't fed upstream. There are some major problems: "yum update" (either global, or anything that touches the modified packages) will replace the custom packages and break stuff, mplayer has very few codecs (no xvid), installing vlc requires manually linking directories, the menu system is non-standard so installed software isn't in the menu (in fact, the custom openoffice packages do set up menu entries, but if you update them to the fedora packages they break), and you can't update to firefox 3 without major hassles.
A hacked up Chinese Fedora clone isn't exactly a great example of a fluid Linux experience - stock Ubuntu would have been far preferable, and saved me from all those "this is Linux? I thought you said it was better than Windows. This would've been far easier in Windows." comments.
So, Paypal (Europe) Ltd. may well have an e-money license, but you aren't a customer of them anymore :-)
Whether they're still regulated by the FSA is an interesting question. they claim to be. However, this may just be a relic of the pre-Luxembourg organisation, or it may be a way of avoiding the need to be re-approved as an e-money issuer.
E-Money balances aren't covered by any form of insurance or security - the FSA e-money sourcebook states that companies must invest in high-security bonds etc., but ultimately a company could go bust and their users would have no more claim on their holdings than any other creditor. The £1000 amount is just an upper limit on an account balance before they have to tell you, and you have to agree that you know, that you have no right to compensation. You have no right to compensation anyway (ELM section 1.5) - the notification regulation is just there to limit the losses of those people who would claim that they never knew.
It is - in Europe Paypal is a licensed, regulated bank based in Luxembourg.
Not true - the Provisional IRA managed to successfully launch attacks against the British for 40 years, and they had a hierarchical organisational structure.
There's no simple answer here. As the FSF say in the answer you link to "This is a legal question, which ultimately judges will decide." And you missed the rest of the answer following your quote "But if the semantics of the communication are intimate enough, exchanging complex internal data structures, that too could be a basis to consider the two parts as combined into a larger program." It could certainly be argued in a court that distributing an application that performs video encoding by transferring commands and data frames across an IPC link constitutes a derivative work, especially if there's no way for the encoding application to work if the GPL component were removed (as would be the case here). The mere aggregation clause was meant to apply more to clear cut cases like distributing a Linux distribution, where differently licensed unconnected software like, say, emacs and skype, could be incorporated on the same CD.
Well, if they don't distribute, then the GPL indeed doesn't apply. But if it they do, then an argument could definitely be made that the GPL would apply if the GPL code were an essential part of the software as a whole (ie. it couldn't be replaced), and they were distributing both sets of code together as a single software suite. The GPL license doesn't say anything about code running in a different process being automatically excluded from the licensing requirements, it only talks about "derivative work", without specifying exactly what that means. The same argument exists over the question of closed source kernel modules - many prominent kernel developers believe that they're illegal. The GPL FAQ says:
"I'd like to incorporate GPL-covered software in my proprietary system. Can I do this by putting a 'wrapper' module, under a GPL-compatible lax permissive license (such as the X11 license) in between the GPL-covered part and the proprietary part?
No. The X11 license is compatible with the GPL, so you can add a module to the GPL-covered program and put it under the X11 license. But if you were to incorporate them both in a larger program, that whole would include the GPL-covered part, so it would have to be licensed as a whole under the GNU GPL.
The fact that proprietary module A communicates with GPL-covered module C only through X11-licensed module B is legally irrelevant; what matters is the fact that module C is included in the whole."
"There have been claims that warming on Mars and Pluto are proof that the recent warming on Earth is caused by an increase in solar activity, and not by greenhouses gases. But we can say with certainty that, even if Mars, Pluto or any other planets have warmed in recent years, it is not due to changes in solar activity. The Sun's energy output has not increased since direct measurements began in 1978 (see Climate myth special: Global warming is down to the Sun, not humans). If increased solar output really was responsible, we should be seeing warming on all the planets and their moons, not just Mars and Pluto." source
Climate scientists are well aware that the sun contributes to Earth's climate. However, it is only one factor, and greenhouse gases provide a more accurate model of recent and historical temperature trends.
"The Sun's energy output has not increased since direct measurements began in 1978."
"According to solar physicists, the sun emitted a third less energy about 4 billion years ago and has been steadily brightening ever since. Yet for most of this time, Earth has been even warmer than today, a phenomenon sometimes called the faint sun paradox. The reason: higher levels of greenhouse gases trapping more of the sun's heat."
Taxes that apply the real world cost of an externality are common in market based systems. See, for example, the California air pollution tax - for a nice introductory economics analysis see "The Undercover Economist".
Also, agreeing with trading partners to implement similar taxation systems is possible - cross-border taxation with the E.U. is an obvious example, where countries are in effect regulated through general agreement in to not dropping VAT or corporation tax to "harmful" levels (where harmful is something that encourages artificial tax structures within the otherwise free-trade market).
The global absolute mean is difficult to calculate, since it requires all equipment and readings to be calibrated globally. That's why climate scientists tend to calculate the relative changes at whatever geographical point they're sampling. The sequences of relative changes can then be combined to produce statistically accurate portraits of the global change (relative change, not absolute values). The absolute value of the global mean doesn't really tell us anything, as we are concerned with regional variations.
Irrefutable proofs are for mathematics. In the real world, we know that CO2 is a greenhouse gas, and the model of how we expect increased levels of it to behave with respect to temperature matches observations from the real world. If you have a model that better matches the data, then go ahead and get it published.
Am I missing something here? It appears that the treaty has been ratified, and U.S. citizens can be extradited to the U.K. for any crime without any evidence being presented in a U.S. court.
This means that for terrorism crimes, it's very likely that extradition requests to the U.S. will have to be denied, since the U.S. carries out activities that the U.K. considers torture. And a "no-torture" guarantee is worthless, since the U.S. doesn't consider the acts as torture in the first place. At a minimum, expect this issue to be brought up in legal challenges to extradition.
Of course, I would like to see the UK extradite a U.S. business man (Ian Norris, Morgan Crucible), or even an internet pirate (Hew Raymond Griffiths, DrinkOrDie). I imagine many people would claim such a thing to be unconstitutional - the alternative, that any crime committed in a globalised post-internet world can be prosecuted by any extradition treaty nation, regardless of the laws of the nation in which the defendant actually resides.
There are many documented stories from drowning survivors who recall the pain before losing consciousness. A quick search brought up the following quote, I'm sure you can find others.
The protester videos are street theater. They use a waterproof covering to ensure that no water is actually inhaled. The truth is that, by all accounts, drowning is a very painful experience. In Algeria, people often died from waterboarding (see Henri Alleg). In the Middle Ages waterboarding was generally used when the rack failed. Anything that's more painful than having your limbs pulled from their sockets has got to be pretty bad.
Europe is fucking tiny. Ok?
Europe 10,180,000 km2
United States 9,826,630 km2