I would add that advances in PV should make rooftop PV at the home/office very useful in the future energy grid, as a method for lopping to top off of peak load in warm climates, and sidestepping the bulk of the distribution problem. However as you point out, that will never replace the base load which a Nuclear or Coal plant can supply.
The future energy grid will need to use almost every tech we've got, but you're right that no combination of renewables can scale to meet the *whole* need in the next 20-40 years, and it's pretty clear we need to do something before then.
I said we have no control over stopping earthquakes.
Ok, so let's exercise our control over not starting earthquakes, mr. selective pedant. Do you think that when someone says they want to "stop crime" that they mean that they want to let the crimes start, and then stop them, rather than prevent them?
And I didn't say geothermal is "our single energy savior" - just a good replacement for nukes.
Cool, so you are ok with all the rest coming from coal and the other existing sources? We need the 20% of Nuclear energy to scale to 80%, not something else to take the current 20% and everything else stay the same. When other people pointed out that it is impractical for wind, hydro, or solar to replace everything, you say "what about geothermal", in which case that better be practical enough to replace coal.
Geothermal is available in nearly every country, in most parts of the US, not just Iceland.
You are ignorant of facts and fallacies, rolling them all up into one.
So far you've tried to weasel around your incorrect statements ("preventing is not stopping!"), and made another unsubstantiated claim (geothermal may be "available" everywhere, but it is impractical many places without using hydraulic fracturing, which is precisely the method that has unknown but nonzero risk). I can only cite what's known and try to inform others. It's clear you pay no heed to such things, so continue to believe what you want to believe.
Ah, looks like another Googler did a driveby downmod.
Alternatively, maybe people tired of you posting the same thing for the 3rd or 4th time in the same story, on like your 10th post.
Hmm, did you know, when I worked at Google personal property would disappear from my desk from time to time? A signed copy of a book by Daniel Lyons, a nice fleece jacket, couple of other things.
I've never lost anything from my desk, nor have any of my immediate coworkers, and I've been at Google longer than you were. Folks in the building have posted on mailing lists from time to time about lost items, but that was pretty much the same as when I was at Microsoft. It was far worse at my university. What kind of mythical 10k+ person organization do you work for now where everyone is 100% honest?
It is interesting that after the rampant dishonesty at Google and how it bothered you, you decided to move to the upstanding world of investment banking and HFT.
Kind of shocking to me at the time, you bet. But there's a pattern.
The pattern is that your experience at Google doesn't match anyone else I've ever talked to, including folks who left, and you seem to have some huge axe to grind. Did your project get cancelled? I used to believe what you said about Google, but then you described TGIF in a way quite different than reality, and after that I started noticing that almost everything you described about Google appeared to be embellished with excessive drama.
I wish you'd stop pretending to be an expert on Google. You've been gone from Google almost as long as you worked there, and you appear to have worked on a siloed project as you seem to know little about most projects or their development practices. You make statements of fact about the performance of executives you've never met, and you don't allow for the fact that people or their roles might have changed a little bit in the last 2.5 years.
To put it terms you'll better understand since you're a kernel person: You sound a lot like someone who worked on a kernel feature for a year or two, then left in a tizzy because they didn't get their way on something. They then go on to complain about how the whole Linux community is broken, and that their bad experience is the norm. The reality is that while there are problems, the main issue is that they didn't really bother to understand the community.
Google has certainly had its issues with open source relations, as was clear with Android kernel modifications and Chromium community relations. Asking folks to wait for Ice Cream for an integrated tablet+phone experience and using the latitude given by a BSD licence to do it is not one of those problems. Rubin is trying to avoid the android equivalent of GCC 2.96 (I'm curious what you think of that episode), as there is already past history to support that at least some handset manufacturers would behave exactly that way.
Apart from his recent Linux header copyright brouhaha (which, to be fair, has still not been properly addressed by the appropriate parties),
Linus himself has spoken on the issue, the FSF's stance is pretty clear from what they've said in the past (and haven't changed), it was discussed in detail on Groklaw, and a ZDNet contributor got a statement from a technology lawyer contradicting the claims of Naughton. What more do you wantt? -- a Ietter signed by all the 1000s of Linux contributors, living and dead?
his writings are pretty accurate,
Really? He's been wrong on almost all of his opinions ever since he tried to block the Sun sale to Oracle along with Monty Wiidenius based on a ridiculous interpretation of the GPL (hey that seems like a theme, doesn't it?). At the time he was smaked down by Eblen Moglen himself (see Groklaw's story).
and he does have a much greater knowledge of patent matters than the random slashdotter or AC.
That's a pretty low bar. It's clear he understands a lot, but he only reports strangely twisted facts so that it always comes out as something Microsoft would agree with. In the process he makes absurd statements that have been debunked repeatedly by folks with more knowledge or better track records such as the FSF or Groklaw. You can look on his own blog where he has trouble defending his position -- he has to pretend various other comments, threads, and stories don't exist to defend his position, while others are providing clear references and links.
In the end, he looks like a fool precisely because he can't say what is true, but what his apparent sponsor needs him to say, no matter what the evidence to the contrary.
I see no FUD if he keeps pointing out how patents affect free software, especially since he is talking about actual lawsuits in progress.
You don't seem to understand the definition of FUD. The best FUD is real facts, used irrelevantly or out of context. For example, it is FUD for a large car dealership to say: "Even though he's got a lower price, I'd be worried from Dealer X, as he has a small business. Did you know that many small businesses that size fail every year? They might fail too and not be able to live up to their support contract like we can. With such low prices, he's probably losing money right now and could fail at any time. But of course, make your own choice based on what you are comfortable with."
See, nothing there is actually untrue, it's just that such stats taken out of context give you no real data on how these *specific* businesses are doing. The idea is to put a kernel of doubt in the person hearing it, so they will go along with you while thinking they are making an informed choice.
As others have pointed out, Florian is clearly trying to imply an extrapolation using a scary red color, he starts his graph at *negative* six lawsuits, ignores that lawsuits take years on an OS only a few years old, and ignores the outcome or expected outcome of any lawsuit. The idea is simply to make you worry about using Android. Yet if you think about it, if you were to graph the total awards in *successful* patent lawsuits against Microsoft since 1978, that also will go up and to the right. Does that mean Windows is in danger, or just that there's a big successful pie and plenty of people with patents who want to try to grab a slice?
And when he brings up indemnity, how can you *not* remember SCO and their eerily similar (and ultimately unsuccessful) FUD campaign?
I like how you say we have no control over earthquakes, and then you cite the one power generation method known to increase the probability of earthquakes. It's even on the wikipedia page you linked. I think geothermal is a great source where available (Iceland, for example), but to tout it as our single energy savior is ignorant of the facts.
Traditional plant designs melt down, but many new designs would not. Of course we don't have any new designs in running plants since it's become basically impossible to build a new nuclear plant, and research has mostly dried up since there is so little chance of commercial implementation.
So, we're stuck with old designs in practice while new designs sit on the drawing board, and folks such as yourself telling us we can't build the new designs because the old ones are too dangerous.
Geothermal is currently 0.2% of world energy production. I'm sorry to burst your bubble, but we aren't going to be able to scale it up 500x, even when using controversial techniques such as enhanced geothermal. As you can see from the nature article, even experts at MIT say the most we're going to get is something like 10% of US energy that way, and that's if we find a good way to understand and deal with the earthquakes it can cause.
Geothermal may be part of the puzzle, just like PV or Wind, but none of those is a panacea, even when combined. We'll need something scalable and available everywhere for baseload. Right now that's either coal, oil, natural gas, or nuclear. Only one of those won't run out in the next 100 years, thus there are folks like me who support nuclear as the only currently practical option to scale up. At the same time, we should pour research into other low-carbon methods and use them where practical.
In the various news articles' rush to be sensationalist, they forgot to note that the low tax rate was only for EU income, because the EU has a tax loophole you can drive a bus through. In order to average 19% given the significant overseas income, that means the US tax rate is pretty high. With a higher tax rate on EU income, it doesn't mean any of that extra tax revenue would go anywhere except to the EU (so they can just change the regulations if they'd like that money from Google and other corporations).
Of course, some folks are angry that Google is not repatriating foreign income, but considering most of the costs for that income are also overseas (datacenters, network connectivity, and engineering offices), I don't see any reason that money should come to the US first (just to pay extra extra taxes and drop profits?). And anyway, that kind of geographically diversified operations and holdings are pretty much the definition of a multi-national corporation.
We don't live in a system where everything must be a law: http://en.wikipedia.org/wiki/Common_law For something like this, you're almost certainly looking for case law and legal precedent.
Sorry, I cannot quote such a source for you, as IANAL.
You didn't reply to skids' point at all. Nobody is claiming that h264, its patent pool, or its licensing method is a problem here. What they are complaining about is that MPEG-LA basically said "Hey everybody, our current pool is insufficient, so send us other possible VP8-related patents and we'll all collude to take down this new competitor." They also announced this in a very public way to generate a FUD effect.
There's a difference between operating as a patent pool to enable interoperability and simplify licencing, and acting as a patent-wielding cartel to stifle competition. Regulators are fine with agreements that promote active competition in a space, but they're not so happy when agreements try to prevent competition. This is why many standards bodies with patents explicitly set out RAND licencing practices -- with unequal license terms you'd try to maximize what you extract from weak members or licensees. But that can quickly be seen as harmful collusion if used against any potential competitor, and that could get you in hot water with anti-trust regulations. As far as what's going on here, regulators will investigate and decide if they want to pursue action.
Your post seems to have little to do with this immediate discussion. Maybe you don't read TFA, but at least read TFS: "...anti-trust investigation of MPEG-LA's purported efforts to prevent Google's VP8 codec from widespread adoption.". There's nothing about h.264 licensing there, yet that's your focus of your rebuttal. The only thing that matters at all regarding their existing codecs is that it makes them a competitor of those who are pushing VP8. Few would dispute that the two camps are competitors.
Just because you sell me something doesn't mean you have the right to my information.
You can opt out if you want to: When publishers use One Pass, which for now is limited to online newspapers and magazines, Google will also share the customer’s name, ZIP code and e-mail address, unless a user decides to opt out. http://venturebeat.com/2011/02/16/google-one-pass/
In the paper world subscriber lists have full addresses, which can be used to get masses of data from traditional direct mail databases. So zip+email is already an improvement, especially if you set up a special email account for subscriptions. When that isn't sufficient, you can opt out.
It's strange that folks think the 30% won't be passed on to the consumer one way or another (if not by price, via ads or decreases in real content). So in some sense you're paying 20% more so that you don't have to uncheck a "share my email" box.
Ah ok thanks for the correction. I had heard that that was a hardware option for GAFYD, but maybe there's nothing suitable available right now, or it isn't an option in the plain-gmail version. Hopefully some new demand will change that in short order:)
There are some hardware options, such as yubikey. Another alternative if you don't mind the extra weight is to find a used android phone -- since the app doesn't require a sim, you only need wifi to get it set up (actually if you wanted to you could use USB and install it directly).
Also, check around to see if there's a Symbian app that implements HOTP (RFC 4226), since that's what Google uses. I imagine that if there isn't one yet, there will be one if this becomes popular.
Google is getting this flak not because the need is not there, but because the implementation sucks
So if these are all unacceptable: (1) getting called at a fixed number (2) receiving an SMS (3) running a smartphone app on Android or iPhone which requires no network access (4) using a 3rd party hardware token (yubikey) (5) using a printed sheet of access codes
What the heck would your "preferred" system look like?
On top of that the summary ominously mentions that one day this optional feature may become mandatory.
TFA made that up; it's not in the official blog. That random speculation did its job though, by sounding scary and increasing page-views.
It's in the TFA, which cites the official Google blog post as its only source, but that in turn has no mention of this every becoming mandatory. Mandatory is just pure speculation at this point, and given that smartphones aren't *that* widespread yet, I really doubt gmail would be that stupid.
I really dislike how Slashdot never bothers to cite original sources, instead going with an N-th level removed blog post.
So if you were to steal my phone, you'd only have access until I noticed I lost the phone and revoked those keys from home. I'd still be able to get to gmail from a home computer using the 30-day per-computer 2-factor allowance, or by using one of the one-time-use keys printed out at setup.
It looks like someone has through this through a bit:)
Okay, I know this is/. and no one expects you to RTFA, but FFS, at least read the summary.
The new two-factor authentication system is a voluntary program right now, although it could become mandatory at some point in the future.
And I did the unthinkable and confirmed that it IS in fact, in TFA.
Now click through to the original blog post from Google itself, and look for any hint that this will ever be made mandatory. Nope, nothing there. I wonder if TFA's author was just trying to sound dramatic, like so many of the "news" sites out there.
I can play that game too: "Gmail is currently a free service, although it could have a monthly fee at some point in the future."
Sure, either could happen, but it's pure speculation.
Perhaps you missed the part about being able to optionally associate it with an IP for 30 days? Entering a 6-digit code 12 times a year doesn't sound too onerous to me.
Not true, if you install the Android app it works completely offline. It imports data via a 2D-barcode displayed in gmail when you set up 2-factor. You don't even need a phone with a working SIM.
The problem is that whatever they're doing to eliminate the so-called spam is also making the use of exacting search terminology useless - literally, 100%, useless. Good luck finding anything useful from a line of output from `dmesg` - you'll find a dozen unanswered messages from Ubuntu forums from 4 years ago, and not much else.
So exact string searches are useless because they turn up those exact strings, but it doesn't have the answer you want? I like searching for error messages too, but sometimes there simply isn't an answer online. Google can't fix that. We as users can help though, by posting to forums, blogs and whatnot when we do find a solution.
I know it's pretty non-obvious, but quoting a string and adding a + in front will make Google much more strict about string matching all of the words. It does strip out some symbols and numbers, but it is really hard to build a search engine without making some simplifications like that.
Let's try an example. Looking at my dmesg, I see this error: ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PCI0.LPC_.EC__.PUBS._STA] (Node ffff88007cb2e0e0), AE_TIME
Searching for that yields no results. With some search/computer skill, I look and notice that Node is a number that might simply be a memory location, and thus make it nearly impossible to find a full string match. If I drop that, I get a few results: ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PCI0.LPC_.EC__.PUBS._STA] AE_TIME
That gets a couple of results, but probably not enough to find a good answer. Now I try dropping the strange path-looking thing: ACPI Error (psparse-0537): Method parse/execution failed AE_TIME
This gets lots of results that seem to be related. Now, I may not find my answer there, but the search engine has done its job IMHO.
Daniel, you're starting to sound like the people who go on LKML and ask why the Kernel can't be rewritten in another language. I'm sure you understand that is crazy. Much like Linux's 8 million lines of code, Android is large at 11 million lines of code. All of the libraries and services a native app might use are not going to be rewritten in a point release. How else would you expect the system to work right now except as a mixed mode?
I personally detest Java, and would rather use either C++ or a real modern language rather than the "C++ 1.1" that Java is. That said, there is a whole lot of existing Java code out there, and lots of developers who are familiar with it (more than C/C++ now), in particular in the mobile space. One has to be a realist.
I hope that someday everything on Android is handled through kernel interfaces and native libraries, with LLVM there to allow a rich set of frontend programming languages while keeping binaries from being tied to specific processor instruction sets. Even if that does happen though, it's certainly not going to be overnight.
I was curious about stuxnet so I read the various stories (though many have quite flimsy evidence), and read up on gas centrifuges as used in enrichment.
Why would they use a PLC to control a centrifuge?
These things operate pretty close to their mechanical limits, and exact speed control is apparently necessary to make sure the forces applied match the gas flow rates and thermal gradient used. Mess up any of those and you won't get good separation. Also, the controllers are designed to power quickly through speeds that are harmonic with the resonant frequency of centrifuge or its housing.
Why was there no overspeed safety on the centrifuge?
The worm caused the controller to lie to the process monitoring system, so those safety controls would not apply. As I mentioned earlier, the centrifuges don't appear to have a lot of margin between nominal operation and mechanical limits, so their may not be any fail-safe speed limits on the device itself (instead it would just be in the controller). Additionally in a radiation environment it probably makes sense to put all controls on the controller and make the centrifuge itself a "dumb" assemblage.
This is just speculation on my part, but my guess is that instead of running the centrifuge at overspeed, the worm ran them at a resonant frequency, which would shake the centrifuge apart while staying within speed limits. Even if this is the case though, I doubt you'll see it in a news story since it is not something a typical journalist would understand, and if they did they might not want to explain it to their readers.
Is there proof that actually happened? This sounds more like bad design or BS.
The evidence right now is very circumstantial, but consistently points in the direction of a directed attack. Few parties have both the technical means and motive for an attack like this, so it seems likely to be a state-sponsored. It will be interesting to see if more information eventually comes to light.
Slashdot Mirror
I would add that advances in PV should make rooftop PV at the home/office very useful in the future energy grid, as a method for lopping to top off of peak load in warm climates, and sidestepping the bulk of the distribution problem. However as you point out, that will never replace the base load which a Nuclear or Coal plant can supply.
The future energy grid will need to use almost every tech we've got, but you're right that no combination of renewables can scale to meet the *whole* need in the next 20-40 years, and it's pretty clear we need to do something before then.
I said we have no control over stopping earthquakes.
Ok, so let's exercise our control over not starting earthquakes, mr. selective pedant. Do you think that when someone says they want to "stop crime" that they mean that they want to let the crimes start, and then stop them, rather than prevent them?
And I didn't say geothermal is "our single energy savior" - just a good replacement for nukes.
Cool, so you are ok with all the rest coming from coal and the other existing sources? We need the 20% of Nuclear energy to scale to 80%, not something else to take the current 20% and everything else stay the same. When other people pointed out that it is impractical for wind, hydro, or solar to replace everything, you say "what about geothermal", in which case that better be practical enough to replace coal.
Geothermal is available in nearly every country, in most parts of the US, not just Iceland.
Take a look at this map:
http://en.wikipedia.org/wiki/File:Geothermal_resource_map_US.png
Almost all the potential is in the western states.
You are ignorant of facts and fallacies, rolling them all up into one.
So far you've tried to weasel around your incorrect statements ("preventing is not stopping!"), and made another unsubstantiated claim (geothermal may be "available" everywhere, but it is impractical many places without using hydraulic fracturing, which is precisely the method that has unknown but nonzero risk). I can only cite what's known and try to inform others. It's clear you pay no heed to such things, so continue to believe what you want to believe.
Ah, looks like another Googler did a driveby downmod.
Alternatively, maybe people tired of you posting the same thing for the 3rd or 4th time in the same story, on like your 10th post.
Hmm, did you know, when I worked at Google personal property would disappear from my desk from time to time? A signed copy of a book by Daniel Lyons, a nice fleece jacket, couple of other things.
I've never lost anything from my desk, nor have any of my immediate coworkers, and I've been at Google longer than you were. Folks in the building have posted on mailing lists from time to time about lost items, but that was pretty much the same as when I was at Microsoft. It was far worse at my university. What kind of mythical 10k+ person organization do you work for now where everyone is 100% honest?
It is interesting that after the rampant dishonesty at Google and how it bothered you, you decided to move to the upstanding world of investment banking and HFT.
Kind of shocking to me at the time, you bet. But there's a pattern.
The pattern is that your experience at Google doesn't match anyone else I've ever talked to, including folks who left, and you seem to have some huge axe to grind. Did your project get cancelled? I used to believe what you said about Google, but then you described TGIF in a way quite different than reality, and after that I started noticing that almost everything you described about Google appeared to be embellished with excessive drama.
I wish you'd stop pretending to be an expert on Google. You've been gone from Google almost as long as you worked there, and you appear to have worked on a siloed project as you seem to know little about most projects or their development practices. You make statements of fact about the performance of executives you've never met, and you don't allow for the fact that people or their roles might have changed a little bit in the last 2.5 years.
To put it terms you'll better understand since you're a kernel person: You sound a lot like someone who worked on a kernel feature for a year or two, then left in a tizzy because they didn't get their way on something. They then go on to complain about how the whole Linux community is broken, and that their bad experience is the norm. The reality is that while there are problems, the main issue is that they didn't really bother to understand the community.
Google has certainly had its issues with open source relations, as was clear with Android kernel modifications and Chromium community relations. Asking folks to wait for Ice Cream for an integrated tablet+phone experience and using the latitude given by a BSD licence to do it is not one of those problems. Rubin is trying to avoid the android equivalent of GCC 2.96 (I'm curious what you think of that episode), as there is already past history to support that at least some handset manufacturers would behave exactly that way.
Apart from his recent Linux header copyright brouhaha (which, to be fair, has still not been properly addressed by the appropriate parties),
Linus himself has spoken on the issue, the FSF's stance is pretty clear from what they've said in the past (and haven't changed), it was discussed in detail on Groklaw, and a ZDNet contributor got a statement from a technology lawyer contradicting the claims of Naughton. What more do you wantt? -- a Ietter signed by all the 1000s of Linux contributors, living and dead?
his writings are pretty accurate,
Really? He's been wrong on almost all of his opinions ever since he tried to block the Sun sale to Oracle along with Monty Wiidenius based on a ridiculous interpretation of the GPL (hey that seems like a theme, doesn't it?). At the time he was smaked down by Eblen Moglen himself (see Groklaw's story).
and he does have a much greater knowledge of patent matters than the random slashdotter or AC.
That's a pretty low bar. It's clear he understands a lot, but he only reports strangely twisted facts so that it always comes out as something Microsoft would agree with. In the process he makes absurd statements that have been debunked repeatedly by folks with more knowledge or better track records such as the FSF or Groklaw. You can look on his own blog where he has trouble defending his position -- he has to pretend various other comments, threads, and stories don't exist to defend his position, while others are providing clear references and links.
In the end, he looks like a fool precisely because he can't say what is true, but what his apparent sponsor needs him to say, no matter what the evidence to the contrary.
I see no FUD if he keeps pointing out how patents affect free software, especially since he is talking about actual lawsuits in progress.
You don't seem to understand the definition of FUD. The best FUD is real facts, used irrelevantly or out of context. For example, it is FUD for a large car dealership to say:
"Even though he's got a lower price, I'd be worried from Dealer X, as he has a small business. Did you know that many small businesses that size fail every year? They might fail too and not be able to live up to their support contract like we can. With such low prices, he's probably losing money right now and could fail at any time. But of course, make your own choice based on what you are comfortable with."
See, nothing there is actually untrue, it's just that such stats taken out of context give you no real data on how these *specific* businesses are doing. The idea is to put a kernel of doubt in the person hearing it, so they will go along with you while thinking they are making an informed choice.
As others have pointed out, Florian is clearly trying to imply an extrapolation using a scary red color, he starts his graph at *negative* six lawsuits, ignores that lawsuits take years on an OS only a few years old, and ignores the outcome or expected outcome of any lawsuit. The idea is simply to make you worry about using Android. Yet if you think about it, if you were to graph the total awards in *successful* patent lawsuits against Microsoft since 1978, that also will go up and to the right. Does that mean Windows is in danger, or just that there's a big successful pie and plenty of people with patents who want to try to grab a slice?
And when he brings up indemnity, how can you *not* remember SCO and their eerily similar (and ultimately unsuccessful) FUD campaign?
I hope it pays well, Florian.
I like how you say we have no control over earthquakes, and then you cite the one power generation method known to increase the probability of earthquakes. It's even on the wikipedia page you linked. I think geothermal is a great source where available (Iceland, for example), but to tout it as our single energy savior is ignorant of the facts.
Traditional plant designs melt down, but many new designs would not. Of course we don't have any new designs in running plants since it's become basically impossible to build a new nuclear plant, and research has mostly dried up since there is so little chance of commercial implementation.
So, we're stuck with old designs in practice while new designs sit on the drawing board, and folks such as yourself telling us we can't build the new designs because the old ones are too dangerous.
Geothermal is currently 0.2% of world energy production. I'm sorry to burst your bubble, but we aren't going to be able to scale it up 500x, even when using controversial techniques such as enhanced geothermal. As you can see from the nature article, even experts at MIT say the most we're going to get is something like 10% of US energy that way, and that's if we find a good way to understand and deal with the earthquakes it can cause.
Geothermal may be part of the puzzle, just like PV or Wind, but none of those is a panacea, even when combined. We'll need something scalable and available everywhere for baseload. Right now that's either coal, oil, natural gas, or nuclear. Only one of those won't run out in the next 100 years, thus there are folks like me who support nuclear as the only currently practical option to scale up. At the same time, we should pour research into other low-carbon methods and use them where practical.
From Google's Q4 2010 earnings report:
Income Taxes -- Our effective tax rate was 19% for the fourth quarter of 2010.
In the various news articles' rush to be sensationalist, they forgot to note that the low tax rate was only for EU income, because the EU has a tax loophole you can drive a bus through. In order to average 19% given the significant overseas income, that means the US tax rate is pretty high. With a higher tax rate on EU income, it doesn't mean any of that extra tax revenue would go anywhere except to the EU (so they can just change the regulations if they'd like that money from Google and other corporations).
Of course, some folks are angry that Google is not repatriating foreign income, but considering most of the costs for that income are also overseas (datacenters, network connectivity, and engineering offices), I don't see any reason that money should come to the US first (just to pay extra extra taxes and drop profits?). And anyway, that kind of geographically diversified operations and holdings are pretty much the definition of a multi-national corporation.
We don't live in a system where everything must be a law:
http://en.wikipedia.org/wiki/Common_law
For something like this, you're almost certainly looking for case law and legal precedent.
Sorry, I cannot quote such a source for you, as IANAL.
You didn't reply to skids' point at all. Nobody is claiming that h264, its patent pool, or its licensing method is a problem here. What they are complaining about is that MPEG-LA basically said "Hey everybody, our current pool is insufficient, so send us other possible VP8-related patents and we'll all collude to take down this new competitor." They also announced this in a very public way to generate a FUD effect.
There's a difference between operating as a patent pool to enable interoperability and simplify licencing, and acting as a patent-wielding cartel to stifle competition. Regulators are fine with agreements that promote active competition in a space, but they're not so happy when agreements try to prevent competition. This is why many standards bodies with patents explicitly set out RAND licencing practices -- with unequal license terms you'd try to maximize what you extract from weak members or licensees. But that can quickly be seen as harmful collusion if used against any potential competitor, and that could get you in hot water with anti-trust regulations. As far as what's going on here, regulators will investigate and decide if they want to pursue action.
Your post seems to have little to do with this immediate discussion. Maybe you don't read TFA, but at least read TFS: "...anti-trust investigation of MPEG-LA's purported efforts to prevent Google's VP8 codec from widespread adoption.". There's nothing about h.264 licensing there, yet that's your focus of your rebuttal. The only thing that matters at all regarding their existing codecs is that it makes them a competitor of those who are pushing VP8. Few would dispute that the two camps are competitors.
Just because you sell me something doesn't mean you have the right to my information.
You can opt out if you want to:
When publishers use One Pass, which for now is limited to online newspapers and magazines, Google will also share the customer’s name, ZIP code and e-mail address, unless a user decides to opt out.
http://venturebeat.com/2011/02/16/google-one-pass/
In the paper world subscriber lists have full addresses, which can be used to get masses of data from traditional direct mail databases. So zip+email is already an improvement, especially if you set up a special email account for subscriptions. When that isn't sufficient, you can opt out.
It's strange that folks think the 30% won't be passed on to the consumer one way or another (if not by price, via ads or decreases in real content). So in some sense you're paying 20% more so that you don't have to uncheck a "share my email" box.
Ah ok thanks for the correction. I had heard that that was a hardware option for GAFYD, but maybe there's nothing suitable available right now, or it isn't an option in the plain-gmail version. Hopefully some new demand will change that in short order :)
There are some hardware options, such as yubikey. Another alternative if you don't mind the extra weight is to find a used android phone -- since the app doesn't require a sim, you only need wifi to get it set up (actually if you wanted to you could use USB and install it directly).
Also, check around to see if there's a Symbian app that implements HOTP (RFC 4226), since that's what Google uses. I imagine that if there isn't one yet, there will be one if this becomes popular.
Good luck, and no you can't have my Nexus S :)
If you don't trust the app, inspect the source here and compile it yourself:
http://code.google.com/p/google-authenticator/
If you don't trust the compiler, get a yubikey which implements the same standard.
If you don't trust a 3rd party vendor, implement something for RFC-4226 yourself:
http://tools.ietf.org/html/rfc4226
If you still don't trust that, I suggest you get a different email provider :)
Google is getting this flak not because the need is not there, but because the implementation sucks
So if these are all unacceptable:
(1) getting called at a fixed number
(2) receiving an SMS
(3) running a smartphone app on Android or iPhone which requires no network access
(4) using a 3rd party hardware token (yubikey)
(5) using a printed sheet of access codes
What the heck would your "preferred" system look like?
On top of that the summary ominously mentions that one day this optional feature may become mandatory.
TFA made that up; it's not in the official blog. That random speculation did its job though, by sounding scary and increasing page-views.
You don't need phone, SMS, or wifi access if you have a smartphone. The authenticator app works completely offline.
It's in the TFA, which cites the official Google blog post as its only source, but that in turn has no mention of this every becoming mandatory. Mandatory is just pure speculation at this point, and given that smartphones aren't *that* widespread yet, I really doubt gmail would be that stupid.
I really dislike how Slashdot never bothers to cite original sources, instead going with an N-th level removed blog post.
Note the added support for "application specific passwords" on the blog post:
http://googleblog.blogspot.com/2011/02/advanced-sign-in-security-for-your.html
Now you can make a single-use revocable password for your phone, and it no longer needs to know your actual gmail password.
So if you were to steal my phone, you'd only have access until I noticed I lost the phone and revoked those keys from home. I'd still be able to get to gmail from a home computer using the 30-day per-computer 2-factor allowance, or by using one of the one-time-use keys printed out at setup.
It looks like someone has through this through a bit :)
Okay, I know this is /. and no one expects you to RTFA, but FFS, at least read the summary.
The new two-factor authentication system is a voluntary program right now, although it could become mandatory at some point in the future.
And I did the unthinkable and confirmed that it IS in fact, in TFA.
Now click through to the original blog post from Google itself, and look for any hint that this will ever be made mandatory. Nope, nothing there. I wonder if TFA's author was just trying to sound dramatic, like so many of the "news" sites out there.
I can play that game too: "Gmail is currently a free service, although it could have a monthly fee at some point in the future."
Sure, either could happen, but it's pure speculation.
Perhaps you missed the part about being able to optionally associate it with an IP for 30 days? Entering a 6-digit code 12 times a year doesn't sound too onerous to me.
Not true, if you install the Android app it works completely offline. It imports data via a 2D-barcode displayed in gmail when you set up 2-factor. You don't even need a phone with a working SIM.
Go here and click on "permissions":
https://market.android.com/details?id=com.google.android.apps.authenticator
Note that no internet or phone access is listed.
The problem is that whatever they're doing to eliminate the so-called spam is also making the use of exacting search terminology useless - literally, 100%, useless. Good luck finding anything useful from a line of output from `dmesg` - you'll find a dozen unanswered messages from Ubuntu forums from 4 years ago, and not much else.
So exact string searches are useless because they turn up those exact strings, but it doesn't have the answer you want? I like searching for error messages too, but sometimes there simply isn't an answer online. Google can't fix that. We as users can help though, by posting to forums, blogs and whatnot when we do find a solution.
I know it's pretty non-obvious, but quoting a string and adding a + in front will make Google much more strict about string matching all of the words. It does strip out some symbols and numbers, but it is really hard to build a search engine without making some simplifications like that.
Let's try an example. Looking at my dmesg, I see this error:
ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PCI0.LPC_.EC__.PUBS._STA] (Node ffff88007cb2e0e0), AE_TIME
Searching for that yields no results. With some search/computer skill, I look and notice that Node is a number that might simply be a memory location, and thus make it nearly impossible to find a full string match. If I drop that, I get a few results:
ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PCI0.LPC_.EC__.PUBS._STA] AE_TIME
That gets a couple of results, but probably not enough to find a good answer. Now I try dropping the strange path-looking thing:
ACPI Error (psparse-0537): Method parse/execution failed AE_TIME
This gets lots of results that seem to be related. Now, I may not find my answer there, but the search engine has done its job IMHO.
Daniel, you're starting to sound like the people who go on LKML and ask why the Kernel can't be rewritten in another language. I'm sure you understand that is crazy. Much like Linux's 8 million lines of code, Android is large at 11 million lines of code. All of the libraries and services a native app might use are not going to be rewritten in a point release. How else would you expect the system to work right now except as a mixed mode?
I personally detest Java, and would rather use either C++ or a real modern language rather than the "C++ 1.1" that Java is. That said, there is a whole lot of existing Java code out there, and lots of developers who are familiar with it (more than C/C++ now), in particular in the mobile space. One has to be a realist.
I hope that someday everything on Android is handled through kernel interfaces and native libraries, with LLVM there to allow a rich set of frontend programming languages while keeping binaries from being tied to specific processor instruction sets. Even if that does happen though, it's certainly not going to be overnight.
I was curious about stuxnet so I read the various stories (though many have quite flimsy evidence), and read up on gas centrifuges as used in enrichment.
Why would they use a PLC to control a centrifuge?
These things operate pretty close to their mechanical limits, and exact speed control is apparently necessary to make sure the forces applied match the gas flow rates and thermal gradient used. Mess up any of those and you won't get good separation. Also, the controllers are designed to power quickly through speeds that are harmonic with the resonant frequency of centrifuge or its housing.
Why was there no overspeed safety on the centrifuge?
The worm caused the controller to lie to the process monitoring system, so those safety controls would not apply. As I mentioned earlier, the centrifuges don't appear to have a lot of margin between nominal operation and mechanical limits, so their may not be any fail-safe speed limits on the device itself (instead it would just be in the controller). Additionally in a radiation environment it probably makes sense to put all controls on the controller and make the centrifuge itself a "dumb" assemblage.
This is just speculation on my part, but my guess is that instead of running the centrifuge at overspeed, the worm ran them at a resonant frequency, which would shake the centrifuge apart while staying within speed limits. Even if this is the case though, I doubt you'll see it in a news story since it is not something a typical journalist would understand, and if they did they might not want to explain it to their readers.
Is there proof that actually happened? This sounds more like bad design or BS.
The evidence right now is very circumstantial, but consistently points in the direction of a directed attack. Few parties have both the technical means and motive for an attack like this, so it seems likely to be a state-sponsored. It will be interesting to see if more information eventually comes to light.
Polygamy?