No, sorry. I heard it during a talk and checked the computation. You can check yourself: assume a strong radio transmitter (say, 2MW), spread the power out over the surface a sphere with 1-10ly, compute how much hits your antenna (say, 300m diameter), and compare with the noise floor. I think such a signal ends up being about 40-50dB below the minimum detectable level at 10 ly.
SETI can succeed for signals directed at us, and for enormously strong isotropic transmissions. But we're not going to pick up anything like human-like broadcasts, and human broadcasts have gotten less and less detectable since I Love Lucy.
No, probably not. It turns out that radio emissions don't even make it out of the Oort cloud before they become lost in background noise. It gets worse once a civilization starts using cellular technologies, mesh networking, compression, and encryption, because then signals require less and less power and look more and more like noise.
Set up a commission, give them unlimited fact-finding authority over the ISPs. They examine network load, operating costs, and approve new budgets and prices. Charter them to permit a steady, single-digit profit margin, while ensuring adequate ongoing investment and modernization.
We had such regulated monopolies in the past; they don't work, the regulators themselves end up being corrupt (influenced by the companies they are supposed to be regulating), and the companies get lazy. There are better ways of dealing with this.
No amount of ideology can give you a laissez faire market in wireless broadband.... Since we inevitably have to have a quasi-governmental broadband industry, I'm all for regulating it better.
Just because the bad regulation we have causes market failures doesn't mean that the fix is to turn broadband into a public utility. It isn't all that hard to fix the problems we have: force companies to split into those who provide cables/towers and those who provide services; outlaw any form of contractual lock-in; mandate equipment compatibility; ensure that spectrum and wires-in-the-ground are subject to competitive bidding every few years.
One of the few areas where Europe is really doing better than the US is in terms of wired and wireless Internet access. How? They didn't turn those companies into regulated monopolies as you suggest, they did many of the things I listed.
It's rather the intent of every single Randian faithiest to INSIST that any failure in the Free Market is due to government interference.
There are many different kinds of government interference, some of them good, a some of them bad. The failure in the US wireless industry is clearly due to government interference, more specifically, the wrong kind of government interference. The solution is not to abolish government interference altogether, but to replace the interference we have with something that works better.
In the case of wireless, the right kind of government interference is preventing the formation of huge corporations, forcing companies to adopt interoperable standards, and forcing companies to make it easy to switch carriers without losing money. In the case of wired, it may be splitting the providers of cables from the providers of services.
The wrong kind of government interference is the kind we have, which encourages consolidation into huge "can't fail" companies, allows customer lock-in, gives corporations rights to bandwidth and access effectively in perpetuity, and charges all sorts of extraneous fees that they can use to cushion their bottom line. And to fix these problems, people like you want to pile more bad regulation on top of already bad regulation.
We need better regulation. And once you have better regulation, you also end up having less of it, because you don't need to pile bad regulation on top of bad regulation to fix the problems that you created in the first place.
I believe that copyright law in Germany is less draconian than that in the US, so there is more freedom of speech in that arena in Germany than the US.
I'm curious: what does German copyright law let you do that US copyright law prohibits?
When a US company does business in a European country they have to comply with the laws in that country, yes. Facebook provides services directly aimed a German people via facebook.de.
Yes, and like any other European company, they have a right to challenge administrative decisions in court. That's all they have announced they will do.
And by what "definition of free speech" do you have more freedom of speech in Germany? I don't know of anything kind of speech that is legal in Germany but illegal in the US, but there are numerous examples of speech that are legal in the US but illegal in Germany.
And not being allowed to say whatever you want everywhere is not a limit, how, exactly?
What are you trying to say? The fact remains that speech that is legal in Germany is legal in the US, but speech that is legal in the US is not legal in Germany. Hence, there are fewer restrictions on free speech in the US than in Germany.
And the US has obscenity and defamation.. so.. not really free speech there neither.
There are all sorts of invalid laws on the book that conflict with the US Constitution; those laws are simply not enforceable anymore. So, citing one of those laws doesn't establish that there's a problem. Obscenity laws have largely been invalidated on Constitutional grounds. You can now pretty much say whatever you want (but there may be restrictions on where you can say it).
Defamation in the US is a civil matter, so it isn't a government restriction on speech. If your speech harms someone, you may have to compensate them for the harm, but the speech is still legal. Furthermore, truth is an absolute defense against defamation. In Germany, defamation is a criminal matter and you can defame people even if what you say is true.
Germany has strong restrictions on free speech, the US has almost none. Whether you like that situation is a different matter, but you can't argue that the two are similar.
Code reuse is possible because we carefully avoid making every piece interact with every other piece. We deliberately restrict the ability for small changes in one are to have global effects on the rest of the program, preferring to create small, self-contained modules with well-defined interfaces. DNA is just the opposite: a single huge parallel program, with patches layered on top of patches, and no organizing structure to be found anywhere
You can think of a DNA program like the output from an optimizing compiler: it will reuse storage locations and share blocks of code, but there may still be high level modularity that you simply don't see easily at the binary level.
An American company really believes they can force Germany to change their laws or allow Facebook to operate outside of the law?
No, they simply announced that they will fight a legal order in Germany that they believe is wrong, just like any other company has a right to.
Just WOW. What the hell kind of shenanigans are they pulling over here, then?
When US law enforcement orders US subsidiaries of European companies to comply with US law, you people complain about overbearing US law enforcement. When German law enforcement orders European subsidiaries of US companies to do something and they announce legal steps, you complain about overbearing US corporations. That's how companies and law enforcement interact in Western societies. Both law enforcement and companies are acting correctly in both cases, and these cases will get settled in the courts.
The only thing that is amazing about these cases is the degree of nationalism that Europeans display. Apparently you think that the entire rest of the world should obey European laws, while European companies can operate overseas free from the restraints of local laws.
It's immensely useful, actually. It saves a lot of money on storage, particularly when it's quasi-archival storage (ie an archive which is accessed frequently
And special uses like that have excellent support on Linux, for example via lessfs and DedupFS, in addition to various backup solutions.
It's just not been historically all that cost effective (from a computing resource viewpoint) to implement and not all that stable,
Yes, that's because it's a complex feature with a big impact on performance and lots of potential for bugs and data loss. And because 99.9% of users/uses don't need it, normal file systems don't offer this feature.
making even the inefficient "script and file based dedup" seem appealing.
Actually, script-based deduping is highly efficient, probably far more efficient than anything you can build into the file system. It simply isn't instantaneous. If you need instantaneous deduplication, you can use one of the FUSE systems.
As usual, Windows picked the wrong technical solution...
ZFS is the only FS to checksum data, ReFS checksums the metadata.
What problem does this actually solve? Has anybody demonstrated that this actually improves something measurably? Has anybody demonstrated that it doesn't lead to higher failure rates?
Lots of UNIX systems, as well as Linux, have offered pretty much everything you list. It's not part of the default Linux installs because they have turned out not to be all that useful, but impose a cost in terms of administrative complexity, failure rates, and/or performance.
Microsoft has a habit of pushing a lot of advanced-sounding features into their OS because that's what their employees get rewarded for and because the marketing department wants it. But just because it sounds advanced and complex doesn't make it an "enterprise feature".
The complexity and number of features of Windows is one of the reasons people prefer not to use it in many large-scale applications; there are just too many things that can go wrong in Windows, and it is too hard to disable features that get in the way.
Apparently, Microsoft still hasn't learned their lesson.
Um, if this is even remotely possible in your mind, you don't understand dedup. That, and you probably don't understand the word "passive", either.
Well, I certainly don't. NTFS doesn't support deduplication AFAIK. For Linux, on the other hand, there are several file systems that do, both natively and through FUSE, plus several user land solutions. Various UNIX systems have had deduplication for decades. The feature never caught on, probably because it actually isn't useful.
and the visual style of the window's title bar makes it blindingly obvious which window you're working with. Therefore, when you pull down a menu, you can instantly see what window will be the target of that action
With Mac-style menus, when you're looking at the menu bar, you're not looking at the title bar, and that's exactly the problem. You can rationalize as much as you like, my parents really did get confused by this (and I tripped over it too when I was still using a Mac).
As for not caring what application a window belongs to, that's only true for very simple, document-based apps.
Kurzweil started the company Kurzweil Computer Products, Inc. and led development of the first omni-font optical character recognition system
Well, see, Kurzweil's company did not develop "the first omni-font optical character recognition system", let alone the technology behind omni-font optical character recognition. And even if his company had been the first, it still wouldn't tell us what his personal technical contributions were.
My god, you are a lazy, dumb shit.
Apparently, you simply don't understand the difference between making a product and making a technical contribution. I'm asking about Kurzweil's personal technical contributions, not about his (clearly successful) entrepreneurial career.
Thanks for your response; I think it really completes the picture of the guy and the kind of people that constitute his fan base.
I would have said the reverse. The menu bar being at the top creates modality that makes it easy to discover which windows belonging to a given application. In the Windows/X11 world, trying to figure out which application a particular window came from can be a usability nightmare
People don't usually care what "application" a window belongs to; the fact that you care on the Mac is a holdover from the Mac's single tasking heritage (where the entire menu bar paradigm originated). What people do care about is that the menu entry they select operates on the document they are working on, and people get confused about that relationship on the Mac.
Or SSH or iChat/Messages screen sharing. The latter makes more sense for home use, IMO.
SSH isn't a good option because OSX command line administration is extremely obscure. iChat is mac specific.That points out another problem with switching to Mac: if you switch your parents, you really have to buy another Mac for yourself and set up Apple-related accounts and infrastructure everywhere. You can't maintain a Mac if you don't use one yourself, it is just too different.
I went down that road; bought a Mac for my parents and a MacBook and desktop for myself. It was a lot of work. In the end, the small benefits of OS X over Windows just didn't justify the big expense and work. A couple of machine generations later, my parents are on Linux, I'm back on Windows and Linux, and we're all a lot happier.
In my experience, switching people from Windows to Linux is a lot less work than switching them from Windows to Mac: pretty as it is, the Mac has just too many annoying differences and annoying little usability problems. My parents could never get used to global menus on the Mac, for example. And remote system management on the Mac is also harder (the best you can do is try and set up remote desktop access). And, of course, there is the obvious advantage that people using Linux can continue to use the hardware they are already used to.
(Besides, you seem to be off your Apple marketing script: I thought the party line among Mac folks was that Mac is UNIX but Linux is not.)
With right-to-work, people don't want pay for a union because they don't want to join the union. They don't want to be covered by union rules or union agreements. They just want to be completely left alone by the union.
Slashdot Mirror
You don't need C++11 to replace MOC. MOC was obsoleted by C++ features many years ago.
So, is it fully standard C++ now or do you still have to use their hokey preprocessor?
No, sorry. I heard it during a talk and checked the computation. You can check yourself: assume a strong radio transmitter (say, 2MW), spread the power out over the surface a sphere with 1-10ly, compute how much hits your antenna (say, 300m diameter), and compare with the noise floor. I think such a signal ends up being about 40-50dB below the minimum detectable level at 10 ly.
SETI can succeed for signals directed at us, and for enormously strong isotropic transmissions. But we're not going to pick up anything like human-like broadcasts, and human broadcasts have gotten less and less detectable since I Love Lucy.
No, probably not. It turns out that radio emissions don't even make it out of the Oort cloud before they become lost in background noise. It gets worse once a civilization starts using cellular technologies, mesh networking, compression, and encryption, because then signals require less and less power and look more and more like noise.
We had such regulated monopolies in the past; they don't work, the regulators themselves end up being corrupt (influenced by the companies they are supposed to be regulating), and the companies get lazy. There are better ways of dealing with this.
Just because the bad regulation we have causes market failures doesn't mean that the fix is to turn broadband into a public utility. It isn't all that hard to fix the problems we have: force companies to split into those who provide cables/towers and those who provide services; outlaw any form of contractual lock-in; mandate equipment compatibility; ensure that spectrum and wires-in-the-ground are subject to competitive bidding every few years.
One of the few areas where Europe is really doing better than the US is in terms of wired and wireless Internet access. How? They didn't turn those companies into regulated monopolies as you suggest, they did many of the things I listed.
There are many different kinds of government interference, some of them good, a some of them bad. The failure in the US wireless industry is clearly due to government interference, more specifically, the wrong kind of government interference. The solution is not to abolish government interference altogether, but to replace the interference we have with something that works better.
In the case of wireless, the right kind of government interference is preventing the formation of huge corporations, forcing companies to adopt interoperable standards, and forcing companies to make it easy to switch carriers without losing money. In the case of wired, it may be splitting the providers of cables from the providers of services.
The wrong kind of government interference is the kind we have, which encourages consolidation into huge "can't fail" companies, allows customer lock-in, gives corporations rights to bandwidth and access effectively in perpetuity, and charges all sorts of extraneous fees that they can use to cushion their bottom line. And to fix these problems, people like you want to pile more bad regulation on top of already bad regulation.
We need better regulation. And once you have better regulation, you also end up having less of it, because you don't need to pile bad regulation on top of bad regulation to fix the problems that you created in the first place.
I'm curious: what does German copyright law let you do that US copyright law prohibits?
Yes, and like any other European company, they have a right to challenge administrative decisions in court. That's all they have announced they will do.
And by what "definition of free speech" do you have more freedom of speech in Germany? I don't know of anything kind of speech that is legal in Germany but illegal in the US, but there are numerous examples of speech that are legal in the US but illegal in Germany.
What are you trying to say? The fact remains that speech that is legal in Germany is legal in the US, but speech that is legal in the US is not legal in Germany. Hence, there are fewer restrictions on free speech in the US than in Germany.
There are all sorts of invalid laws on the book that conflict with the US Constitution; those laws are simply not enforceable anymore. So, citing one of those laws doesn't establish that there's a problem. Obscenity laws have largely been invalidated on Constitutional grounds. You can now pretty much say whatever you want (but there may be restrictions on where you can say it).
Defamation in the US is a civil matter, so it isn't a government restriction on speech. If your speech harms someone, you may have to compensate them for the harm, but the speech is still legal. Furthermore, truth is an absolute defense against defamation. In Germany, defamation is a criminal matter and you can defame people even if what you say is true.
Germany has strong restrictions on free speech, the US has almost none. Whether you like that situation is a different matter, but you can't argue that the two are similar.
You can think of a DNA program like the output from an optimizing compiler: it will reuse storage locations and share blocks of code, but there may still be high level modularity that you simply don't see easily at the binary level.
No, they simply announced that they will fight a legal order in Germany that they believe is wrong, just like any other company has a right to.
When US law enforcement orders US subsidiaries of European companies to comply with US law, you people complain about overbearing US law enforcement. When German law enforcement orders European subsidiaries of US companies to do something and they announce legal steps, you complain about overbearing US corporations. That's how companies and law enforcement interact in Western societies. Both law enforcement and companies are acting correctly in both cases, and these cases will get settled in the courts.
The only thing that is amazing about these cases is the degree of nationalism that Europeans display. Apparently you think that the entire rest of the world should obey European laws, while European companies can operate overseas free from the restraints of local laws.
And special uses like that have excellent support on Linux, for example via lessfs and DedupFS, in addition to various backup solutions.
Yes, that's because it's a complex feature with a big impact on performance and lots of potential for bugs and data loss. And because 99.9% of users/uses don't need it, normal file systems don't offer this feature.
Actually, script-based deduping is highly efficient, probably far more efficient than anything you can build into the file system. It simply isn't instantaneous. If you need instantaneous deduplication, you can use one of the FUSE systems.
As usual, Windows picked the wrong technical solution...
What problem does this actually solve? Has anybody demonstrated that this actually improves something measurably? Has anybody demonstrated that it doesn't lead to higher failure rates?
Lots of UNIX systems, as well as Linux, have offered pretty much everything you list. It's not part of the default Linux installs because they have turned out not to be all that useful, but impose a cost in terms of administrative complexity, failure rates, and/or performance.
Microsoft has a habit of pushing a lot of advanced-sounding features into their OS because that's what their employees get rewarded for and because the marketing department wants it. But just because it sounds advanced and complex doesn't make it an "enterprise feature".
The complexity and number of features of Windows is one of the reasons people prefer not to use it in many large-scale applications; there are just too many things that can go wrong in Windows, and it is too hard to disable features that get in the way.
Apparently, Microsoft still hasn't learned their lesson.
Well, I certainly don't. NTFS doesn't support deduplication AFAIK. For Linux, on the other hand, there are several file systems that do, both natively and through FUSE, plus several user land solutions. Various UNIX systems have had deduplication for decades. The feature never caught on, probably because it actually isn't useful.
With Mac-style menus, when you're looking at the menu bar, you're not looking at the title bar, and that's exactly the problem. You can rationalize as much as you like, my parents really did get confused by this (and I tripped over it too when I was still using a Mac).
Yes, like the kind parents tend to run.
Well, see, Kurzweil's company did not develop "the first omni-font optical character recognition system", let alone the technology behind omni-font optical character recognition. And even if his company had been the first, it still wouldn't tell us what his personal technical contributions were.
Apparently, you simply don't understand the difference between making a product and making a technical contribution. I'm asking about Kurzweil's personal technical contributions, not about his (clearly successful) entrepreneurial career.
Thanks for your response; I think it really completes the picture of the guy and the kind of people that constitute his fan base.
then you stop providing services to them. It's as simple as that.
People don't usually care what "application" a window belongs to; the fact that you care on the Mac is a holdover from the Mac's single tasking heritage (where the entire menu bar paradigm originated). What people do care about is that the menu entry they select operates on the document they are working on, and people get confused about that relationship on the Mac.
SSH isn't a good option because OSX command line administration is extremely obscure. iChat is mac specific.That points out another problem with switching to Mac: if you switch your parents, you really have to buy another Mac for yourself and set up Apple-related accounts and infrastructure everywhere. You can't maintain a Mac if you don't use one yourself, it is just too different.
I went down that road; bought a Mac for my parents and a MacBook and desktop for myself. It was a lot of work. In the end, the small benefits of OS X over Windows just didn't justify the big expense and work. A couple of machine generations later, my parents are on Linux, I'm back on Windows and Linux, and we're all a lot happier.
Gosh, maybe you should express yourself more clearly then!
In my experience, switching people from Windows to Linux is a lot less work than switching them from Windows to Mac: pretty as it is, the Mac has just too many annoying differences and annoying little usability problems. My parents could never get used to global menus on the Mac, for example. And remote system management on the Mac is also harder (the best you can do is try and set up remote desktop access). And, of course, there is the obvious advantage that people using Linux can continue to use the hardware they are already used to.
(Besides, you seem to be off your Apple marketing script: I thought the party line among Mac folks was that Mac is UNIX but Linux is not.)
I didn't say it wasn't. I'm not trying to prove a point here.
I just would like to know what he has done technically.
With right-to-work, people don't want pay for a union because they don't want to join the union. They don't want to be covered by union rules or union agreements. They just want to be completely left alone by the union.