I'm there under a pseudonym for exactly that reason. I detest with a passion their real-names-only policy, and their way of forcing you to engage on every forum & on every topic with the same identity & tying it back to your entire social graph and history. It's the anti-Internet:(
This is part of a bigger play by Facebook.
Most mail accounts in use right now are password SMTP over TLS/SSL. Yet most services on the net assume that people are in full control of their primary mailboxes.
By going multi-factor on their login system, Facebook wants to establish their messaging system as a more secure, more trusted endpoint (especially for the average user with zero understanding of password hygiene) than good old email. Once they do so, and get their users trained up softly-softly on multi-factor authentication, they then quietly pitch to organizations and service providers (banks, government services, utilities,...) to request Facebook, rather than email, as the preferred primary mechanism for staying in touch with customers.
After all, if Facebook accounts are harder to spoof than an email address -- and with the continual life history & social graph data they contain, they surely are -- why wouldn't an organization want to stay in touch with its customers that way? From the point of view of a big org concerned with identity theft and fraud prevention, it's surely a tempting way to arrange things. Facebook owns your digital identity and theirs, phishing becomes much more difficult to execute as senders are authenticated & easily verified.
Post 9/11, two years or so after the accident, they were brought back in to commercial service (with additional safety modifications - Kevlar lining in the fuel tanks being the main one) for a couple of years. However, they'd become uneconomical (never mind now with $100/barrel oil) and as parts were no longer made, they were having to cannibalize some planes for spares.
Concorde didn't have the range to go cross the Pacific - to even go trans-Atlantic, it had to be given landing priority at the airports it serviced. The major trans-Pacific routes are a good 50% longer than trans-Atlantic.
What it takes to secure a (user desktop) system is:- - Not leaving ports open unless they absolutely need to be - Not running Outlook Express and randomly downloaded crapware - Using a reasonably secure browser, and avoiding popup-infested porn and warez sites - Not running britney_spears.exe when it arrives in your inbox.. etc.
A secure permissions model has very little to do with it, IMHO.
"the day they cure aids" is a nice idea, thing is it probably won't work out like that. It's gone from "Aids is a death sentence" to "Aids is a lifetime on pretty awful drugs that'll probably take a decade or two off your life", in a few years it'll be "Aids is a lifetime on slightly less awful drugs that'll take 5yrs off your life" or "we've got a vaccine but it only works 80% of the time, and then only on two or three common strains". Already youth here (UK) seem to regard it as less of a mortal threat than 10-15 years ago.. hence rocketing rates of syphilis and lesser STDs.
Interesting point of view, but perhaps a bit optimistic:- human lifespan is already up at the top of the range for mammals -- even if the 'turning knob' can be fixed (double or triple the maximum cell division count, and suitably increase the metabolic / nutritional budget for tumor suppression to compensate), our evolutionary line has had an upper limit of a hundred years or so since the earliest mammals evolved 100m years ago -- nothing that we've evolved in that time has built for multi-century endurance. I'd wager that it'd be substantially easier to extend the life of a rat to 20 years, or a dog to 50, than to get a human being to 200.
Germany's courts and officials are noted for this kind of thing. Very easy there to get somebody's show booth shut down with scant allegations of trademark or copyright infringement.
The primary problem for orbital flight isn't altitude, it's delta-V.. and even if you have a platform cruising at 33KM at perhaps Mach 2, you still need a heck of a lot of fuel to get to orbit. A staged rocket uses only a relatively small proportion of its fuel in getting to 33KM or Mach 2.
I disagree. If the likes of the Gates foundation can go some way to fixing the problems of the developing world (not to mention the US school system), in the long term that will do more for the space programme than investing $billions in todays' rocket scientists. Right now, a lack of realistic breakthrough ideas is holding space exploration up more than a lack of funds. If humanity really is to go places in space, we need to be thinking about the problems of long-duration interstellar travel, not blowing vast sums on touch-and-go men-on-Mars projects. If we want to move forward, we need to get the other 9/10ths of humanity (you know, the ones without clean water or schools) involved. Right now there's a hell of a lot of untapped potential out in the world.. to keep things moving forward, we're going to need it. It's a long way off, but imagine today's developing world producing scientific discoveries at the same rate per-capita that the US, Japan and NW Europe do today.
I was particularly thinking of the behaviour of the Installer when dealing with non-trivial installs. Also parts of the Window Manager, and the interaction between Carbon and Cocoa windows in mixed apps.
Yeah, right. At system level, there's still so much broken with OS X -- in some ways it's as bad as Windows 3.1
"Non-blocking" sync primitives that actually block? Check.
Huge chunks of the API that are not thread safe, and will crash if you call them from two threads at once? Check.
API calls that have only one or two lines of documentation, and even then don't actually do what that documentation says they'll do? Check.
Parts of the API that are in such constant flux that you can't rely on them from one OS revision to the next? Check.
A permissions model that, as implemented, is far too complex and fragile for single-user or family desktop machines, and breaks very easily? Check.
Major system components that appear to have only three or four engineers working on them, with little in the way of oversight? Check.
Sure, the user experience design of OS X, especially in relation to 'newbie' user experience, is excellent. What they've acheived with the resources they have is fairly impressive. Once you start using their APIs, though, it's chewing-gum-and-string compared to Win32 or Linux... they do some impressive stuff, but the comparative lack of engineering resources has to bite somewhere, and one of those places is API consistency, quality and developer documentation.
It's got to be way off -- chemical engine technology won't get you anywhere near 1/3c, even launching a Sputnik-sized probe with a Saturn-sized rocket.
A quick Google suggests that the Voyager probes are travelling at about 1/20000c; New Horizons is somewhat faster again, but will still need ten years to get to Pluto (a distance of 5 light hours). My guess would be, if we used our largest available rocket and the smallest available payload capable of phoning home over multi light year distances, it might be possible to improve on that by a factor of 20, but you're still looking at millennia to get to the nearest star.
Nuclear fusion or matter / antimatter reactions are likely the only way to get anywhere near 1/3c.
Concorde only used afterburner at take-off and acceleration to supersonic. Once it reached cruising speed (mach 2) they could turn the afterburner off.
You have to take in to account that fossilization, followed by intact survival and subsequent discovery, makes for an *extremely* rare set of events.
There are a few thousand good-condition dinosaur skeletons now existing in human ownership worldwide. Dinosaurs ruled the earth for about 200M years -- if we assume that they lived on average 10 years and the world supported an average of just 1M individuals at any given time (and it was probably more like 100M), you are looking at 20 trillion individuals during the dinosaur era.
So, even given those conservative figures, our recovery rate so far is less than 1 individual in a billion, probably lower -- apply the same figures to the human population, and you'd have a tiny handful of "freak ape" skeletons found in a couple of sites worldwide. There'd perhaps be a controversy as to whether the remains even represented a species of their own, or a gorilla with hydrocephalus and other defects.
The problem is not whether or not it's possible, but whether it's feasible for a development team to do it well enough (for Mac users, who expect much higher standards of such things than Linux users) and in a short enough time frame.
The vast majority of the Win32 API on Windows also runs in userland. The APIs that userland Win32 uses to talk to drivers look pretty different to the APIs that most application developers are using.
I don't know about Google, but the Internet as a whole certainly does. It's not surprising that people are finding it hard to see the next big thing -- we're most likely at the tail end of the IT revolution -- there are surely still advances to be made, but they'll probably not compare to the advances made between about 1970 and 2000.. in the same way that aircraft have advanced in the 35 or so years since the 747 and Concorde first flew, but not at anything like the pace at which they advanced in the previous 35 years.
I think people are almost certainly right in looking at medicine and biotech for the next big thing -- there seems to be far more potential for useful progress there than in large-scale engineering (be it space, transport, civil...), or in IT. Augmentation will be big (indeed, it IS big already -- personal mobile telephones are essentially used that way by many, even if they're not yet permanent attachments drawing power from the human body).
Never mind that.. imagine the kind of conversational skills and knowledge a wireless Google wetware implant would give you? You'd be rid forever of that "who?" moment when the girl you're chatting up starts talking about how great some movie director or fashion designer you've never heard of is. Never fail another exam, get lost in another town, or be unable to find somewhere still serving drinks at 12pm.
Of course, whether you'd want those amoral masters of search having access to your innermost thoughts is an entirely different matter (in Soviet Russia, Google indexes YOU!) - would you trust their guarantee that the device is one-way-only?
Windows "has problems", and OS X "has few", mostly because there are still 50 times as many people using Windows as OS X.
That means most h/crackers, spyware authors and virus writers target Windows -- and whilst the BSD layer of OS X is secure and rock-solid, the upper layers (Carbon etc.) are anything but. Plus the fact that most of the l33t kids on Mac are fanboys, where many on Windows love to hate M$.
Believe me, as a developer, trying to keep a complex piece of software working right on every OS X sub-release from 10.2.8 to the present is a nightmare. As to the documentation.. let's just put it this way:- M$ has ten times the resources of Apple, and when it comes to API documentation, it shows!
FWIW, I've never had hardware-related problems on properly built Windows machines (I'm not talking about $299 junk PCs).. also, the Mac hardware has changed a great deal over the past few years, while there aren't as many designs as there are PC motherboards, there's still quite a few.
True, but it makes it easier than ever for them to do spot checks.. one person sat at a desk could get through a few thousand vehicles per day. Also, whilst doing optical recognition of a particular model is (relatively) hard, colour is easy.
And no doubt, once this is in place, they'll make the penalties for having fake plates sufficiently stiff that no petty crook is going to want to drive around with them on a day-to-day basis.
... and despite all that redundancy and safety, two have been lost in around a hundred missions.
If space travel is to be scaled up, and space tourism to catch on, we certainly can't afford to have it any -less- safe -- how many people would fly commercial aircraft if one in 100 airline flights ended in a fatal accident (as opposed to of the order of one in a million)? OK, so space tourism is a bleeding-edge, once-in-a-lifetime experience, but still - a safety record worse than one fatal accident every 1000 flights (roughly equivalent to the very early days of airlines c. 1930) is not going to win a lot of business, and no existing manned rocket system has gotten close to that.
Slashdot Mirror
I'm there under a pseudonym for exactly that reason. I detest with a passion their real-names-only policy, and their way of forcing you to engage on every forum & on every topic with the same identity & tying it back to your entire social graph and history. It's the anti-Internet :(
This is part of a bigger play by Facebook. Most mail accounts in use right now are password SMTP over TLS/SSL. Yet most services on the net assume that people are in full control of their primary mailboxes. By going multi-factor on their login system, Facebook wants to establish their messaging system as a more secure, more trusted endpoint (especially for the average user with zero understanding of password hygiene) than good old email. Once they do so, and get their users trained up softly-softly on multi-factor authentication, they then quietly pitch to organizations and service providers (banks, government services, utilities, ...) to request Facebook, rather than email, as the preferred primary mechanism for staying in touch with customers.
After all, if Facebook accounts are harder to spoof than an email address -- and with the continual life history & social graph data they contain, they surely are -- why wouldn't an organization want to stay in touch with its customers that way? From the point of view of a big org concerned with identity theft and fraud prevention, it's surely a tempting way to arrange things. Facebook owns your digital identity and theirs, phishing becomes much more difficult to execute as senders are authenticated & easily verified.
Post 9/11, two years or so after the accident, they were brought back in to commercial service (with additional safety modifications - Kevlar lining in the fuel tanks being the main one) for a couple of years. However, they'd become uneconomical (never mind now with $100/barrel oil) and as parts were no longer made, they were having to cannibalize some planes for spares.
Concorde didn't have the range to go cross the Pacific - to even go trans-Atlantic, it had to be given landing priority at the airports it serviced. The major trans-Pacific routes are a good 50% longer than trans-Atlantic.
What it takes to secure a (user desktop) system is:- .. etc.
- Not leaving ports open unless they absolutely need to be
- Not running Outlook Express and randomly downloaded crapware
- Using a reasonably secure browser, and avoiding popup-infested porn and warez sites
- Not running britney_spears.exe when it arrives in your inbox
A secure permissions model has very little to do with it, IMHO.
"the day they cure aids" is a nice idea, thing is it probably won't work out like that. It's gone from "Aids is a death sentence" to "Aids is a lifetime on pretty awful drugs that'll probably take a decade or two off your life", in a few years it'll be "Aids is a lifetime on slightly less awful drugs that'll take 5yrs off your life" or "we've got a vaccine but it only works 80% of the time, and then only on two or three common strains". Already youth here (UK) seem to regard it as less of a mortal threat than 10-15 years ago.. hence rocketing rates of syphilis and lesser STDs.
Interesting point of view, but perhaps a bit optimistic:- human lifespan is already up at the top of the range for mammals -- even if the 'turning knob' can be fixed (double or triple the maximum cell division count, and suitably increase the metabolic / nutritional budget for tumor suppression to compensate), our evolutionary line has had an upper limit of a hundred years or so since the earliest mammals evolved 100m years ago -- nothing that we've evolved in that time has built for multi-century endurance. I'd wager that it'd be substantially easier to extend the life of a rat to 20 years, or a dog to 50, than to get a human being to 200.
Germany's courts and officials are noted for this kind of thing. Very easy there to get somebody's show booth shut down with scant allegations of trademark or copyright infringement.
Rock solid maybe (I have one, seems fine) but why the heck did they have to remove the VESA mount that was present on the older flat iMacs :(
The primary problem for orbital flight isn't altitude, it's delta-V.. and even if you have a platform cruising at 33KM at perhaps Mach 2, you still need a heck of a lot of fuel to get to orbit. A staged rocket uses only a relatively small proportion of its fuel in getting to 33KM or Mach 2.
.. except that Cell is completely unsuitable for use as a desktop CPU.
For games consoles with dedicated software? Perhaps.
For scientific computing and HPC? Sure.
As an off-board number cruncher and accelerator chip? Yup.
As a desktop? Heck no, a multi-core x86 or indeed PPC knocks it in to a cocked hat.
BTW, I own both a dual 2GHz G5 and a dual-1.8 iMacIntel. The intel box smokes the G5 by a long distance.
I disagree. If the likes of the Gates foundation can go some way to fixing the problems of the developing world (not to mention the US school system), in the long term that will do more for the space programme than investing $billions in todays' rocket scientists. Right now, a lack of realistic breakthrough ideas is holding space exploration up more than a lack of funds. If humanity really is to go places in space, we need to be thinking about the problems of long-duration interstellar travel, not blowing vast sums on touch-and-go men-on-Mars projects. If we want to move forward, we need to get the other 9/10ths of humanity (you know, the ones without clean water or schools) involved. Right now there's a hell of a lot of untapped potential out in the world.. to keep things moving forward, we're going to need it. It's a long way off, but imagine today's developing world producing scientific discoveries at the same rate per-capita that the US, Japan and NW Europe do today.
I was particularly thinking of the behaviour of the Installer when dealing with non-trivial installs. Also parts of the Window Manager, and the interaction between Carbon and Cocoa windows in mixed apps.
Yeah, right. At system level, there's still so much broken with OS X -- in some ways it's as bad as Windows 3.1
"Non-blocking" sync primitives that actually block? Check.
Huge chunks of the API that are not thread safe, and will crash if you call them from two threads at once? Check.
API calls that have only one or two lines of documentation, and even then don't actually do what that documentation says they'll do? Check.
Parts of the API that are in such constant flux that you can't rely on them from one OS revision to the next? Check.
A permissions model that, as implemented, is far too complex and fragile for single-user or family desktop machines, and breaks very easily? Check.
Major system components that appear to have only three or four engineers working on them, with little in the way of oversight? Check.
Sure, the user experience design of OS X, especially in relation to 'newbie' user experience, is excellent. What they've acheived with the resources they have is fairly impressive. Once you start using their APIs, though, it's chewing-gum-and-string compared to Win32 or Linux... they do some impressive stuff, but the comparative lack of engineering resources has to bite somewhere, and one of those places is API consistency, quality and developer documentation.
It's got to be way off -- chemical engine technology won't get you anywhere near 1/3c, even launching a Sputnik-sized probe with a Saturn-sized rocket.
A quick Google suggests that the Voyager probes are travelling at about 1/20000c; New Horizons is somewhat faster again, but will still need ten years to get to Pluto (a distance of 5 light hours). My guess would be, if we used our largest available rocket and the smallest available payload capable of phoning home over multi light year distances, it might be possible to improve on that by a factor of 20, but you're still looking at millennia to get to the nearest star.
Nuclear fusion or matter / antimatter reactions are likely the only way to get anywhere near 1/3c.
Yep, although the origin is, again, French...
2 0words%20for%20animals%2C%20French%20words%20for%2 0meat
http://everything2.com/index.pl?node=Anglo-Saxon%
("Snail" is presumably related to the German "Schnecke").
Concorde only used afterburner at take-off and acceleration to supersonic. Once it reached cruising speed (mach 2) they could turn the afterburner off.
You have to take in to account that fossilization, followed by intact survival and subsequent discovery, makes for an *extremely* rare set of events.
There are a few thousand good-condition dinosaur skeletons now existing in human ownership worldwide. Dinosaurs ruled the earth for about 200M years -- if we assume that they lived on average 10 years and the world supported an average of just 1M individuals at any given time (and it was probably more like 100M), you are looking at 20 trillion individuals during the dinosaur era.
So, even given those conservative figures, our recovery rate so far is less than 1 individual in a billion, probably lower -- apply the same figures to the human population, and you'd have a tiny handful of "freak ape" skeletons found in a couple of sites worldwide. There'd perhaps be a controversy as to whether the remains even represented a species of their own, or a gorilla with hydrocephalus and other defects.
The problem is not whether or not it's possible, but whether it's feasible for a development team to do it well enough (for Mac users, who expect much higher standards of such things than Linux users) and in a short enough time frame.
Personally I think it's doubtful for that reason.
The vast majority of the Win32 API on Windows also runs in userland. The APIs that userland Win32 uses to talk to drivers look pretty different to the APIs that most application developers are using.
I don't know about Google, but the Internet as a whole certainly does. It's not surprising that people are finding it hard to see the next big thing -- we're most likely at the tail end of the IT revolution -- there are surely still advances to be made, but they'll probably not compare to the advances made between about 1970 and 2000.. in the same way that aircraft have advanced in the 35 or so years since the 747 and Concorde first flew, but not at anything like the pace at which they advanced in the previous 35 years.
I think people are almost certainly right in looking at medicine and biotech for the next big thing -- there seems to be far more potential for useful progress there than in large-scale engineering (be it space, transport, civil...), or in IT. Augmentation will be big (indeed, it IS big already -- personal mobile telephones are essentially used that way by many, even if they're not yet permanent attachments drawing power from the human body).
Never mind that.. imagine the kind of conversational skills and knowledge a wireless Google wetware implant would give you? You'd be rid forever of that "who?" moment when the girl you're chatting up starts talking about how great some movie director or fashion designer you've never heard of is. Never fail another exam, get lost in another town, or be unable to find somewhere still serving drinks at 12pm.
Of course, whether you'd want those amoral masters of search having access to your innermost thoughts is an entirely different matter (in Soviet Russia, Google indexes YOU!) - would you trust their guarantee that the device is one-way-only?
Windows "has problems", and OS X "has few", mostly because there are still 50 times as many people using Windows as OS X.
That means most h/crackers, spyware authors and virus writers target Windows -- and whilst the BSD layer of OS X is secure and rock-solid, the upper layers (Carbon etc.) are anything but. Plus the fact that most of the l33t kids on Mac are fanboys, where many on Windows love to hate M$.
Believe me, as a developer, trying to keep a complex piece of software working right on every OS X sub-release from 10.2.8 to the present is a nightmare. As to the documentation.. let's just put it this way:- M$ has ten times the resources of Apple, and when it comes to API documentation, it shows!
FWIW, I've never had hardware-related problems on properly built Windows machines (I'm not talking about $299 junk PCs).. also, the Mac hardware has changed a great deal over the past few years, while there aren't as many designs as there are PC motherboards, there's still quite a few.
True, but it makes it easier than ever for them to do spot checks.. one person sat at a desk could get through a few thousand vehicles per day. Also, whilst doing optical recognition of a particular model is (relatively) hard, colour is easy.
And no doubt, once this is in place, they'll make the penalties for having fake plates sufficiently stiff that no petty crook is going to want to drive around with them on a day-to-day basis.
... and despite all that redundancy and safety, two have been lost in around a hundred missions.
If space travel is to be scaled up, and space tourism to catch on, we certainly can't afford to have it any -less- safe -- how many people would fly commercial aircraft if one in 100 airline flights ended in a fatal accident (as opposed to of the order of one in a million)? OK, so space tourism is a bleeding-edge, once-in-a-lifetime experience, but still - a safety record worse than one fatal accident every 1000 flights (roughly equivalent to the very early days of airlines c. 1930) is not going to win a lot of business, and no existing manned rocket system has gotten close to that.