Really. Space-X has docked a test capsule to the ISS, and their first cargo delivery launches October 7th. Astronauts will follow on later flights.
NASA hasn't developed a successful new booster in 30 years, despite about three failed attempts. Space-X has a new booster that works. Arianespace has a new booster that works. NASA has old ICBM derivatives and a "Space Launch System" proposal.
Space-X has only 1800 employees. NASA still has 34,000. NASA does some good stuff, but it's far too big for what it accomplishes.
This has been talked up for a decade or two, but needs cost and capacity numbers.
There's also the painful fact that every reactor design that had anything mechanically non-trivial inside the reactor has been a flop. There have been two German pebble-bed reactors, both of which had pebble jams serious enough to cause major accidents with significant radiation leaks. Tsinghua University in China has one that's worked for a while, and that design is being scaled up. The Rongcheng Shidaowan Nuclear Power Plant, with two pebble-bed reactors, is under construction now. Completion in 2015. Maybe they can make it work. We won't really know until there are a few hundred reactor-years on that technology.
High temperature, gas-cooled reactors have been tried, but were troublesome. The only big one was Fort. St. Vrain, which had a lot of troubles with auxiliary equipment and corrosion. It only ran 10 years. No big safety issues, though; just high maintenance costs.
At the high end, there's 3D real time ultrasound. This builds up a 3D image from multiple scans from different directions. The ultrasound part of this is no more complex than the basic machines. The 3D part is knowing where the sensor is and software to build up a 3D image. 3D images are much easier to interpret than simple reflection images, and you can rotate them and look from another angle.
So what's needed is a handheld device with a position tracking system. You could probably kludge something together with a 6DOF INS (an accelerometer and rate gyro chip) and a mouse sensor to track movement along the body. The rest is software.
Here's a Swiss railroad clock in its native habitat, at
Cornavin station. There are clocks at regular intervals along platforms, and the second hands are, of course, in sync. It's part of the Swiss Railways branding - their stations tend to have a large, if not excessive, number of those clocks.
I had to opt out of Pacific Gas and Electric's online billing system and go back to paper bills when they changed their EULA to allow more "disclosure". If I just buy power from them, they're subject to regulatory rules, enforced by the California Public Utilities Commission. But they wanted me to sign up for an "online account", which isn't regulated. If you don't sign up, they're not allowed to redistribute your "smart meter" info. If you sign up, you've consented to distribution to "affiliates".
I strongly recommend opting out to PG&E customers (California, Nevada) who are concerned about privacy issues.
The last Microsoft Office product I bought was Word 97. I've been using OpenOffice, then LIbreOffice, since about 2002. It's a OK word processor, a mediocre but adequate spreadsheet, and a better draw program than Office. What's in Microsoft Office that a home user would need, let alone pay for monthly?
Yes, one new mall opened in 2006. That's it since then. The huge Xanadu mall in New Jersey that was supposed to open in 2009 didn't open. Two bankruptcies and a roof collapse later, opening is now scheduled for 2014. Maybe.
Amazon is taking over more and more of retail. They've already taken over books and DVDs. They took over and re-started Webvan. They're building local distribution centers for same day delivery. They bought Kiva Robotics so those warehouses won't need many people.
The glory years of store-based retail are over. No new enclosed mall has been built in the US in a decade. Most retail areas have vacant stores that will never be used again.
In a few years, retail will consist of recreational shopping and convenience stores. Everything else will be on line.
Then all we have to worry about is downtime. Read E. M. Forster's "The Machine Stops".
Netflix and Amazon don't need tracking of casual browsers, because they have real customers. They have, legitimately, information about what you knowingly bought from them. Businesses that have real sites that sell real stuff don't really need to track browsers, just customers.
Even Facebook doesn't need tracking of casual browsers, since, while they're intrusive, you clearly sign up with and log into Facebook. Google doesn't really need personalization; they were profitable just putting up ads that were relevant to the current search.
So, really, it's the junk sites that need this. Those with Google AdSense junk ads. Most entertainment sites. Slashdot. Crap like that. Getting rid of tracking would hurt them. We might lose some of them. No big loss.
He's right. The issue is not what Facebook shows. It's how the pages are put together. It takes work by a lot of servers to assemble each page. The user-facing servers send out queries to servers which check the feeds of everything being followed - friends, events, calendars, messages, applications - and create a page to display. This page is updated automatically if you keep it open. You can look at any of these items in more detail, and go back into their past if desired.
That's what managers do - follow many changing items superficially and look at some of them in detail. A management version might have feeds for shipments which missed their ship date, incoming orders, customer complaints, personnel absences, due dates for major supplier shipments, and other items of interest. Different users would be watching different things, some info would be available only to some users, and users would set what they wanted to see.
If you've ever used a Bloomberg terminal, it's a lot like that, but with worse graphics.
Facebook has a reasonable platform for that sort of thing. The back end is databases and message passing. The business logic and formatting is mostly in PHP (for which Facebook has a hard-code compiler, so it doesn't take forever). Facebook also has decent solutions to the "tell me if it changed without polling too often" problem.
The AS&E Z-backscatter systems are quite good. There's some X-ray exposure, and they're very expensive, but they're effective. They image by atomic number, and can distinguish elements. It's not just density any more, like transmission X-rays. This technology is much more effective (and intrusive) than microwave-based systems.
The machines need a huge number of X-ray detectors, are physically large, very expensive, and aren't that fast. The latest generation of machines are about 2/3 the size of the previous ones (which were so big that they wouldn't fit through standard doors) and are faster. But they're still big.
If the TSA is going to throw money at something, this is at least something that works.
The San Ysidro border US border checkpoint has several lanes of the big version for cars. This thing is the size of a drive-through carwash. It's too slow to use for primary inspection, though; only some vehicles are run through it.
The biggest weakness of the technology is that it is slow.
Yes, Apple omitting a London Tube station from a map of London is kind of a killer fail. The images show the user about 100m from a Central Line tube station, but Apple is going to make him walk about a mile to another station.
Is turn-by-turn navigation that important in a handheld device? That's more of a feature for a car-mounted device. You shouldn't be looking at a smartphone while driving, anyway.
At the top of Slashdot pages is an icon with a TV screen with rounded corners and a "rabbit ears" antenna.
The "floppy disk" icon for "Save" hasn't made sense in 20 years. Even then, it makes little sense; it ought to mean "copy to removable medium", not "save to local file".
The use of binoculars for "search" never made much sense. For some time, I thought that icon meant "zoom". "Movie camera" icons still appear in some programs, with camera outlines not seen since the 1930s. But they sometimes mean "play" rather than "record". Images of cassette tapes still appear in some audio-related programs.
Abstract icons hold up better. The circular arrow for "refresh" is fine. Symbols from international road signs are well recognized.
Facebook seems to avoid this. The most dated icon on Facebook's main page is a tear-off calendar. Photoshop is at the other extreme, with a collection of icons meaningful only to people who did photographic film darkroom work.
Too many people are trying to emulate Susan Kare's work of 30 years ago. Badly. Kare herself has moved on.
The biggest trend is automatic steering. Self-driving cars are still experimental. Self-driving tractors are mainstream farming. Automatic plow height adjustment is routine. (The bottom of each furrow should be level, so extra water isn't needed to get over the high spots.)
The second biggest trend is integrating sensors and controls. Measuring soil properties and adjusting inputs (fertilizer, seed, sprays) to compensate is a key part of precision farming. The trend is to do this continuously based on real time measurements, rather than based on a few samples taken in advance.
The overall effect is to reduce wasted inputs and thus costs, improving profits. It doesn't increase yield per acre much, but it may make it feasible to farm more acres.
There's this strange illusion in the US that tablets are supposed to cost a few hundred dollars. Go look at tablets on Alibaba. ("Tablet pc: 474,433 products found.") 7-inch devices are mostly in the $50-$100 range. Over $100, you get a 10-inch screen. Most of them use an Allwinner $7 part, which has most of the electronics other than the display.
Look on Amazon. There's a good selection of tablets in the $60 range. Around $70, you start to get all the bells and whistles. And that's retail.
If Microsoft wants to do something useful in this area, they should build an OS that can respond to a "power is failing" signal and reliably get to a safe state in a few hundred milliseconds. Then put enough capacitance in the power supply to provide a second or two of shutdown warning. With that, transient power failures are no problem, nor are short delays during transfer from line power to backup power.
I don't believe the US military ever actually contributed funding to the development of the shuttle.
The USAF built their own Shuttle launch facility at Vandenberg. The first USAF launch was planned for October 1986. After the Challenger disaster, it was clear that the Shuttle wasn't reliable enough for USAF needs. So Vandenberg converted that pad for Titan and Delta boosters, which they continue to launch once in a while.
Just because Apple can't make a rugged phone doesn't mean others can't. There are many good rugged smartphones now. Nokia, Motorola, HTC, Samsung, Sonim, Casio, and even Caterpillar now have rugged smartphones.
The problem remains the C language. C (and C++) is the only remaining major language prone to buffer overflows.
This can be fixed. See "Safe arrays and pointers for C through compatible additions to the language". This is a proposal for a "strict mode" for C which prevents buffer overflows. It's been discussed on Lambda the Ultimate, the C standard newsgroup, and the GCC development list, and with each round of criticisms, the design is tightened up.
This proposal includes a "strict mode", in which the rules are tighter, and ways to talk about the size of arrays. Non-strict code can call strict code, and vice versa. So there's a gentle migration path to all-strict programs, one source file at a time. It's an extension to C, not a new language. Some of the necessary features for this are already in C99 or are GCC extensions, so I'm trying to get this into GCC as an extension so it can be tried in the real world.
It's no longer acceptable to say that this problem can't be solved. It can. It just takes the will to solve it. Prodding from the security community will help.
Strict code is mostly about declarations. For example, the Linux "read" function, which is now declared
int read(int fd, void* buf, size_t len);
would be declared
int read(size_t len; int fd, void_space (buf&)[len], size_t len);
Instead of passing a pointer, you pass a reference to an array, a reference with an associated size. So the language knows how big the array is. Incidentally, the first "size_t len;" is a forward declaration of len. That's an existing but rarely used GCC extension. It's needed because so many C, UNIX, Linux, and POSIX APIs have the buffer param before the buffer size.
(For those few of you who know what a C99 variable length array parameter is, you'll wonder why this syntax differs from that. It's a long story. C99 VLA params are demoted to pointers at function entrance, losing the size info. It turns out nobody uses C99 VLA params; repeated searches have failed to find any of them in open source code. Also, Microsoft refused to implement them in Visual C/C++, they're incompatible with C++, and they've been demoted to an optional feature in the latest C standard draft.)
This is the usual bullshit about how NASA advanced semiconductor and computer technology. About the only real advance to come from NASA was NASTRAN, the first finite-element analysis program. The paper talks about "space and defense". It was DoD, especially the USAF, that pushed semiconductor and computer technology hard. SAGE, the Atlas Missile Guidance Computer, the Navy's nuclear submarine program, and the various huge missile and radar programs of the 1950s and 1960s all advanced computer and electronics technology.
NASA was a consumer of those technologies, and in terms of units purchased, not a big one. NASA bought a few tens of rockets a year; at the peak, missile programs bought hundreds to thousands.
NASA was big on materials and weight reduction, and some interesting materials came out of NASA. But more of them came out of the USAF. At the time, much of that was classified. The SR-71 was a titanium aircraft flown in the 1960s. Lockheed's Skunk Works actually pioneered the use of liquid hydrogen as a propellant, although NASA took the credit. Heat shield materials came from missile nose cones.
NASA was #1 at public relations, and still has a huge PR operation. DoD and the USAF were trying to keep the USSR from finding out what we had. So NASA got to take the credit for a lot of stuff they didn't pioneer.
After all, Alan Shepard went into space atop a Redstone ICBM booster. John Glenn went into space atop an Atlas ICBM booster. The Gemini program used modified Titan II ICBM boosters. Only Apollo had its own booster.
Slashdot Mirror
who needs NASA?
Really. Space-X has docked a test capsule to the ISS, and their first cargo delivery launches October 7th. Astronauts will follow on later flights.
NASA hasn't developed a successful new booster in 30 years, despite about three failed attempts. Space-X has a new booster that works. Arianespace has a new booster that works. NASA has old ICBM derivatives and a "Space Launch System" proposal.
Space-X has only 1800 employees. NASA still has 34,000. NASA does some good stuff, but it's far too big for what it accomplishes.
This has been talked up for a decade or two, but needs cost and capacity numbers.
There's also the painful fact that every reactor design that had anything mechanically non-trivial inside the reactor has been a flop. There have been two German pebble-bed reactors, both of which had pebble jams serious enough to cause major accidents with significant radiation leaks. Tsinghua University in China has one that's worked for a while, and that design is being scaled up. The Rongcheng Shidaowan Nuclear Power Plant, with two pebble-bed reactors, is under construction now. Completion in 2015. Maybe they can make it work. We won't really know until there are a few hundred reactor-years on that technology.
High temperature, gas-cooled reactors have been tried, but were troublesome. The only big one was Fort. St. Vrain, which had a lot of troubles with auxiliary equipment and corrosion. It only ran 10 years. No big safety issues, though; just high maintenance costs.
This doesn't look like Cornavin station:
You're right. The original article with that image said Cornavin, but I've been to Cornavin station and it doesn't look like that.
At the high end, there's 3D real time ultrasound. This builds up a 3D image from multiple scans from different directions. The ultrasound part of this is no more complex than the basic machines. The 3D part is knowing where the sensor is and software to build up a 3D image. 3D images are much easier to interpret than simple reflection images, and you can rotate them and look from another angle.
So what's needed is a handheld device with a position tracking system. You could probably kludge something together with a 6DOF INS (an accelerometer and rate gyro chip) and a mouse sensor to track movement along the body. The rest is software.
Here's a Swiss railroad clock in its native habitat, at Cornavin station. There are clocks at regular intervals along platforms, and the second hands are, of course, in sync. It's part of the Swiss Railways branding - their stations tend to have a large, if not excessive, number of those clocks.
It's a famous design. A home-size version is available from the Museum of Modern Art in New York. (It does not, however, sync to an external time source.)
I had to opt out of Pacific Gas and Electric's online billing system and go back to paper bills when they changed their EULA to allow more "disclosure". If I just buy power from them, they're subject to regulatory rules, enforced by the California Public Utilities Commission. But they wanted me to sign up for an "online account", which isn't regulated. If you don't sign up, they're not allowed to redistribute your "smart meter" info. If you sign up, you've consented to distribution to "affiliates".
I strongly recommend opting out to PG&E customers (California, Nevada) who are concerned about privacy issues.
The last Microsoft Office product I bought was Word 97. I've been using OpenOffice, then LIbreOffice, since about 2002. It's a OK word processor, a mediocre but adequate spreadsheet, and a better draw program than Office. What's in Microsoft Office that a home user would need, let alone pay for monthly?
Yes, one new mall opened in 2006. That's it since then. The huge Xanadu mall in New Jersey that was supposed to open in 2009 didn't open. Two bankruptcies and a roof collapse later, opening is now scheduled for 2014. Maybe.
Amazon is taking over more and more of retail. They've already taken over books and DVDs. They took over and re-started Webvan. They're building local distribution centers for same day delivery. They bought Kiva Robotics so those warehouses won't need many people.
The glory years of store-based retail are over. No new enclosed mall has been built in the US in a decade. Most retail areas have vacant stores that will never be used again. In a few years, retail will consist of recreational shopping and convenience stores. Everything else will be on line.
Then all we have to worry about is downtime. Read E. M. Forster's "The Machine Stops".
I had a Well account back when it was on a VAX/750 and you paid by the hour. But it's been a long time since the Well mattered.
They had a reactor trip. Big deal. It didn't even show up as a power grid event that required emergency action in the PJM dashboard.
Netflix and Amazon don't need tracking of casual browsers, because they have real customers. They have, legitimately, information about what you knowingly bought from them. Businesses that have real sites that sell real stuff don't really need to track browsers, just customers. Even Facebook doesn't need tracking of casual browsers, since, while they're intrusive, you clearly sign up with and log into Facebook. Google doesn't really need personalization; they were profitable just putting up ads that were relevant to the current search.
So, really, it's the junk sites that need this. Those with Google AdSense junk ads. Most entertainment sites. Slashdot. Crap like that. Getting rid of tracking would hurt them. We might lose some of them. No big loss.
"Wikimedia UK would like to announce that Roger Bamkin has stepped down as a Trustee."
He's right. The issue is not what Facebook shows. It's how the pages are put together. It takes work by a lot of servers to assemble each page. The user-facing servers send out queries to servers which check the feeds of everything being followed - friends, events, calendars, messages, applications - and create a page to display. This page is updated automatically if you keep it open. You can look at any of these items in more detail, and go back into their past if desired.
That's what managers do - follow many changing items superficially and look at some of them in detail. A management version might have feeds for shipments which missed their ship date, incoming orders, customer complaints, personnel absences, due dates for major supplier shipments, and other items of interest. Different users would be watching different things, some info would be available only to some users, and users would set what they wanted to see. If you've ever used a Bloomberg terminal, it's a lot like that, but with worse graphics.
Facebook has a reasonable platform for that sort of thing. The back end is databases and message passing. The business logic and formatting is mostly in PHP (for which Facebook has a hard-code compiler, so it doesn't take forever). Facebook also has decent solutions to the "tell me if it changed without polling too often" problem.
The AS&E Z-backscatter systems are quite good. There's some X-ray exposure, and they're very expensive, but they're effective. They image by atomic number, and can distinguish elements. It's not just density any more, like transmission X-rays. This technology is much more effective (and intrusive) than microwave-based systems.
The machines need a huge number of X-ray detectors, are physically large, very expensive, and aren't that fast. The latest generation of machines are about 2/3 the size of the previous ones (which were so big that they wouldn't fit through standard doors) and are faster. But they're still big.
If the TSA is going to throw money at something, this is at least something that works.
The San Ysidro border US border checkpoint has several lanes of the big version for cars. This thing is the size of a drive-through carwash. It's too slow to use for primary inspection, though; only some vehicles are run through it. The biggest weakness of the technology is that it is slow.
Yes, Apple omitting a London Tube station from a map of London is kind of a killer fail. The images show the user about 100m from a Central Line tube station, but Apple is going to make him walk about a mile to another station.
Is turn-by-turn navigation that important in a handheld device? That's more of a feature for a car-mounted device. You shouldn't be looking at a smartphone while driving, anyway.
At the top of Slashdot pages is an icon with a TV screen with rounded corners and a "rabbit ears" antenna.
The "floppy disk" icon for "Save" hasn't made sense in 20 years. Even then, it makes little sense; it ought to mean "copy to removable medium", not "save to local file".
The use of binoculars for "search" never made much sense. For some time, I thought that icon meant "zoom". "Movie camera" icons still appear in some programs, with camera outlines not seen since the 1930s. But they sometimes mean "play" rather than "record". Images of cassette tapes still appear in some audio-related programs.
Abstract icons hold up better. The circular arrow for "refresh" is fine. Symbols from international road signs are well recognized.
Facebook seems to avoid this. The most dated icon on Facebook's main page is a tear-off calendar. Photoshop is at the other extreme, with a collection of icons meaningful only to people who did photographic film darkroom work.
Too many people are trying to emulate Susan Kare's work of 30 years ago. Badly. Kare herself has moved on.
This is a pulsed fusion system. That's technically interesting, but it's a lab apparatus, not a basis for a power plant.
This is just on Slashdot because it's some minor iPad app. This isn't where the action is. Here are the the top 5 tech trends in precision agriculture.
The biggest trend is automatic steering. Self-driving cars are still experimental. Self-driving tractors are mainstream farming. Automatic plow height adjustment is routine. (The bottom of each furrow should be level, so extra water isn't needed to get over the high spots.)
The second biggest trend is integrating sensors and controls. Measuring soil properties and adjusting inputs (fertilizer, seed, sprays) to compensate is a key part of precision farming. The trend is to do this continuously based on real time measurements, rather than based on a few samples taken in advance.
The overall effect is to reduce wasted inputs and thus costs, improving profits. It doesn't increase yield per acre much, but it may make it feasible to farm more acres.
There's this strange illusion in the US that tablets are supposed to cost a few hundred dollars. Go look at tablets on Alibaba. ("Tablet pc: 474,433 products found.") 7-inch devices are mostly in the $50-$100 range. Over $100, you get a 10-inch screen. Most of them use an Allwinner $7 part, which has most of the electronics other than the display.
Look on Amazon. There's a good selection of tablets in the $60 range. Around $70, you start to get all the bells and whistles. And that's retail.
If Microsoft wants to do something useful in this area, they should build an OS that can respond to a "power is failing" signal and reliably get to a safe state in a few hundred milliseconds. Then put enough capacitance in the power supply to provide a second or two of shutdown warning. With that, transient power failures are no problem, nor are short delays during transfer from line power to backup power.
Tandem had that 20 years ago.
I don't believe the US military ever actually contributed funding to the development of the shuttle.
The USAF built their own Shuttle launch facility at Vandenberg. The first USAF launch was planned for October 1986. After the Challenger disaster, it was clear that the Shuttle wasn't reliable enough for USAF needs. So Vandenberg converted that pad for Titan and Delta boosters, which they continue to launch once in a while.
Just because Apple can't make a rugged phone doesn't mean others can't. There are many good rugged smartphones now. Nokia, Motorola, HTC, Samsung, Sonim, Casio, and even Caterpillar now have rugged smartphones.
Here's a Samsung phone being used to crack walnuts. Here's an HTC phone being used as a hammer. And here's an IPhone 4 broken by a 1-foot drop.
The problem remains the C language. C (and C++) is the only remaining major language prone to buffer overflows.
This can be fixed. See "Safe arrays and pointers for C through compatible additions to the language". This is a proposal for a "strict mode" for C which prevents buffer overflows. It's been discussed on Lambda the Ultimate, the C standard newsgroup, and the GCC development list, and with each round of criticisms, the design is tightened up.
This proposal includes a "strict mode", in which the rules are tighter, and ways to talk about the size of arrays. Non-strict code can call strict code, and vice versa. So there's a gentle migration path to all-strict programs, one source file at a time. It's an extension to C, not a new language. Some of the necessary features for this are already in C99 or are GCC extensions, so I'm trying to get this into GCC as an extension so it can be tried in the real world.
It's no longer acceptable to say that this problem can't be solved. It can. It just takes the will to solve it. Prodding from the security community will help.
Strict code is mostly about declarations. For example, the Linux "read" function, which is now declared int read(int fd, void* buf, size_t len); would be declared int read(size_t len; int fd, void_space (buf&)[len], size_t len); Instead of passing a pointer, you pass a reference to an array, a reference with an associated size. So the language knows how big the array is. Incidentally, the first "size_t len;" is a forward declaration of len. That's an existing but rarely used GCC extension. It's needed because so many C, UNIX, Linux, and POSIX APIs have the buffer param before the buffer size.
(For those few of you who know what a C99 variable length array parameter is, you'll wonder why this syntax differs from that. It's a long story. C99 VLA params are demoted to pointers at function entrance, losing the size info. It turns out nobody uses C99 VLA params; repeated searches have failed to find any of them in open source code. Also, Microsoft refused to implement them in Visual C/C++, they're incompatible with C++, and they've been demoted to an optional feature in the latest C standard draft.)
This is the usual bullshit about how NASA advanced semiconductor and computer technology. About the only real advance to come from NASA was NASTRAN, the first finite-element analysis program. The paper talks about "space and defense". It was DoD, especially the USAF, that pushed semiconductor and computer technology hard. SAGE, the Atlas Missile Guidance Computer, the Navy's nuclear submarine program, and the various huge missile and radar programs of the 1950s and 1960s all advanced computer and electronics technology.
NASA was a consumer of those technologies, and in terms of units purchased, not a big one. NASA bought a few tens of rockets a year; at the peak, missile programs bought hundreds to thousands.
NASA was big on materials and weight reduction, and some interesting materials came out of NASA. But more of them came out of the USAF. At the time, much of that was classified. The SR-71 was a titanium aircraft flown in the 1960s. Lockheed's Skunk Works actually pioneered the use of liquid hydrogen as a propellant, although NASA took the credit. Heat shield materials came from missile nose cones.
NASA was #1 at public relations, and still has a huge PR operation. DoD and the USAF were trying to keep the USSR from finding out what we had. So NASA got to take the credit for a lot of stuff they didn't pioneer.
After all, Alan Shepard went into space atop a Redstone ICBM booster. John Glenn went into space atop an Atlas ICBM booster. The Gemini program used modified Titan II ICBM boosters. Only Apollo had its own booster.