Order your manned flight to the ISS
here.
Space-X is doing various "abort tests" NASA insists on, and the first manned launch for NASA isn't supposed to happen until 2017. But Space-X may send their own private astronauts into orbit next year.
Space-X has been sending Dragon spacecraft to the ISS for some time now. The fourth one was supposed to launch this week, but the USAF had a fire at one of their tracking stations and all Cape Canaveral launches are on hold.
After reading all three articles, I'm glad the developers walked out. Now they need to stop apologizing about it. They were recruited by misrepresentation, and when they found out, they didn't like it. They have nothing to apologize for. They don't need to justify their actions. That the sponsor lost $500K is not their problem.
Current cheap chinese-made LED lights are amazingly efficient and put out a very good light, but suffer from capacitor plague.
A solution for that has been developed at Southwest Jiaotong University, Chengdu, Sichuan. Here's the theory paper: "Bridgeless SEPIC-derived LED Driver without Electrolytic
Capacitor for Multistring Application". And here's the product announcement."Dali Power recommends led drive design without electrolytic capacitor, used mainly for bulb lamp. Life is 10 times of the original LED drivers, the design life is as much as 40000 hours or more, can be well matched with the life LED lamp, small size, only forty percent of the original LED driver area. Products are mainly used in household low power lighting, suitable scope is 3W ~ 20W." So this problem is being solved. The new approach is both cheaper and has a longer lifetime.
If those overpriced "smart" bulbs still have electrolytics inside, they're already obsolete.
When you shut off a lamp manually, Hue may not know what state the light is in. Turn it off with the Tap, and it knows the lights are off.
They've replicated the one-way communication of X10, then. That seems rather lame.
Meanwhile, Cree's nice LED replacements for 60W incandescent bulbs are now below $10 at Home Depot. 10 year warranty. They draw 9 watts. Dimmable with existing external dimmers. Just buy a case of those and replace anything that burns out with one.
I'm sure I'll be crucified on a cross of gold for saying this, but the silver mine owners of Nevada were the Koch brothers of that time...
Somebody knows a little financial history. Early thinking about inflation involved "the free and unlimited coinage of silver". Look up the "free silver" movement.
Understanding the implications and biases of the financial system is very important. For example, introducing money into the system via the Fed is a a bias toward the banking system. Introducing it through infrastructure spending (which Japan does) is a bias toward the construction industry. Introducing it by financing export oriented industries (as China does) is a bias toward manufacturing industries.This is a political policy choice, and more than one option is possible.
If you have to kill the engine for some emergency, that shouldn't disable the air bag system. Perhaps the air bag system should be powered whenever either the ignition is on or the vehicle is out of Park.
Now that ex-employees are talking to the press and the cops, we'll find out what was going on.
The Reuters article makes it clear that Karpeles had exclusive personal control over Mt. Gox's cash. That probably means he'll be the one going to jail. I've been writing for months (ever since Mt. Gox suspended US dollar withdrawals last summer) that Mt. Gox was either incompetent, broke, or crooked. Now it looks like all of the above.
Why would Karpeles import a Honda Accord R from the UK to Japan? They're made in Japan.
I'm in Silicon Valley, and cellular just doesn't work very well. At least not Sprint's CDMA network.
At home, I have to go to a window to get one or two bars, because the local community association doesn't want a cell tower nearby. I have a Sprint Airave box, which gives me a femtocell which mooches bandwidth from my IP connection. This gets me VoIP quality at cellular prices. If I lose Internet connectivity, I lose cellular connectivity. The Airave box is badly programmed; when it loses IP connectivity it still captures local handsets and insists it's the best path to the network. You have to disconnect its power to reach a cell tower instead.
At TechShop Menlo Park, which is adjacent to a major freeway, I have to get near a window to get coverage. I'm not sure why there's a coverage hole there.
For a long time, there was no Sprint coverage on the Stanford campus, because Stanford had an exclusive deal with AT&T.
I was in San Jose recently, near PayPal HQ, and couldn't get
Sprint connectivity until I drove up to a closed Sprint store. They have a femtocell so their demos work, and just outside the store, there was good connectivity.
Even when it works, cellular voice quality sucks. Sprint finally seems to have fixed their delay problem, though. For a while I was getting delays as long as a second, with delayed echoes coming back, like some low-end VoIP system.
The land line works great. Voice quality is very good, because it's only about 150 feet of copper to the big underground AT&T vault (the size of a shipping container, air conditioned, and full of racks of gear) out at the street. But there are no cellular antennas at that location; it's all wires and fiber.
How to steal car:
1. Guess username and password.
2. Log in to "https://portal.vn.teslamotors.com".
3. Send GET to "https://portal.vn.teslamotors.com/vehicles" to get list of vehicle IDs for that owner.
4. Send GET to "https://portal.vn.teslamotors.com/vehicles/{id}/command/drive_state" to get vehicle latitude and longitude.
5. Send GET to "https://portal.vn.teslamotors/vehicles//vehicles/{id}/command/door_unlock" to unlock doors.
6. Get in car and plug laptop into onboard Ethernet, where car internals are exposed, unencrypted. ...
And those guys think they're going to do automatic driving. Right.
Remember how photo printers put photo shops out of business?
Well, yes. I haven't seen any photo shops lately. "1 Hour Photo" is dead. Kinkos has photo printers, and so do the local CVS and Walgreens, but they're not used much. Nobody has an in-store film processor any more. Palo Alto still has Keeble and Shugat, a high end photo equipment store with pro darkroom services. Redwood City has some wedding-photographer types and some commercial printers. That's about it.
It's amazing that one company has been able to obtain 80% market share in eyeglass frames in the US. It's not like they're hard to make. Frames start at $0.60 on Alibaba.
For really cheap glasses, you make them round. Ordinary lenses have three parameters - spherical radius, cylindrical radius, and cylinder axis. For round lenses, only the first two matter; the third is determined when the lens goes into the frame. So there's a briefcase-sized kit used in India with a set of standard round lenses moulded from polycabonate, standard round frames, an adjustable temporary frame for the eye exam, an eye chart, and a little gadget to notch the lenses to keep them from rotating once the desired cylinder axis is determined.
The original author is looking at the part of the problem that gets the most attention, but not at the part that causes the most problems. He's looking at programming languages and their expressive problems. Those are real, but they're not why large programs break. Large programs usually break for non-local reasons. Typically, Part A was connected to Part B in some way such that an assumption made by part B was not satisfied by part A.
C is particularly bad in this regard. C programs revolve round three issues: "how big is it", "who owns it", and "who locks it". The language helps with none of these issues, which is why we've been having buffer overflows and low-level security holes for three decades now. Most newer languages (not including C++) address the first two, although few address the third well.
There have been attempts to get hold of this problem formally. "Design by contract" tries to do this. This allows talking about things like "Before calling F, you must call G to do setup.", or "parameter A must be greater than parameter B". Design by contract is a good idea but weighed down by so much formalism that few programmers use it. It's about blame - if a library you call fails, but the caller satisfied the contract in the call, it's the library's fault. Vendors hate that being made so explicit.
it's a concept from aerospace - if A won't connect properly to B, the one which doesn't match the spec is wrong. If you can't decide who's wrong, the spec is wrong. This is why you can take a Pratt and Whitney engine off an airliner, put a Rolls Royce engine on, and have it work. Many aircraft components are interchangeable like that. It takes a lot of paperwork, inspections, and design reviews to make that work.
The price we pay for not having this is a ritual-taboo approach to libraries. We have examples now, not reference manuals. If you do something that isn't in an example, and it doesn't work, it's your fault. This is the cause of much trouble with programming in the large.
Database implementers get this, but almost nobody else in commercial programming does. (Not even the OS people, which is annoying.)
The trouble with VR is that it's hard to do anything in there except move and shoot. Manipulation sucks without force feedback. When VR was first developed, there was a lot of interest in it for CAD. But it didn't help.
Trying to assemble parts in VR is no better than doing it with a mouse and screen. Legos might work. VR Minecraft is quite possible, because things snap into place in easily implemented ways. Real world parts don't fit together as simply.
(Although, thinking about this, it might be possible. You'd need a good game physics engine with very good support for actual geometry. Not just bounding boxes, but good enough to handle bolt-in-hole placement. This is a solved problem in collision detection (I used to work on that), but most games don't bother. Then you'd need to assist the user with something like the "Assemble" feature in Autodesk Inventor. But it would need to be more physical and less symbolic than the way Inventor does it, where assembly means aligning two facing features. Inventor does it that way because you may want to align the features first, then put a hole and bolt through both, then project the hole back onto the original parts so you know where to drill the hole in each part. This is what you need for mechanical design, but not for assembling existing parts.)
This isn't a new idea. It's been tried several times since 1950 for city buses, which are constantly stopping and starting. In 2009, one was developed for use in London. In the 2009 model, the linkage to the flywheel is mechanical, through a continously variable transmission, not electrical. Although this has been in test for several years now, it's only one bus.
That's the same technology Volvo is using. Putting this in a car seems marginal. It makes more sense for buses and delivery vans.
For better or worse, security paranoia after 9/11 has funded substantial computer crime analysis capabilities in the US.
Japan's JPCERT is a small industry-funded nonprofit. US CERT was a small nonprofit before 9/11. It's now part of Homeland Security's empire. The Secret Service and the FBI also have big computer crime units.
All serious Bitcoin mining is now industrial-scale using custom ASICs. CPU-based and GPU-based mining are dead. They can't even cover their own power bill. This guy's setup is primitive compared to this large high-density liquid-cooled mining facility in Hong Kong. The two biggest mining pools control over half of the mining power, and the biggest, "ghash.io", would have over half if they hadn't deliberately split up to avoid that happening.
The thing to remember about Bitcoin mining is that all miners are in competition for a fixed number of Bitcoins produced each week. More mining does not mean more Bitcoins are generated.
"Nurture and grow a civilisation of reactive, living followers who worship you as a god." - product promo for his current game. Talk about an ego trip...
It's an always-on MMORPG, so managing your piece of the world may be a full time job. The graphics suck, apparently by intent. It makes Animal Crossing look realistic.
All that blithering. The business isn't about DVD vs streaming. It's about metered flat rate (Netflix's 2 DVDs at a time) vs pay per item (Amazon, Google streaming).
Pricing models in the movie industry are an interesting subject, but the original poster clearly knows nothing about them. This has been discussed to death in the trades (The Hollywood Reporter and Variety). Hollywood is desperately trying to avoid commodization of movies, something that's already happened to music.
Facebook is way overpaying. All those guys have is yet another piece of 3D headgear. They're not the first, they're not the best, they're not shipping production product, and what they have is rather clunky. It's basically off the shelf displays with the optics to focus them at infinity.
Another one of those "nanomaterial" stories. Claims of "cheaper manufacturing costs" for a product not yet produced in volume. Yet another "solar cell made with printing technology" scheme. Sigh.
So this thing is one micrometer thick and they want to print it on windows. How long does it last, hammered by UV and weather and thermal cycling? Lifetimes for silicon solar panels are now up to 25 years (warranties are available for that long), and there's falloff in output over that time. Can a 1 micrometer film match that? Realistically, it's going to have to be behind a protective layer. Maybe it could be on the surface of the middle layer in double-pane glass, but now you have a complex sandwich to manufacture. There goes the "cheap manufacturing".
Vertical windows are poorly oriented for capturing power. Solar shingles are better oriented. (Also, they exist, and you can buy them now.)
And, as the head of Applied Material Solar pointed out a decade ago, half the cost is installing the thing. These guys need better numbers to back up their cost claims.
Slashdot Mirror
If I want to see nerds talk, there's YouTube.
Order your manned flight to the ISS here. Space-X is doing various "abort tests" NASA insists on, and the first manned launch for NASA isn't supposed to happen until 2017. But Space-X may send their own private astronauts into orbit next year.
Space-X has been sending Dragon spacecraft to the ISS for some time now. The fourth one was supposed to launch this week, but the USAF had a fire at one of their tracking stations and all Cape Canaveral launches are on hold.
After reading all three articles, I'm glad the developers walked out. Now they need to stop apologizing about it. They were recruited by misrepresentation, and when they found out, they didn't like it. They have nothing to apologize for. They don't need to justify their actions. That the sponsor lost $500K is not their problem.
Current cheap chinese-made LED lights are amazingly efficient and put out a very good light, but suffer from capacitor plague.
A solution for that has been developed at Southwest Jiaotong University, Chengdu, Sichuan. Here's the theory paper: "Bridgeless SEPIC-derived LED Driver without Electrolytic Capacitor for Multistring Application". And here's the product announcement. "Dali Power recommends led drive design without electrolytic capacitor, used mainly for bulb lamp. Life is 10 times of the original LED drivers, the design life is as much as 40000 hours or more, can be well matched with the life LED lamp, small size, only forty percent of the original LED driver area. Products are mainly used in household low power lighting, suitable scope is 3W ~ 20W." So this problem is being solved. The new approach is both cheaper and has a longer lifetime.
If those overpriced "smart" bulbs still have electrolytics inside, they're already obsolete.
At this point I'm mostly replacing CFLs.
When you shut off a lamp manually, Hue may not know what state the light is in. Turn it off with the Tap, and it knows the lights are off.
They've replicated the one-way communication of X10, then. That seems rather lame.
Meanwhile, Cree's nice LED replacements for 60W incandescent bulbs are now below $10 at Home Depot. 10 year warranty. They draw 9 watts. Dimmable with existing external dimmers. Just buy a case of those and replace anything that burns out with one.
I'm sure I'll be crucified on a cross of gold for saying this, but the silver mine owners of Nevada were the Koch brothers of that time...
Somebody knows a little financial history. Early thinking about inflation involved "the free and unlimited coinage of silver". Look up the "free silver" movement.
Understanding the implications and biases of the financial system is very important. For example, introducing money into the system via the Fed is a a bias toward the banking system. Introducing it through infrastructure spending (which Japan does) is a bias toward the construction industry. Introducing it by financing export oriented industries (as China does) is a bias toward manufacturing industries.This is a political policy choice, and more than one option is possible.
there are a number of exchanges that pay interest on your holdings on the exchange.
That just screams "Ponzi scheme". Where's the money coming from?
If you have to kill the engine for some emergency, that shouldn't disable the air bag system. Perhaps the air bag system should be powered whenever either the ignition is on or the vehicle is out of Park.
Now that ex-employees are talking to the press and the cops, we'll find out what was going on.
The Reuters article makes it clear that Karpeles had exclusive personal control over Mt. Gox's cash. That probably means he'll be the one going to jail. I've been writing for months (ever since Mt. Gox suspended US dollar withdrawals last summer) that Mt. Gox was either incompetent, broke, or crooked. Now it looks like all of the above.
Why would Karpeles import a Honda Accord R from the UK to Japan? They're made in Japan.
I'm in Silicon Valley, and cellular just doesn't work very well. At least not Sprint's CDMA network.
At home, I have to go to a window to get one or two bars, because the local community association doesn't want a cell tower nearby. I have a Sprint Airave box, which gives me a femtocell which mooches bandwidth from my IP connection. This gets me VoIP quality at cellular prices. If I lose Internet connectivity, I lose cellular connectivity. The Airave box is badly programmed; when it loses IP connectivity it still captures local handsets and insists it's the best path to the network. You have to disconnect its power to reach a cell tower instead.
At TechShop Menlo Park, which is adjacent to a major freeway, I have to get near a window to get coverage. I'm not sure why there's a coverage hole there.
For a long time, there was no Sprint coverage on the Stanford campus, because Stanford had an exclusive deal with AT&T.
I was in San Jose recently, near PayPal HQ, and couldn't get Sprint connectivity until I drove up to a closed Sprint store. They have a femtocell so their demos work, and just outside the store, there was good connectivity.
Even when it works, cellular voice quality sucks. Sprint finally seems to have fixed their delay problem, though. For a while I was getting delays as long as a second, with delayed echoes coming back, like some low-end VoIP system.
The land line works great. Voice quality is very good, because it's only about 150 feet of copper to the big underground AT&T vault (the size of a shipping container, air conditioned, and full of racks of gear) out at the street. But there are no cellular antennas at that location; it's all wires and fiber.
How to steal car:
...
1. Guess username and password.
2. Log in to "https://portal.vn.teslamotors.com".
3. Send GET to "https://portal.vn.teslamotors.com/vehicles" to get list of vehicle IDs for that owner.
4. Send GET to "https://portal.vn.teslamotors.com/vehicles/{id}/command/drive_state" to get vehicle latitude and longitude.
5. Send GET to "https://portal.vn.teslamotors/vehicles//vehicles/{id}/command/door_unlock" to unlock doors.
6. Get in car and plug laptop into onboard Ethernet, where car internals are exposed, unencrypted.
And those guys think they're going to do automatic driving. Right.
Remember how photo printers put photo shops out of business?
Well, yes. I haven't seen any photo shops lately. "1 Hour Photo" is dead. Kinkos has photo printers, and so do the local CVS and Walgreens, but they're not used much. Nobody has an in-store film processor any more. Palo Alto still has Keeble and Shugat, a high end photo equipment store with pro darkroom services. Redwood City has some wedding-photographer types and some commercial printers. That's about it.
It's amazing that one company has been able to obtain 80% market share in eyeglass frames in the US. It's not like they're hard to make. Frames start at $0.60 on Alibaba.
For really cheap glasses, you make them round. Ordinary lenses have three parameters - spherical radius, cylindrical radius, and cylinder axis. For round lenses, only the first two matter; the third is determined when the lens goes into the frame. So there's a briefcase-sized kit used in India with a set of standard round lenses moulded from polycabonate, standard round frames, an adjustable temporary frame for the eye exam, an eye chart, and a little gadget to notch the lenses to keep them from rotating once the desired cylinder axis is determined.
The original author is looking at the part of the problem that gets the most attention, but not at the part that causes the most problems. He's looking at programming languages and their expressive problems. Those are real, but they're not why large programs break. Large programs usually break for non-local reasons. Typically, Part A was connected to Part B in some way such that an assumption made by part B was not satisfied by part A.
C is particularly bad in this regard. C programs revolve round three issues: "how big is it", "who owns it", and "who locks it". The language helps with none of these issues, which is why we've been having buffer overflows and low-level security holes for three decades now. Most newer languages (not including C++) address the first two, although few address the third well.
There have been attempts to get hold of this problem formally. "Design by contract" tries to do this. This allows talking about things like "Before calling F, you must call G to do setup.", or "parameter A must be greater than parameter B". Design by contract is a good idea but weighed down by so much formalism that few programmers use it. It's about blame - if a library you call fails, but the caller satisfied the contract in the call, it's the library's fault. Vendors hate that being made so explicit.
it's a concept from aerospace - if A won't connect properly to B, the one which doesn't match the spec is wrong. If you can't decide who's wrong, the spec is wrong. This is why you can take a Pratt and Whitney engine off an airliner, put a Rolls Royce engine on, and have it work. Many aircraft components are interchangeable like that. It takes a lot of paperwork, inspections, and design reviews to make that work.
The price we pay for not having this is a ritual-taboo approach to libraries. We have examples now, not reference manuals. If you do something that isn't in an example, and it doesn't work, it's your fault. This is the cause of much trouble with programming in the large.
Database implementers get this, but almost nobody else in commercial programming does. (Not even the OS people, which is annoying.)
The trouble with VR is that it's hard to do anything in there except move and shoot. Manipulation sucks without force feedback. When VR was first developed, there was a lot of interest in it for CAD. But it didn't help.
Trying to assemble parts in VR is no better than doing it with a mouse and screen. Legos might work. VR Minecraft is quite possible, because things snap into place in easily implemented ways. Real world parts don't fit together as simply.
(Although, thinking about this, it might be possible. You'd need a good game physics engine with very good support for actual geometry. Not just bounding boxes, but good enough to handle bolt-in-hole placement. This is a solved problem in collision detection (I used to work on that), but most games don't bother. Then you'd need to assist the user with something like the "Assemble" feature in Autodesk Inventor. But it would need to be more physical and less symbolic than the way Inventor does it, where assembly means aligning two facing features. Inventor does it that way because you may want to align the features first, then put a hole and bolt through both, then project the hole back onto the original parts so you know where to drill the hole in each part. This is what you need for mechanical design, but not for assembling existing parts.)
He's well known if you're into the low-level machinery of game graphics.
This isn't a new idea. It's been tried several times since 1950 for city buses, which are constantly stopping and starting. In 2009, one was developed for use in London. In the 2009 model, the linkage to the flywheel is mechanical, through a continously variable transmission, not electrical. Although this has been in test for several years now, it's only one bus.
That's the same technology Volvo is using. Putting this in a car seems marginal. It makes more sense for buses and delivery vans.
There are serious doubts as to Japanese law enforcement's abilities to deal with the technical issues involved.
Surprisingly, this is correct. The National Police Agency, as of last summer, was just setting up their computer crime unit. It's mostly aimed at infrastructure protection. The Tokyo Metropolitan Police also set up a cybercrime squad in 2013. So they're just getting started on this.
For better or worse, security paranoia after 9/11 has funded substantial computer crime analysis capabilities in the US. Japan's JPCERT is a small industry-funded nonprofit. US CERT was a small nonprofit before 9/11. It's now part of Homeland Security's empire. The Secret Service and the FBI also have big computer crime units.
Check out TheCleanGame[.]Net/scc if you're looking for .5% a day on your btc... :)
Sigh. Another Bitcoin Ponzi.
There seem to be a significant number of Bitcoin users who can't recognize an obvious Ponzi scheme on sight.
All serious Bitcoin mining is now industrial-scale using custom ASICs. CPU-based and GPU-based mining are dead. They can't even cover their own power bill. This guy's setup is primitive compared to this large high-density liquid-cooled mining facility in Hong Kong. The two biggest mining pools control over half of the mining power, and the biggest, "ghash.io", would have over half if they hadn't deliberately split up to avoid that happening.
The thing to remember about Bitcoin mining is that all miners are in competition for a fixed number of Bitcoins produced each week. More mining does not mean more Bitcoins are generated.
"Nurture and grow a civilisation of reactive, living followers who worship you as a god." - product promo for his current game. Talk about an ego trip...
It's an always-on MMORPG, so managing your piece of the world may be a full time job. The graphics suck, apparently by intent. It makes Animal Crossing look realistic.
All that blithering. The business isn't about DVD vs streaming. It's about metered flat rate (Netflix's 2 DVDs at a time) vs pay per item (Amazon, Google streaming).
Pricing models in the movie industry are an interesting subject, but the original poster clearly knows nothing about them. This has been discussed to death in the trades (The Hollywood Reporter and Variety). Hollywood is desperately trying to avoid commodization of movies, something that's already happened to music.
Facebook is way overpaying. All those guys have is yet another piece of 3D headgear. They're not the first, they're not the best, they're not shipping production product, and what they have is rather clunky. It's basically off the shelf displays with the optics to focus them at infinity.
$2 billion for that is a bit much.
Another one of those "nanomaterial" stories. Claims of "cheaper manufacturing costs" for a product not yet produced in volume. Yet another "solar cell made with printing technology" scheme. Sigh.
So this thing is one micrometer thick and they want to print it on windows. How long does it last, hammered by UV and weather and thermal cycling? Lifetimes for silicon solar panels are now up to 25 years (warranties are available for that long), and there's falloff in output over that time. Can a 1 micrometer film match that? Realistically, it's going to have to be behind a protective layer. Maybe it could be on the surface of the middle layer in double-pane glass, but now you have a complex sandwich to manufacture. There goes the "cheap manufacturing".
Vertical windows are poorly oriented for capturing power. Solar shingles are better oriented. (Also, they exist, and you can buy them now.) And, as the head of Applied Material Solar pointed out a decade ago, half the cost is installing the thing. These guys need better numbers to back up their cost claims.