What's needed is a big lawsuit by a big bank against Microsoft for willful negligence. (Def: Intentional performance of an unreasonable act in disregard of a known risk, making it highly probable that harm will be caused.")
Knowingly distributing operating systems which are known to be remotely exploitable to attack other systems fits that definition.
Microsoft's EULA doesn't protect them here. The victim is a third party, not their own customer, and not a signatory to the EULA. Nor does this require a class action. There are single banks big enough to take this on.
These guys are claiming to have controlled thermonuclear fusion above breakeven. That's huge. No one has ever done that. If it works, we have a new major power source.
They write: "Now, the key will be combining each isolated test into a final experiment that produces fusion using this technology". That's a Nobel prize if they succeed.
I read the article and went to edX. At edX I signed up; (semicolon should be a comma) but, (misplaced comma) I can not (should be "cannot") find out about this system. Quite Frankly,(lower case!) I am a teacher and I need my students to be writing more; (Run-on sentence. Use a period.) however, I do not have the time to grade all of their papers (comma here) so I have been assigning more objective homework that I would like.
All banks can only pay out a fraction of deposits at any given time.
Mt. Gox is not a bank. Mt. Gox is a payment services firm under Japan's Payment Services Act of 2009. That law allowed non-bank businesses to do payment services, and, since then, many of the mobile operators in Japan run payment services. But payment service firms in Japan are not allowed to engage in fractional reserve banking.
"The PSA will impose an obligation on an operator to secure
the assets in amounts equal to or more than the total
amount of: (i) funds which an operator is transmitting; and (ii) procedural costs in relation to reimbursement of such funds as set out in (i), so that the transferred funds can reach the recipient even in the event of an operator's insolvency."
So Mt. Gox has to have at least 100% of the deposited funds as hard assets.
Almost all the "online wallet" companies have at some point lost customer money. Instalwallet is just the latest. Bitomat, MyBitcoin, and some others also tanked.
Bitcoin.org now has a warning: "Web wallets host your bitcoins. That means it is possible for them to lose your bitcoins following any incident on their side. As of today, no web wallet service provide enough insurance to be used to store value like a bank."
They're unregulated depositary institutions. Historically, those don't end well. Keeping much money in Bitcoin "exchanges" is iffy, too. Mt. Gox has withdrawal rate limits on Bitcoins, which is suspicious. They should be able to pay out 100% of their Bitcoin balances at any time. If they can't, they're skimming.
Like Erlang, it provides the shared-nothing actor model for concurrent programming at the language level, instead of mucking with threads and global state directly.
That's encouraging. The Go language designers tried to do that, but didn't quite get it right. They talk a good game of "Don't communicate by sharing memory; share memory by communicating". But in reality, Go shares data (usually, too much data) between concurrent tasks. When you pass a reference through a Go channel, you're sharing data. Go is prone to race conditions. You can even break through the memory model and craft exploits by exploiting race conditions. As a result, Google's AppEngine has to lock down Go programs to one thread.
(There are three big memory safety holes in C: "How big is it", "Who can delete it", and "Who locks it". C++ tries to address #1, but doesn't quite succeed. Languages with GC address #2. Few languages (Ada and Erlang being exceptions) address #3 well. Go does not, Rust apparently does.)
We'll have to see how the Rust people do. Stuff that comes out of Mozilla tends to have a huge number of half-implemented features. (That's so open-source.) Go is a coherent design, even if flawed. It's from some of the old Plan9 people at Bell Labs.
Aereo is a legal hack. Each user has their very own UHF antenna. The receiving center has thousands of tiny UHF antennas, one per user, each driving their own private file store. It's a remote DVR.
Remember netbooks? They got to be rather good under-$300 laptop computers before the industry killed them off for not being expensive enough and not requiring any expensive or intrusive "cloud service".
Provide a few key apps, and wearing Google Goggles will be made illegal.
CopWatch Whenever a uniformed police officer or a police car appears, log badge number, faces, location, time, and date. Upload to tracking web site for map overlay. Process face image for face recognition. Match face against other faces seen on any device subscribing to the service. If matching person is in a vehicle, upload license plate info. Add vehicle to tracking list.
BribeWatch Like CopWatch, but for elected officials. Preload system with pictures of elected officials from news media. Also preload with list of all lobbyists registered with Congress (a public record). Record who politicians are seen with. Feed lobbyist location data, contribution data, and vote data into a machine learning algorithm to generate probable cause information for bribery investigations.
Bitcoin itself doesn't need much regulation, but some of the companies profiting from it do. Entities which hold onto your money, from PayPal to Mt. Gox, are depository institutions, and should be regulated as such. Customers need to know who's behind them and that they really have the money. There's a long list of defunct "Bitcoin exchanges" which took the money and ran. That's where regulation is needed.
Customers need strong enforceable rights against exchanges. Mt. Gox is notorious for blaming their own business partners (OKPay, Dwolla, etc.) when something goes wrong. They also limit how fast you can take money out (even in Bitcoins) change those limits from time to time, and the limits are very low by banking standards. Since Mt Gox supposedly has 100% of the deposited Bitcoins, they should be able to pay out 100% of any account on demand.
DARPA has had some of those two-handed robots for a few years now, and has researchers working with them. So far, the results have been disappointing. They can pick up a screwdriver, but not use it. Robots manipulating a powered socket wrench in a well-defined environment aren't a big deal; auto assembly lines have had those for years. Putting a key in a lock, slowly, is about the upper limit of that project.
Part of the problem is that simulators for manipulation aren't very good. Willow Garage is funded to take the current version of the Gazebo simulator and make it work for manipulation. But I doubt they'll be able to make it good enough to do manipulation using force feedback. The physics engines they're using are for game physics; they cheat on frictional contacts. You need a really good simulator to debug control software for putting a nut on a bolt by feel.
They don't seem to be able to accomplish much or cooperate very well. Better work with swarm robots has been done.
One of the better robotics ideas of the 1980s was a pair of small (about 1 cubic foot) robots with small forklifts. They could cooperate to move large objects, like a couch. One was in charge, and one was the helper. Once both robots were in position, force feedback and very limited communication was enough to coordinate them.
That's a useful concept to develop further today. In the 1980s, navigation and vision weren't good enough to make this work reliably. Now they are.
Very nice. Festo, which is a German industrial robotics firm, does a technical tour de force every year. They built a robotic bird two years ago. In 2009, they built a robot penguin" which swims beautifully.
Festo does this to sell their industrial robotics systems, which are very well made.
This collision has one particular effect that does not meet standard thoughts on free markets; competition brings prices UP.
That's very common. In the antiquated "free market" view, competition inevitably drives prices down. In modern marketing, competition is by features, conveniance, marketing, and status symbol value. (Academic journals are in the status symbol category.) Pricing is driven by implicit or explicit collusion, with competitors striving to push prices upwards.
This model applies to appliances, autos, cell phones, music, movie tickets, etc. Some things still have price competition, but they're mostly commodities.
Back in my DARPA Grand Challenge days, I saw fully automated parking as the first "killer app" for automated driving. Everybody was obsessed with automated freeway driving, but that's not what annoys people. Looking for parking annoys people. The general idea is that you get out of your car at your destination, and it goes and parks itself somewhere. When you want your car back, you call it and it comes to you. Parking then need not be as close to the destination; a big parking garage a mile away is fine.
The first application of this should have been for airport rental cars. You rent the car via your phone, and the car comes to the loading area near baggage claim and picks you up. When you're done with the car and at the airport, you get out at the departure area, and it drives itself to rental car return. Customers would save an hour on every plane trip. That would sell.
It's workable. At no time is autonomous operation above about 20MPH necessary, which means slamming on the brakes is sufficient to deal with most problems. All the rental cars are new and under common ownership and maintenance, so the self-driving systems can be checked out on every rental. The system could be expanded to include the top 10 destinations for rental cars - major hotels, convention centers, etc.
After 9/11, no way would autonomous vehicles be allowed in an airport terminal area. So that didn't look promising back in the mid-2000s. Today, though, with terrorism down to nuisance levels, it's worth looking at again.
As for VW thinking that automated driving is more than a decade away, both Ford and Mercedes have said they expect to have it in production vehicles in five years.
This could be interesting. Servers are still designed like PCs. That's not fundamental. One could have compute servers which have network hardware that is configured during the boot process and which restricts what the machine can talk to. Their storage is all external, on file servers elsewhere. They have no peripherals other than the network. They barely need an operating system - the remote management hardware in the network interface handles administration.
With Linux at 15,000,000 lines of code, there's a bloat problem. There's still a need for a run-time library, but it will be more like the C run time system than an OS.
Shouldn't we have some book that explains how to quickly bootstrap ourselves back to current levels of civilization?
Such information was collected in the 1950s, stored on microfiche, and copies provided to major US fallout shelters, along with a reader that only needed sunlight. I have no idea where to get that now.
There's a classic set of books, Build Your Own Metalworking Shop From Scrap, by David Gingery, just for this purpose. The first book, Build Your Own Charcoal Foundry, starts the process. The books do assume there's plenty of metal scrap around; you're not expected to start from ore.
Getting up to pre-WWII levels from very little can be done. Much of Europe and Japan had to do it after WWII. Although many of the skills have been lost. How many people today can operate a lathe?
Where's the announcement from China? Canonical has a long history of bullshit announcements that some big vendor is going to use their product. They've made that claim in the past for both Asus and Dell. In both cases, the Canonical product never appeared on those platforms, or was a very minor niche announcement.
Here's a recent policy announcement on open source from CSIP. They encourage using Linux, but Canonical is not mentioned. The action agency on this is the "China Innovative Strategic Alliance for Open Source", but I'm not seeing them associated with Canonical.
A book written by a committee will be a painful read.
If you want to do something useful, create one of those single plastic sheet two page guides to the language. Boiling the language down to two pages of small type with a few diagrams is a useful exercise. More useful than another thousand page book of blithering.
There's a tradeoff between lifespan and maintenance requirements. For fun, I restore old Teletype machines from the 1920s and 1930s. I have four of them running.
A normal maintenance operation on early Teletypes is to remove the two electrical parts (the motor and the selector magnet) and soak the entire machine in cleaning solution to get rid of dust and dead oil. For machines in heavy use, Western Union did that annually. Then they had to be oiled again (there are several hundred oiling points and six pages of lubrication instructions), gears and sliding joints greased, the electrical parts re-installed, adjustment procedures performed, and the machines re-tested.
Because of this design for maintainability, I've been able to take 80 year old machines that were covered with rust and dirt, and restore them to full operation. But who would put up with something today which required that kind of maintenance?
Getting people to clean or change the filters on their desktop computers is difficult.
The big problem is that almost everybody in US "mobile", telephony, and cable already has a "walled garden". From Comcast to Apple, everybody in that space has a tight grip on their users. Most "apps" are really just a form of DRM.
This is a US thing, though. In Europe, and most of the rest of the world, everybody uses interchangeable GSM phones. The carriers have less control over handsets.
This business of storing used fuel rods at working reactors needs to stop. Once they've cooled off enough for dry cask storage, they should go to dry storage in a mountain somewhere.
The US repository was supposed to be in Yucca Mountain. Early plans called for a second repository somewhere on the east coast, probably in hard-rock mountains in Maine or Vermont. There are mountain ranges that haven't done anything exciting geologically in the last 20 million years or so, and those are the best sites.
Google used to offer search through a SOAP API. With no ads. They discontinued that. Then they offered search through a "Web Search API", again with no ads. That's deprecated, rate limited, and going away, although it still works if you have an old API key for it. ("Google Custom Search" does not support general web searches, and costs $5 per 1000 queries.)
Slashdot Mirror
What's needed is a big lawsuit by a big bank against Microsoft for willful negligence. (Def: Intentional performance of an unreasonable act in disregard of a known risk, making it highly probable that harm will be caused.") Knowingly distributing operating systems which are known to be remotely exploitable to attack other systems fits that definition.
Microsoft's EULA doesn't protect them here. The victim is a third party, not their own customer, and not a signatory to the EULA. Nor does this require a class action. There are single banks big enough to take this on.
These guys are claiming to have controlled thermonuclear fusion above breakeven. That's huge. No one has ever done that. If it works, we have a new major power source. They write: "Now, the key will be combining each isolated test into a final experiment that produces fusion using this technology". That's a Nobel prize if they succeed.
This is too important to let NASA fuck up.
I read the article and went to edX. At edX I signed up; (semicolon should be a comma) but, (misplaced comma) I can not (should be "cannot") find out about this system. Quite Frankly,(lower case!) I am a teacher and I need my students to be writing more; (Run-on sentence. Use a period.) however, I do not have the time to grade all of their papers (comma here) so I have been assigning more objective homework that I would like.
Grade: D
All banks can only pay out a fraction of deposits at any given time.
Mt. Gox is not a bank. Mt. Gox is a payment services firm under Japan's Payment Services Act of 2009. That law allowed non-bank businesses to do payment services, and, since then, many of the mobile operators in Japan run payment services. But payment service firms in Japan are not allowed to engage in fractional reserve banking. "The PSA will impose an obligation on an operator to secure the assets in amounts equal to or more than the total amount of: (i) funds which an operator is transmitting; and (ii) procedural costs in relation to reimbursement of such funds as set out in (i), so that the transferred funds can reach the recipient even in the event of an operator's insolvency."
So Mt. Gox has to have at least 100% of the deposited funds as hard assets.
Almost all the "online wallet" companies have at some point lost customer money. Instalwallet is just the latest. Bitomat, MyBitcoin, and some others also tanked. Bitcoin.org now has a warning: "Web wallets host your bitcoins. That means it is possible for them to lose your bitcoins following any incident on their side. As of today, no web wallet service provide enough insurance to be used to store value like a bank."
They're unregulated depositary institutions. Historically, those don't end well. Keeping much money in Bitcoin "exchanges" is iffy, too. Mt. Gox has withdrawal rate limits on Bitcoins, which is suspicious. They should be able to pay out 100% of their Bitcoin balances at any time. If they can't, they're skimming.
Like Erlang, it provides the shared-nothing actor model for concurrent programming at the language level, instead of mucking with threads and global state directly.
That's encouraging. The Go language designers tried to do that, but didn't quite get it right. They talk a good game of "Don't communicate by sharing memory; share memory by communicating". But in reality, Go shares data (usually, too much data) between concurrent tasks. When you pass a reference through a Go channel, you're sharing data. Go is prone to race conditions. You can even break through the memory model and craft exploits by exploiting race conditions. As a result, Google's AppEngine has to lock down Go programs to one thread.
(There are three big memory safety holes in C: "How big is it", "Who can delete it", and "Who locks it". C++ tries to address #1, but doesn't quite succeed. Languages with GC address #2. Few languages (Ada and Erlang being exceptions) address #3 well. Go does not, Rust apparently does.)
We'll have to see how the Rust people do. Stuff that comes out of Mozilla tends to have a huge number of half-implemented features. (That's so open-source.) Go is a coherent design, even if flawed. It's from some of the old Plan9 people at Bell Labs.
Actually, I think it was Microsoft that killed them off...
Right. And they heavily pressured vendors not to sell Linux netbooks.
There is, though, a thriving Linux netbook industry in China.
Aereo is a legal hack. Each user has their very own UHF antenna. The receiving center has thousands of tiny UHF antennas, one per user, each driving their own private file store. It's a remote DVR.
Remember netbooks? They got to be rather good under-$300 laptop computers before the industry killed them off for not being expensive enough and not requiring any expensive or intrusive "cloud service".
"All these worlds are yours, except Europa. Attempt no landing there."
Provide a few key apps, and wearing Google Goggles will be made illegal.
Bitcoin itself doesn't need much regulation, but some of the companies profiting from it do. Entities which hold onto your money, from PayPal to Mt. Gox, are depository institutions, and should be regulated as such. Customers need to know who's behind them and that they really have the money. There's a long list of defunct "Bitcoin exchanges" which took the money and ran. That's where regulation is needed.
Customers need strong enforceable rights against exchanges. Mt. Gox is notorious for blaming their own business partners (OKPay, Dwolla, etc.) when something goes wrong. They also limit how fast you can take money out (even in Bitcoins) change those limits from time to time, and the limits are very low by banking standards. Since Mt Gox supposedly has 100% of the deposited Bitcoins, they should be able to pay out 100% of any account on demand.
DARPA has had some of those two-handed robots for a few years now, and has researchers working with them. So far, the results have been disappointing. They can pick up a screwdriver, but not use it. Robots manipulating a powered socket wrench in a well-defined environment aren't a big deal; auto assembly lines have had those for years. Putting a key in a lock, slowly, is about the upper limit of that project.
Part of the problem is that simulators for manipulation aren't very good. Willow Garage is funded to take the current version of the Gazebo simulator and make it work for manipulation. But I doubt they'll be able to make it good enough to do manipulation using force feedback. The physics engines they're using are for game physics; they cheat on frictional contacts. You need a really good simulator to debug control software for putting a nut on a bolt by feel.
They don't seem to be able to accomplish much or cooperate very well. Better work with swarm robots has been done.
One of the better robotics ideas of the 1980s was a pair of small (about 1 cubic foot) robots with small forklifts. They could cooperate to move large objects, like a couch. One was in charge, and one was the helper. Once both robots were in position, force feedback and very limited communication was enough to coordinate them.
That's a useful concept to develop further today. In the 1980s, navigation and vision weren't good enough to make this work reliably. Now they are.
Very nice. Festo, which is a German industrial robotics firm, does a technical tour de force every year. They built a robotic bird two years ago. In 2009, they built a robot penguin" which swims beautifully.
Festo does this to sell their industrial robotics systems, which are very well made.
This collision has one particular effect that does not meet standard thoughts on free markets; competition brings prices UP.
That's very common. In the antiquated "free market" view, competition inevitably drives prices down. In modern marketing, competition is by features, conveniance, marketing, and status symbol value. (Academic journals are in the status symbol category.) Pricing is driven by implicit or explicit collusion, with competitors striving to push prices upwards.
This model applies to appliances, autos, cell phones, music, movie tickets, etc. Some things still have price competition, but they're mostly commodities.
Back in my DARPA Grand Challenge days, I saw fully automated parking as the first "killer app" for automated driving. Everybody was obsessed with automated freeway driving, but that's not what annoys people. Looking for parking annoys people. The general idea is that you get out of your car at your destination, and it goes and parks itself somewhere. When you want your car back, you call it and it comes to you. Parking then need not be as close to the destination; a big parking garage a mile away is fine.
The first application of this should have been for airport rental cars. You rent the car via your phone, and the car comes to the loading area near baggage claim and picks you up. When you're done with the car and at the airport, you get out at the departure area, and it drives itself to rental car return. Customers would save an hour on every plane trip. That would sell.
It's workable. At no time is autonomous operation above about 20MPH necessary, which means slamming on the brakes is sufficient to deal with most problems. All the rental cars are new and under common ownership and maintenance, so the self-driving systems can be checked out on every rental. The system could be expanded to include the top 10 destinations for rental cars - major hotels, convention centers, etc.
After 9/11, no way would autonomous vehicles be allowed in an airport terminal area. So that didn't look promising back in the mid-2000s. Today, though, with terrorism down to nuisance levels, it's worth looking at again.
As for VW thinking that automated driving is more than a decade away, both Ford and Mercedes have said they expect to have it in production vehicles in five years.
This could be interesting. Servers are still designed like PCs. That's not fundamental. One could have compute servers which have network hardware that is configured during the boot process and which restricts what the machine can talk to. Their storage is all external, on file servers elsewhere. They have no peripherals other than the network. They barely need an operating system - the remote management hardware in the network interface handles administration.
With Linux at 15,000,000 lines of code, there's a bloat problem. There's still a need for a run-time library, but it will be more like the C run time system than an OS.
Shouldn't we have some book that explains how to quickly bootstrap ourselves back to current levels of civilization?
Such information was collected in the 1950s, stored on microfiche, and copies provided to major US fallout shelters, along with a reader that only needed sunlight. I have no idea where to get that now.
There's a classic set of books, Build Your Own Metalworking Shop From Scrap, by David Gingery, just for this purpose. The first book, Build Your Own Charcoal Foundry, starts the process. The books do assume there's plenty of metal scrap around; you're not expected to start from ore.
Getting up to pre-WWII levels from very little can be done. Much of Europe and Japan had to do it after WWII. Although many of the skills have been lost. How many people today can operate a lathe?
Where's the announcement from China? Canonical has a long history of bullshit announcements that some big vendor is going to use their product. They've made that claim in the past for both Asus and Dell. In both cases, the Canonical product never appeared on those platforms, or was a very minor niche announcement.
I'm not finding any announcement about this on China government sites. However, the Ministry of Information Industry Software and Integrated Circuit Promotion Center is listed by Microsoft as an Microsoft Embedded Partner.
Here's a recent policy announcement on open source from CSIP. They encourage using Linux, but Canonical is not mentioned. The action agency on this is the "China Innovative Strategic Alliance for Open Source", but I'm not seeing them associated with Canonical.
A book written by a committee will be a painful read.
If you want to do something useful, create one of those single plastic sheet two page guides to the language. Boiling the language down to two pages of small type with a few diagrams is a useful exercise. More useful than another thousand page book of blithering.
There's a tradeoff between lifespan and maintenance requirements. For fun, I restore old Teletype machines from the 1920s and 1930s. I have four of them running.
A normal maintenance operation on early Teletypes is to remove the two electrical parts (the motor and the selector magnet) and soak the entire machine in cleaning solution to get rid of dust and dead oil. For machines in heavy use, Western Union did that annually. Then they had to be oiled again (there are several hundred oiling points and six pages of lubrication instructions), gears and sliding joints greased, the electrical parts re-installed, adjustment procedures performed, and the machines re-tested.
Because of this design for maintainability, I've been able to take 80 year old machines that were covered with rust and dirt, and restore them to full operation. But who would put up with something today which required that kind of maintenance? Getting people to clean or change the filters on their desktop computers is difficult.
The big problem is that almost everybody in US "mobile", telephony, and cable already has a "walled garden". From Comcast to Apple, everybody in that space has a tight grip on their users. Most "apps" are really just a form of DRM.
This is a US thing, though. In Europe, and most of the rest of the world, everybody uses interchangeable GSM phones. The carriers have less control over handsets.
This business of storing used fuel rods at working reactors needs to stop. Once they've cooled off enough for dry cask storage, they should go to dry storage in a mountain somewhere.
The US repository was supposed to be in Yucca Mountain. Early plans called for a second repository somewhere on the east coast, probably in hard-rock mountains in Maine or Vermont. There are mountain ranges that haven't done anything exciting geologically in the last 20 million years or so, and those are the best sites.
Google used to offer search through a SOAP API. With no ads. They discontinued that. Then they offered search through a "Web Search API", again with no ads. That's deprecated, rate limited, and going away, although it still works if you have an old API key for it. ("Google Custom Search" does not support general web searches, and costs $5 per 1000 queries.)