This is the first step in GM's plan to revolutionize the automobile.
They've been working on a "platform" (literally) that uses electric drive and everything else just stacks on top of it. There was some discussion of even have replacable bodies.
They had been planning on using fuel cells to provide the power, but apparently have decided that a traditional IC (and hopefully diesel option) running a generator is a useful stop-gap.
There were several articles, including some in Wired IIRC, about 4 or 5 years ago predicting this new platform would be available in about 10 years. Looks like they're more or less on schedule.
Saving the DNA might be useful, but for many mammals and birds, there's much more to behavior than just DNA.
While it is not as dramatic as aliens saving human DNA without any of our culture, many animals don't function well if they don't have their parents (or other members of their species) to teach them how to survive.
Combine it with needing the rest of their habitat, and it is almost meaningless to talk about trying to "preserve" the species that way.
The original poster was suggesting just saving money and paying cash for services instead of getting insurance.
And you can negotiate with providers before you receive services to lower their prices if you're paying for it out of pocket. i.e. if you don't have insurance.
My point wasn't about negotiating insurance policies, but negotiating the actual price of specific services, which, as I pointed out, can be problematic in some circumstances.
The insurance companies negotiate with all of your providers, including some you're not even aware exist, for lower rates. And while you can do some negotiation yourself, that is a very difficult thing if you're lying on a stretcher unconcious.
At my most recent physical, the lab billed $900 for all of the tests. The insurance company paid $300 and the rest was the "negotiated discount".
The medical system in the US is fundamentally flawed, and facing it WITHOUT insurance could easily bancrupt you.
And while not an epidemic, from the article mentioned elsewhere:
"An interesting and reliable way of viewing graduation trends over time is to consider a statistic reported in the Digest of Education Statistics, 2000, Table 101. It reports the ratio of regular high school graduates (excluding GEDs) to the total 17 year-old population in the United States going back as far as 1870. This ratio is a reasonable approximation of a national graduation rate and can be consistently calculated for more than a century. The table shows that graduation rates steadily climbed to a peak of 77.1% in 1969 and have since fallen back to 70.6% in 2000, a level that was first achieved in 1963."
I was unable to extract the table from the government website (http://nces.ed.gov/programs/digest/) referenced, but perhaps someone else can.
So in any case, the graduation rate may have dropped 10% in the last 37 years. Not a good sign for the future.
While your numbers show 10% or so "drop-out" rates, the numbers from here (http://www.csmonitor.com/2006/0621/p03s02-ussc.ht ml) and here (http://www.manhattan-institute.org/html/cr_baeo.h tm) say closer to 30%.
Some of this is due to the latter studies not including GEDs (which are an indication of school failure, IMO) and some of it is probably due to sampling or other errors (discussed in the article).
Of the 50 largest school districts in the country, according to these numbers, 5 graduate less than half of their 8th graders. Only 5/50 graduate 80% of their students.
I have no idea what the numbers were like in 1972, but I can't imagine that things are significantly better now.
If you'd bothered to read their IPO or other documents, you'd know that common shareholders have nearly no control over the company due to the voting structure the owners set up.
They note that their policies may not maximize profit. Their policies are set based on doing what's right, as deemed by the founders.
Supporting the resale market increases the prices on the secondary market, and increases the value of the original goods.
If you know you can buy an Xbox, and resell it for 70% of the original price, it is a much better "investment" than if you can sell it for 30% of the original price.
If you report that the Chinese government is systemically jailing a class of people, that would undermine China's national unity.
If you report that the Chinese government is using slave labor to do something, that hurts China's reputation.
If you report that the Chinese government is performing industrial espionage, that hurts their interests.
If you report on their treatment of religious minorities or otherwise exercise what we in the US would consider our First Amendment rights, you violate their "religious policies".
Reporting what happened in Tienamen square probably counts as "inciting hatred".
And all of these very appropriate subjects for news reporting can be blocked by the Chinese government. Sorry, but government censorship is scary.
And they are technically "Debit Cards", so the money IS an asset instead of a liability. (And for that matter, having a net positive balance on your credit card is an asset as well..)
And last, money in your bank account, or spent on a credit card, is pretty much the same as cash. i.e. having it in the bank is the same as having cash. Spending it by using your credit card is the same as spending cash.
If you're using a "charge card" or a "debit card", instead of a "credit card", there's nothing wrong using plastic. It's only borrowing money at absurd rates for trivial purchases that's a problem.
What you fail to understand is that, for traditional TV, we are not customers. We're product. The TV stations are delivering us to their advertisers. The advertisers pay them, not us.
If the networks could have us chained to our sofas and forced to watch advertising for 8 hours a day, kept awake by electrical jolts, they'd do it in a heartbeat.
So anything that makes their advertisers unhappy results in worse conditions for the herd. I mean viewers.
Costco, however, pays significantly above what WalMart or SAMS club does. They also pay significantly above what Target and other retailers do.
I didn't mention WalMart in my post. I strictly talked about CostCo and how they treat their suppliers and employees, which is about as good as it gets in mainstream retail these days.
We can move the discussion to which company is the worst. And perhaps WalMart isn't the worst in terms of total compensation, turnover, shrinkage, etc. But they're close.
CostCo pays their employees very well for retail and treats their suppliers with respect. In return, their suppliers try hard to keep them as a customer, and CostCo's "shrinkage" (i.e. mostly employee theft) is the lowest in the industry.
Their profits are essentially the annual membership fee. Once you've paid that, you're buying everything pretty much at cost (including those higher salaries.)
They do not advertise and dispense with most of the corporate BS. Which means the customer doesn't have to pay for all of that overhead either.
That's why the compiler and underlying OS are tested and certified to the same level as the application in avionics applications.
That's why it is preferred to actually do the certification testing on the real hardware, using real inputs and outputs, so you exercise the hardware micro-code, compiler, OS, etc. when doing certification.
And for most life-critical systems, you end up detecting failures and dropping into some kind of "fail safe" mode rather than trying for perfect reliability. The interactions and failure modes of your correction algorithms and hardware are harder to guarantee correct than just detecting the failure and dropping into a fail-safe (which varies depending on the type of system from a special mode to a second redundant piece of hardware). Of course you can minimize failures to begin with by using appropriate system design, hardware and software techniques. But for almost every component in the system, you have to be able to function with the component failed, as it could be physically damaged, have power fail, etc. while in flight.
And yes, the avionics industry uses lots of very expensive processes, hardware, and software techniques and tools as necessary to meet the very strict safety requirements. "Good, Fast, Cheap, FAA Certified. Pick Two."
While I'll certainly agree that certain best practices that are also used by Agile can be applied in a safety critical environment, I think the biggest problem is that safety critical processes tend to have more people, and more non-developers, involved than pure agile methods can easily handle.
Many of the advantages from Agile come from the fact that you are strongly encourage to ASK to the person who's the expert how something works. You pair program. You communication verbally. This doesn't leave the paper trail needed to do a safety analysis on your changes.
Agile methods also eliminate redundancy in documentation by writing as little documentation as possible at high levels, and using lower level forms (such as unit test cases) to express them precisely.
This doesn't scale well to larger organizations. Somewhere around 10-20 developers is usually given as the threshold, although I think I've seen a few numbers higher.
It also doesn't scale well to an environment where you have safety engineers, certification engineers and systems engineers who have to read the requirements and understand how they change especially if those people do not read code. (Not to mention the people at the certifying agency.) In fact, the people who are certifying (testing) the product against the requirements are not ALLOWED to be the people who write the code in our environment. This means a very detailed requirement specification has to be written (vs. use cases on index cards.)
And things that are a very good idea, like refactoring often, become a bigger issue when you have to document every change so that several different people can analyze the safety implications of your code change. There can be significant regulatory implications with significant customer expense if you decide to touch too many functions in the system, even if the external system behavior does not change.
If you run through the extreme model, about half the things Just Don't Work (or at least are Very Insufficient) in a safety critical environment. (And things like automated builds and solid SCM are a no-brainer in ANY environment, and are pretty much required in a safety critical environment.)
Certainly minimizing the documentation, audits, etc. is a laudable goal, and we do as much of it as possible. And everyone should take advantage of best practices as much as possible. And automating as much of the process (both for the developer and the organization) is a fine thing. But I have yet to see a safety critical process that is even close to Agile.
Speaking as someone who does life critical software development, the LAST thing they are going to be able to use with FDA approval is an Agile process.
These processes are all about mega-reams of documentation, proof that you've done all kinds of analysis, and audits of all of your process.
I haven't investigated this product, but I think this is "per developer", not a floating seat license.
Realistically, when the time comes to do debugging, you wouldn't want fewer seats than developers anyway.
And you're not really running on a PC here. You'd almost certainly be running on proprietary hardware, hooked up to special power supplies, with bizarre aviation hardware interfaces. Each target box might run a substantial fraction of $18k.
And to be honest, the cost of generating DO-178B Level A documentation on any substantial body of code is certainly worth the a couple hundred thousand dollars they're charging. (Depending on the costs of integrating their documentation into your process.)
But it's still a LOT of money. (Which is why people don't write open source aviation equipment..)
You are correct. If it is actually panel mounted, and not "removable", it will be certified to some degree. It maybe just be for "situational awareness" and not connected to any kind of flight guidance or be certified for GPS approaches or even WAAS.
I'm just more familiar with the "handheld" units (Frequently semi-permanently mounted) that aren't certified at all.
And a little research turns up per-developer pricing, although not the per-unit run-time license cost. That's not actually unreasonable, given the cost of DO-178B Level A documentation, but still. Ouch.
Price and Availability
In addition to the LynxOS-178 kernel, the offering also includes a complete artifacts package for the kernel and user library, DO-178B required documentation, code coverage test suites and analysis for 100% modified condition/decision coverage of the kernel and libraries, a full suite of standards-based development tools, and support. The company will also soon release the industry's first commercial-off-the-shelf certifiable TCP/IP stack. Development seats, including the LynxOS-178 kernel and one year of priority support, start at $18,000.
Handheld and panel mount GPS systems are typically not certified for use in aircraft. They may be used to "aid situational awareness" but are not to be used to navigate the aircraft.
It isn't completely clear to me what you're trying to do.
What is the cost/impact of a failure in your system? Are we talking a little bit of money (website), bunch of money (stock trading) or are we talking people dying? Or just your boss saying "I want 24x7 availability for this app darnit!"?
I work on embedded systems using a limited dialect of C++ on safety critical systems. We, for instance, don't get to use dynamic memory allocation because it makes a system's behavior non-deterministic. (And besides, what do you do if you try to allocate memory and it fails?) Multi-tasking is just right out for formal certification of the application.
Also, are you looking at high availability or high reliability? They're similar, and use some of the same techniques, but have different aims. In my environment, we'd rather fail and bring down a system than leave a system up that is producing even slightly incorrect data. (And we have a redundant unit in case of hardware failure, etc.) But in high availability, you want to keep operating, even if degraded. (Think telephone switches, where the ability to carry half the call with some static is better than nothing.)
In general, however, you want as few components (both software and hardware) as necessary to do the job. Assuming a single system can handle the work, splitting it across three boxes just adds two more points of hardware failure, 2 possible communication failures, and a bunch of software to handle inter-machine communication. Every library, every OS call, and every component is a potential source of failure; use with caution.
And in the end, if you really can't have failures, go look at formal development, certification and testing practices specified by the FDA or FAA. They don't guarantee it'll be perfect, but they give you an idea of the kind of work you need to get close. For example, testing that has MCDC code coverage of >90%. The cost will be an order of magnitude higher than for normal software, but it won't fail. And they have advice on use and certification of third party libraries and OSes. (Summary: Libraries verifiable to these levels also cost an order of magnitude more money.)
Those methods will also give you advice on availability assessments of hardware, etc. Is an average server or PC sufficiently reliable for your application? This has to be one of your first assessments. (i.e. how quickly can you get hard drives replaced, and what's the simultaneous failure probability of 2 drives in a RAID/mirror setup?)
Slashdot Mirror
This is the first step in GM's plan to revolutionize the automobile.
They've been working on a "platform" (literally) that uses electric drive and everything else just stacks on top of it. There was some discussion of even have replacable bodies.
They had been planning on using fuel cells to provide the power, but apparently have decided that a traditional IC (and hopefully diesel option) running a generator is a useful stop-gap.
There were several articles, including some in Wired IIRC, about 4 or 5 years ago predicting this new platform would be available in about 10 years. Looks like they're more or less on schedule.
Saving the DNA might be useful, but for many mammals and birds, there's much more to behavior than just DNA.
While it is not as dramatic as aliens saving human DNA without any of our culture, many animals don't function well if they don't have their parents (or other members of their species) to teach them how to survive.
Combine it with needing the rest of their habitat, and it is almost meaningless to talk about trying to "preserve" the species that way.
The original poster was suggesting just saving money and paying cash for services instead of getting insurance.
And you can negotiate with providers before you receive services to lower their prices if you're paying for it out of pocket. i.e. if you don't have insurance.
My point wasn't about negotiating insurance policies, but negotiating the actual price of specific services, which, as I pointed out, can be problematic in some circumstances.
Here's a good reason why you can't:
The insurance companies negotiate with all of your providers, including some you're not even aware exist, for lower rates. And while you can do some negotiation yourself, that is a very difficult thing if you're lying on a stretcher unconcious.
At my most recent physical, the lab billed $900 for all of the tests. The insurance company paid $300 and the rest was the "negotiated discount".
The medical system in the US is fundamentally flawed, and facing it WITHOUT insurance could easily bancrupt you.
And while not an epidemic, from the article mentioned elsewhere:
"An interesting and reliable way of viewing graduation trends over time is to consider a statistic reported in the Digest of Education Statistics, 2000, Table 101. It reports the ratio of regular high school graduates (excluding GEDs) to the total 17 year-old population in the United States going back as far as 1870. This ratio is a reasonable approximation of a national graduation rate and can be consistently calculated for more than a century. The table shows that graduation rates steadily climbed to a peak of 77.1% in 1969 and have since fallen back to 70.6% in 2000, a level that was first achieved in 1963."
I was unable to extract the table from the government website (http://nces.ed.gov/programs/digest/) referenced, but perhaps someone else can.
So in any case, the graduation rate may have dropped 10% in the last 37 years. Not a good sign for the future.
While your numbers show 10% or so "drop-out" rates, the numbers from here (http://www.csmonitor.com/2006/0621/p03s02-ussc.ht ml) and here (http://www.manhattan-institute.org/html/cr_baeo.h tm) say closer to 30%.
Some of this is due to the latter studies not including GEDs (which are an indication of school failure, IMO) and some of it is probably due to sampling or other errors (discussed in the article).
Of the 50 largest school districts in the country, according to these numbers, 5 graduate less than half of their 8th graders. Only 5/50 graduate 80% of their students.
I have no idea what the numbers were like in 1972, but I can't imagine that things are significantly better now.
If you'd bothered to read their IPO or other documents, you'd know that common shareholders have nearly no control over the company due to the voting structure the owners set up.
They note that their policies may not maximize profit. Their policies are set based on doing what's right, as deemed by the founders.
Maybe you should sell your shares.
Supporting the resale market increases the prices on the secondary market, and increases the value of the original goods.
If you know you can buy an Xbox, and resell it for 70% of the original price, it is a much better "investment" than if you can sell it for 30% of the original price.
The same can be said for the games.
Let's see:
If you report that the Chinese government is systemically jailing a class of people, that would undermine China's national unity.
If you report that the Chinese government is using slave labor to do something, that hurts China's reputation.
If you report that the Chinese government is performing industrial espionage, that hurts their interests.
If you report on their treatment of religious minorities or otherwise exercise what we in the US would consider our First Amendment rights, you violate their "religious policies".
Reporting what happened in Tienamen square probably counts as "inciting hatred".
And all of these very appropriate subjects for news reporting can be blocked by the Chinese government. Sorry, but government censorship is scary.
Except the masons are humans and conspiring to destroy humanity.
The Scarlet Crusade is human, and they're evil.
Most of the pirates (nice folks there) are humans.
Oh yeah, the Barov family are great examples of human kind.
Stratholm is a human city, right?
Let's see, what else have I missed...
Can you show me a model of VCR that allows me to watch one recorded show while it is simultaneously recording another one?
So, teaching kids that cash is a toy is better?
And they are technically "Debit Cards", so the money IS an asset instead of a liability. (And for that matter, having a net positive balance on your credit card is an asset as well..)
And last, money in your bank account, or spent on a credit card, is pretty much the same as cash. i.e. having it in the bank is the same as having cash. Spending it by using your credit card is the same as spending cash.
If you're using a "charge card" or a "debit card", instead of a "credit card", there's nothing wrong using plastic. It's only borrowing money at absurd rates for trivial purchases that's a problem.
What you fail to understand is that, for traditional TV, we are not customers. We're product. The TV stations are delivering us to their advertisers. The advertisers pay them, not us.
If the networks could have us chained to our sofas and forced to watch advertising for 8 hours a day, kept awake by electrical jolts, they'd do it in a heartbeat.
So anything that makes their advertisers unhappy results in worse conditions for the herd. I mean viewers.
Costco, however, pays significantly above what WalMart or SAMS club does. They also pay significantly above what Target and other retailers do.
I didn't mention WalMart in my post. I strictly talked about CostCo and how they treat their suppliers and employees, which is about as good as it gets in mainstream retail these days.
We can move the discussion to which company is the worst. And perhaps WalMart isn't the worst in terms of total compensation, turnover, shrinkage, etc. But they're close.
CostCo pays their employees very well for retail and treats their suppliers with respect. In return, their suppliers try hard to keep them as a customer, and CostCo's "shrinkage" (i.e. mostly employee theft) is the lowest in the industry.
Their profits are essentially the annual membership fee. Once you've paid that, you're buying everything pretty much at cost (including those higher salaries.)
They do not advertise and dispense with most of the corporate BS. Which means the customer doesn't have to pay for all of that overhead either.
That's why the compiler and underlying OS are tested and certified to the same level as the application in avionics applications.
That's why it is preferred to actually do the certification testing on the real hardware, using real inputs and outputs, so you exercise the hardware micro-code, compiler, OS, etc. when doing certification.
And for most life-critical systems, you end up detecting failures and dropping into some kind of "fail safe" mode rather than trying for perfect reliability. The interactions and failure modes of your correction algorithms and hardware are harder to guarantee correct than just detecting the failure and dropping into a fail-safe (which varies depending on the type of system from a special mode to a second redundant piece of hardware). Of course you can minimize failures to begin with by using appropriate system design, hardware and software techniques. But for almost every component in the system, you have to be able to function with the component failed, as it could be physically damaged, have power fail, etc. while in flight.
And yes, the avionics industry uses lots of very expensive processes, hardware, and software techniques and tools as necessary to meet the very strict safety requirements. "Good, Fast, Cheap, FAA Certified. Pick Two."
While I'll certainly agree that certain best practices that are also used by Agile can be applied in a safety critical environment, I think the biggest problem is that safety critical processes tend to have more people, and more non-developers, involved than pure agile methods can easily handle.
Many of the advantages from Agile come from the fact that you are strongly encourage to ASK to the person who's the expert how something works. You pair program. You communication verbally. This doesn't leave the paper trail needed to do a safety analysis on your changes.
Agile methods also eliminate redundancy in documentation by writing as little documentation as possible at high levels, and using lower level forms (such as unit test cases) to express them precisely.
This doesn't scale well to larger organizations. Somewhere around 10-20 developers is usually given as the threshold, although I think I've seen a few numbers higher.
It also doesn't scale well to an environment where you have safety engineers, certification engineers and systems engineers who have to read the requirements and understand how they change especially if those people do not read code. (Not to mention the people at the certifying agency.) In fact, the people who are certifying (testing) the product against the requirements are not ALLOWED to be the people who write the code in our environment. This means a very detailed requirement specification has to be written (vs. use cases on index cards.)
And things that are a very good idea, like refactoring often, become a bigger issue when you have to document every change so that several different people can analyze the safety implications of your code change. There can be significant regulatory implications with significant customer expense if you decide to touch too many functions in the system, even if the external system behavior does not change.
If you run through the extreme model, about half the things Just Don't Work (or at least are Very Insufficient) in a safety critical environment. (And things like automated builds and solid SCM are a no-brainer in ANY environment, and are pretty much required in a safety critical environment.)
Certainly minimizing the documentation, audits, etc. is a laudable goal, and we do as much of it as possible. And everyone should take advantage of best practices as much as possible. And automating as much of the process (both for the developer and the organization) is a fine thing. But I have yet to see a safety critical process that is even close to Agile.
Speaking as someone who does life critical software development, the LAST thing they are going to be able to use with FDA approval is an Agile process.
These processes are all about mega-reams of documentation, proof that you've done all kinds of analysis, and audits of all of your process.
Do a google on Alex the gray and Dr. Pepperburg.
She's at University of Arizona last I heard.
Parrots can recognize and name multiple objects, colors, numbers, and materials and use the words correctly.
I haven't investigated this product, but I think this is "per developer", not a floating seat license.
Realistically, when the time comes to do debugging, you wouldn't want fewer seats than developers anyway.
And you're not really running on a PC here. You'd almost certainly be running on proprietary hardware, hooked up to special power supplies, with bizarre aviation hardware interfaces. Each target box might run a substantial fraction of $18k.
And to be honest, the cost of generating DO-178B Level A documentation on any substantial body of code is certainly worth the a couple hundred thousand dollars they're charging. (Depending on the costs of integrating their documentation into your process.)
But it's still a LOT of money. (Which is why people don't write open source aviation equipment..)
You are correct. If it is actually panel mounted, and not "removable", it will be certified to some degree. It maybe just be for "situational awareness" and not connected to any kind of flight guidance or be certified for GPS approaches or even WAAS.
I'm just more familiar with the "handheld" units (Frequently semi-permanently mounted) that aren't certified at all.
And something running NT4.0? *shiver*
Handheld and panel mount GPS systems are typically not certified for use in aircraft. They may be used to "aid situational awareness" but are not to be used to navigate the aircraft.
Everyone does, but they're not certified.
Except that you don't get a better rate plan when you provide your own phone.
So you're still paying $10-$15 a month in subsidy for a phone you didn't even get "for free".
It isn't completely clear to me what you're trying to do.
What is the cost/impact of a failure in your system? Are we talking a little bit of money (website), bunch of money (stock trading) or are we talking people dying? Or just your boss saying "I want 24x7 availability for this app darnit!"?
I work on embedded systems using a limited dialect of C++ on safety critical systems. We, for instance, don't get to use dynamic memory allocation because it makes a system's behavior non-deterministic. (And besides, what do you do if you try to allocate memory and it fails?) Multi-tasking is just right out for formal certification of the application.
Also, are you looking at high availability or high reliability? They're similar, and use some of the same techniques, but have different aims. In my environment, we'd rather fail and bring down a system than leave a system up that is producing even slightly incorrect data. (And we have a redundant unit in case of hardware failure, etc.) But in high availability, you want to keep operating, even if degraded. (Think telephone switches, where the ability to carry half the call with some static is better than nothing.)
In general, however, you want as few components (both software and hardware) as necessary to do the job. Assuming a single system can handle the work, splitting it across three boxes just adds two more points of hardware failure, 2 possible communication failures, and a bunch of software to handle inter-machine communication. Every library, every OS call, and every component is a potential source of failure; use with caution.
And in the end, if you really can't have failures, go look at formal development, certification and testing practices specified by the FDA or FAA. They don't guarantee it'll be perfect, but they give you an idea of the kind of work you need to get close. For example, testing that has MCDC code coverage of >90%. The cost will be an order of magnitude higher than for normal software, but it won't fail. And they have advice on use and certification of third party libraries and OSes. (Summary: Libraries verifiable to these levels also cost an order of magnitude more money.)
Those methods will also give you advice on availability assessments of hardware, etc. Is an average server or PC sufficiently reliable for your application? This has to be one of your first assessments. (i.e. how quickly can you get hard drives replaced, and what's the simultaneous failure probability of 2 drives in a RAID/mirror setup?)