The cockpit is not a separate pressure vessel. Whether there is a cockpit door or not, the cockpit and cabin will de-pressurize together.
That said, de-pressurization at altitude is a serious situation, but should be survivable by most of the passengers most of the time. I would think that people with medical conditions might well suffer serious injury or die, but that is certainly not a forgone conclusion.
One friend used to fly an airplane that would commonly experience catastrophic de-pressurization events at 25,000 feet (it was a turbo-charged piston twin). There were never any injuries, although it probably isn't the most pleasant way to spend your day.
Me too. There would not be ANY question of letting strangers walk away with my 2yo. I probably would allow them to wand a young child while I held them, but nothing more.
If they couldn't accommodate that, I would not fly.
I agree with the general sentiment that all these invasive searches suck.
Perspective students submit a detailed self-evalution for the degree of their choice, BA, BS, MA, MBA, or PhD. A TSU registrar will evaluate your application within 5-7 days and contact you via email with the results of their evaluation.
Re:Good news, folks
on
RT Linux Patches
·
· Score: 2, Informative
Actually, I think you are a bit confused. LynxOS is NOT Linux. The company (LynuxWorks) sells two operating systems, their own hard-realtime OS (LynxOS) and also Linux. They tend to do a bait-and-switch thing: you bring them in to buy Linux and they try to talk you into using their LynxOS operating system instead. Depending on what you are doing, LynxOS can be a fine hard realtime OS. I just don't appreciate the whole bait-and-switch thing.
As for DO178B, (I've worked on a DO178B avionics project) it is indeed very difficult to achieve level A certification. WindRiver's VxWorks is level A certifiable (which means if you give them a large percentage of a million dollars they will help you get your VxWorks implementation certified to level A).
What you should realize about how DO-178B works for software, however, is that it is mostly about having a documented process for software development. You come up with a process, and then show through a paper trail that you have followed the process. You can sort of do it backwards by reverse-engineering the paperwork (which is what WindRiver tends to do) but that is perverting the aim - to improve software engineering quality by having and following a good process for developing software).
In any case, DO-178B would be a difficult process for Linux. It's certainly not impossible, but it would be wildly expensive. DO-178B is not very compatible with a fast moving, ever changing development process. (In fact, it's pretty much designed to prevent just that!).
Level-A certification is probably best left to the small dedicated RTOS market, but Level C might be possible for Linux if there was enough interest. Level D is for stuff like coffee pots and would probably be easy (but I don't remember the details for Level D, so I'm just talking out of my ass about it).
The main DO-178B RTOS market is Enea's OSE (which we used), VxWorks, and now LynxOS. There are probably others that I'm not aware of, but it's a pretty small community of OSs.
"ground effect... essential for... especially an emergency autorotation landing in something [other] than an autogyro".
Not sure why you think this. During an autorotation in a helicopter or an autogyro, airflow is upward through the rotor. Even during the flare phase, airflow is upward, and thus ground effect does not apply. During the landing phase (hovering autorotation) the airflow is briefly downward, but I doubt that ground effect has much effect - this phase only lasts 2-5 seconds. I've never seen any studies, but I doubt you get much of a vortex going in that amount of time. Besides which - you usually have enough forward airspeed to be in translational lift, even at touchdown.
Experience: helicopter CFI/CFII for the last 25 years...
Check out my website sometime if you are interested in helicopters. www.copters.com.
The FAA is very strict about the stuff giving the pilot false indications, causing a crash. You should be glad they are anal about this stuff!
The big worry everybody has is during the final parts of an instrument approach. There are lots of things that can interfere with the glideslope and localizer. Vehicles parked too close to the glide slope antenna, for instance. The FAA takes steps to minimize the chance that something unlikely like a vehicle parked next to the antenna could screw up the signal in a way to make it give a false reading. They test all the navigation aids on a regular basis by having specially instrumented airplanes fly the approach and compare actual signals to expected signals. Really, a lot of work goes into trying to make sure there can not be interference with these signals, because the possible result is a lot of dead people.
There have already been known cases where avionics with microprocessors can take out radios when the interference happens to coincide with a harmonic of the aircraft radio frequency. A guy I used to know had that exact thing happen to him on a regular basis until the avionics company tweaked his box. I don't see why a microprocessor in the back is any less likely to cause interference (except distance, but you don't know where the antenae are).
So, do you really want the FAA, taking all these precautions to avoid interference with the navigation radios, to just ignore a huge potential source of interference from the cabin of the airplane? If they did, the first time a crash was attributed to that, heads would roll (and a lot of people might be dead).
The critical part is typically the last 2-5 minutes of the flight, when the aircraft is navigating using the ILS system, and obstructions are very close (sometimes 10s of feet - it's scary sometimes to fly instrument approaches in visual conditions and see just how close to stuff the approach takes you!).
It might not seem dangerous, and the chances of an accident being caused by this kind of interference is probably pretty low, but we really don't know for sure, so why take the chance?
The "West Wing" scenerio might seem stupid, but isn't. The poster to which you replied, in my opinion, was right on the mark.
My background is as an instrument flying instructor and a avionics architect/developer.
The reason given for the State Police at Logan Airport in Boston to carry silenced MP5 submachine guns is to not scare the public when they start shooting! (I heard it on the radio, so it must be true);-)
My personal belief is that they are carrying those weapons because they intend on using them under certain circumstances, i.e. I doubt it is a blufff. I don't know what restrictions are placed on the officers, but I'll bet they would shoot, even in a crowd, if they think their target is enough of a danger to themselves or to the public.
Similar to aviation - large commercial airliners, used in the attack, were operational again just a few days later. Small aircraft, not used in the attack, were basicallly grounded for a month.:-(
The "killer app" is realtime high def video delivery over the internet. Not just from a Blockbuster type web site, but peer to peer as well. Once you have that type of bandwidth availble to the average subscriber, you'll see a ton of video getting exchanged over the 'Net.
Can I really opt out of Microsoft if my employer uses it, my clients use it, and Microsoft makes their interfaces proprietary and/or patent encumbered in order to defeat open source alternatives?
At my last job, it was dictated that everybody use Microsoft. At my current job they simply make it difficult to not use Microsoft.
At some point, it becomes in theory possible to opt out, but in practice impossible.
Why are you booting it? Put it to sleep for goodness sake! Then it takes 2-3 seconds to be ready for use. Reboot it every 6 months when you upgrade the software!
I just wanted to put in a "job well done" plug for Applecare. Of course, Applecare costs a non-trivial amount ($350 or so, depending on the computer). I guess it's a question of you get what you pay for.
First of all, the wait times are typically 2-5 minutes. I wish I could turn off the background music so that I could just leave it on speakerphone until it gets transferred, but 2-5 minutes a couple times a year strikes me as okay.
Secondly, the TS specialists seem to know their stuff. They seem to have been well trained. I've been getting people in Austin... anyone know whether they are Apple empolyees or whether it is outsourced? I'd guess they are Apple employees, but I don't know for sure.
In any case, I called them twice recently. The first call was because the email application started croaking a lot (once or twice a day). The technical support specialist was very knowledgeable, spent quite a bit of time with me discussing what it might be, etc. He actually listened to my theory that it was related to the junk-mail rules engine, pointed out how I could reset the rules engine, and that in fact seemed to solve the problem. It's nice when tech support will actually listen to your theories:-)
The other call was because my laptop hard drive started making bad noises. I felt like they might have made me go through more steps than absolutely necessary (OS reinstall, disk-erase/OS reinstall) than absolutely necessary, but the trouble shooting they wanted me to do was not unreasonable. Also, they (after some persuading) were willing to send me a drive and let me replace it, rather than have to send them the laptop for a week (it's my everything-including-work computer, so it's tough to let it go for even a day, nevermind a week).
All in all, I've never felt like they wasted my time, the stuff they ask me to try makes sense, and ultimately I've had my problems solved with a reasonable minimum of fuss.
Applecare is not exactly cheap, but given that I tend to keep my Macs for 4-5 years, it's not an outrageous expense. I think other computer manufacturers could win some points with their customers by being more like Apple.
On the subject of 35mm (still) versus digital photos... I shoot and print both 35mm and professional digital (Nikon D1X) and I'd have to say they are comparable. Each has a couple characteristics in which it is superior to the other. The D1X is a 5 megapixel camera, btw.
Be careful about just comparing number of pixels, though. You need to look at stuff like noise, color rendition, etc. as well.
I'll second this - I used it to bridge my Internet connection from my townhouse to my neighbor's townhouse (so I could have Internet when I was visiting her). Security is probably just as bad as Wireless (anyone on the same transformer can tap in) but it's cheap and easy.
I got a Ethernet->Powerline, and then a Powerline->Wireless so that I can use my laptop via wireless, but I didn't have to run any wires, or do a pringles point-to-point link between the townhouses.
not to mention that #3 the iTunes hack was done on a windows machine! But, that was not mentioned, and the implication, as I read the article, was that it was an example of a MacOS-X weakness (even though that was not explicitly stated).
Not only that, but it wasn't even a hack of the DRM, it was an exploit of the analog hole - "Hey dude, I stuck a microphone next to your computer and recorded your music... I HACKED your DRM!"
Sigh. The article was one of the most biased, least informative I've seen to date. Somebody feels threatened.
I said: You act as if you can't pay people to write open source code.
You said: No I don't. I just didn't address that topic.
I think you did address it. You said:
It seems to me that a closed source shop would be theoretically better positioned to meet an immutable deadline (such as an election date). At least when you own your employees you can mandate overtime and crack the whip harder. When the software is open source you cannot enter "crunch mode" and make the scattered developers put in long hours.
In that paragraph, you are clearly linking open/closed source to being able to pay people, and thus "crack the whip" to meet deadlines, or not. I don't believe open/closed dictates anything about how the software gets written. It dictates who owns the software once it is written. I think it should be owned by the taxpayers who paid for it, and open to scrutiny by them.
>Tell me, who should pay?
The same people who are paying for proprietary solutions now! The government bodies who are paying Diebold and other companies to create propriatary e-vote solutions.
The "many eyes" argument is merely a shotgun approach to quality control. What is needed is strong leadership implemeting a plan which includes rigorous and ongoing testing. Open source does not guarantee this any more than closed source guarantees its absence.
Again, I have to disagree with you. The issue is not just quality assurance. It is a question of trust. In a closed source situation, we all have to trust the manufacturer that they did not put in trogan horses, etc. With open source, anyone, including the common citizen, can verify the operation of the software. He can scrutinize it, test it, and even verify that the binary that HE builds is the same as the binary running on the actual voting machine.
That's my issue - I don't want to have to trust that a corporation, or some employee of that corporation, doesn't have an agenda that includes tampering with votes.
There are still ways to spoof with open source, so I still want a paper trail. But I think open source makes it that much more difficult to launch an attack on an election. Right now, it appears to be laughably easy, and that makes me very very nervous.
You act as if you can't pay people to write open source code. That's not true. A company could be hired to produce and maintain an e-vote program, the source could be open - open to scrutiny, open to testing by third parties, open so that people can suggest improvements.
I think that closed software/hardware solutions need to be declared unconstitutional. If we can't trust that the results are accurate, we risk our democracy.
I also agree with the people that say you need to have a paper trail...
That's not true in my experience. My 266Mhz G3 with 192 meg runs OS X just fine... It's not as fast as my 800Mhz TiBook, but it runs pretty much as fast under OS X as it ever did under OS 9... I still use it for Photoshop work...
So, if my salary costs my company 4 times what one of the IT guys makes, and I'm forced to use a computer that makes me half as efficient as I could be, that benefits the company how?
Horror story at my last job: company IT policy: only PC hardware connected to the network (unless you were buddies with someone, then you could attach anything you wanted). User's can't install software - only IT can do that. All sorts of Nazi controls in place to prevent users from hosing the network. Except, hey! Several times a year my snazzy PC would hose the disk, the IT department would "reinstall" the OS, and I'd lose all my data, applications, and about 3 days of work. So, all their Nazi techniques made me unproductive, and the computer system STILL wasn't reliable. Their response? *SHRUG*.
Switched to a Mac running OS-X. No more problems with file systems suddenly going away. No more crashes. No more lost productivity. Typical uptime? 6 months - mostly reboots to load new versions of the operating system.
Predictably, IT was pissed that I was connecting a Mac, even tho a) it worked perfectly and b) it didn't need any attention from them at all (I'm sure that's what they were scared of).
The IT staff there was SO USED to the problem being the flakey Microsoft OS, that they would always just reload the OS, even when anyone with half a brain could see the problem was elsewhere. I had one PC that would reboot itself occasionally. I get grilled: no, I didn't install any funky software. No, no funky hardware. No, I didn't reboot it myself. No, nobody else walked up and rebooted it. Did you guys think about power? Maybe you should put a AC monitor in my office in case the power is flakey? No, they don't have anything like that.... Let's reload the OS again! Guess what it was... Power! Duh! The point is that they are so used to it being the lousy OS, they just always suspect that any problem is caused by the lousy OS. And... running this lousy OS benefits me how?
I'm a pilot, and in the aircraft I fly with NiCd batteries, there are many warnings systems, etc., because the batteries can thermally run away. If you're lucky, the battery just melts through the aircraft and fall out the bottom. If you're unlucky, it explodes and takes the nose off the aircraft. Seriously.
QNX isn't the only one. If you pay Wind River barrels of money (a good chunk of a million dollars) they'll help you get your project certified for life threatening projects (read: avionics and health).
The other one is Enea Data Systems, now known as OSE. I've used their OS, and it is designed for use in safety critical applications, including avionics, chemical plants, health, etc. It's a great little OS, btw.
Slashdot Mirror
That said, de-pressurization at altitude is a serious situation, but should be survivable by most of the passengers most of the time. I would think that people with medical conditions might well suffer serious injury or die, but that is certainly not a forgone conclusion.
One friend used to fly an airplane that would commonly experience catastrophic de-pressurization events at 25,000 feet (it was a turbo-charged piston twin). There were never any injuries, although it probably isn't the most pleasant way to spend your day.
If they couldn't accommodate that, I would not fly.
I agree with the general sentiment that all these invasive searches suck.
>>Microsoft may not be here in five years either.
Maybe they will exist, but with continued predictions-without-merit, maybe they won't be relevant?
Actually, I think you are a bit confused. LynxOS is NOT Linux. The company (LynuxWorks) sells two operating systems, their own hard-realtime OS (LynxOS) and also Linux. They tend to do a bait-and-switch thing: you bring them in to buy Linux and they try to talk you into using their LynxOS operating system instead. Depending on what you are doing, LynxOS can be a fine hard realtime OS. I just don't appreciate the whole bait-and-switch thing. As for DO178B, (I've worked on a DO178B avionics project) it is indeed very difficult to achieve level A certification. WindRiver's VxWorks is level A certifiable (which means if you give them a large percentage of a million dollars they will help you get your VxWorks implementation certified to level A). What you should realize about how DO-178B works for software, however, is that it is mostly about having a documented process for software development. You come up with a process, and then show through a paper trail that you have followed the process. You can sort of do it backwards by reverse-engineering the paperwork (which is what WindRiver tends to do) but that is perverting the aim - to improve software engineering quality by having and following a good process for developing software). In any case, DO-178B would be a difficult process for Linux. It's certainly not impossible, but it would be wildly expensive. DO-178B is not very compatible with a fast moving, ever changing development process. (In fact, it's pretty much designed to prevent just that!). Level-A certification is probably best left to the small dedicated RTOS market, but Level C might be possible for Linux if there was enough interest. Level D is for stuff like coffee pots and would probably be easy (but I don't remember the details for Level D, so I'm just talking out of my ass about it). The main DO-178B RTOS market is Enea's OSE (which we used), VxWorks, and now LynxOS. There are probably others that I'm not aware of, but it's a pretty small community of OSs.
Not sure why you think this. During an autorotation in a helicopter or an autogyro, airflow is upward through the rotor. Even during the flare phase, airflow is upward, and thus ground effect does not apply. During the landing phase (hovering autorotation) the airflow is briefly downward, but I doubt that ground effect has much effect - this phase only lasts 2-5 seconds. I've never seen any studies, but I doubt you get much of a vortex going in that amount of time. Besides which - you usually have enough forward airspeed to be in translational lift, even at touchdown.
Experience: helicopter CFI/CFII for the last 25 years...
Check out my website sometime if you are interested in helicopters. www.copters.com.
Paul
The big worry everybody has is during the final parts of an instrument approach. There are lots of things that can interfere with the glideslope and localizer. Vehicles parked too close to the glide slope antenna, for instance. The FAA takes steps to minimize the chance that something unlikely like a vehicle parked next to the antenna could screw up the signal in a way to make it give a false reading. They test all the navigation aids on a regular basis by having specially instrumented airplanes fly the approach and compare actual signals to expected signals. Really, a lot of work goes into trying to make sure there can not be interference with these signals, because the possible result is a lot of dead people.
There have already been known cases where avionics with microprocessors can take out radios when the interference happens to coincide with a harmonic of the aircraft radio frequency. A guy I used to know had that exact thing happen to him on a regular basis until the avionics company tweaked his box. I don't see why a microprocessor in the back is any less likely to cause interference (except distance, but you don't know where the antenae are).
So, do you really want the FAA, taking all these precautions to avoid interference with the navigation radios, to just ignore a huge potential source of interference from the cabin of the airplane? If they did, the first time a crash was attributed to that, heads would roll (and a lot of people might be dead).
The critical part is typically the last 2-5 minutes of the flight, when the aircraft is navigating using the ILS system, and obstructions are very close (sometimes 10s of feet - it's scary sometimes to fly instrument approaches in visual conditions and see just how close to stuff the approach takes you!).
It might not seem dangerous, and the chances of an accident being caused by this kind of interference is probably pretty low, but we really don't know for sure, so why take the chance?
The "West Wing" scenerio might seem stupid, but isn't. The poster to which you replied, in my opinion, was right on the mark.
My background is as an instrument flying instructor and a avionics architect/developer.
The reason given for the State Police at Logan Airport in Boston to carry silenced MP5 submachine guns is to not scare the public when they start shooting! (I heard it on the radio, so it must be true) ;-)
My personal belief is that they are carrying those weapons because they intend on using them under certain circumstances, i.e. I doubt it is a blufff. I don't know what restrictions are placed on the officers, but I'll bet they would shoot, even in a crowd, if they think their target is enough of a danger to themselves or to the public.
Similar to aviation - large commercial airliners, used in the attack, were operational again just a few days later. Small aircraft, not used in the attack, were basicallly grounded for a month. :-(
The "killer app" is realtime high def video delivery over the internet. Not just from a Blockbuster type web site, but peer to peer as well. Once you have that type of bandwidth availble to the average subscriber, you'll see a ton of video getting exchanged over the 'Net.
At my last job, it was dictated that everybody use Microsoft. At my current job they simply make it difficult to not use Microsoft.
At some point, it becomes in theory possible to opt out, but in practice impossible.
Why are you booting it? Put it to sleep for goodness sake! Then it takes 2-3 seconds to be ready for use. Reboot it every 6 months when you upgrade the software!
First of all, the wait times are typically 2-5 minutes. I wish I could turn off the background music so that I could just leave it on speakerphone until it gets transferred, but 2-5 minutes a couple times a year strikes me as okay.
Secondly, the TS specialists seem to know their stuff. They seem to have been well trained. I've been getting people in Austin... anyone know whether they are Apple empolyees or whether it is outsourced? I'd guess they are Apple employees, but I don't know for sure.
In any case, I called them twice recently. The first call was because the email application started croaking a lot (once or twice a day). The technical support specialist was very knowledgeable, spent quite a bit of time with me discussing what it might be, etc. He actually listened to my theory that it was related to the junk-mail rules engine, pointed out how I could reset the rules engine, and that in fact seemed to solve the problem. It's nice when tech support will actually listen to your theories :-)
The other call was because my laptop hard drive started making bad noises. I felt like they might have made me go through more steps than absolutely necessary (OS reinstall, disk-erase/OS reinstall) than absolutely necessary, but the trouble shooting they wanted me to do was not unreasonable. Also, they (after some persuading) were willing to send me a drive and let me replace it, rather than have to send them the laptop for a week (it's my everything-including-work computer, so it's tough to let it go for even a day, nevermind a week).
All in all, I've never felt like they wasted my time, the stuff they ask me to try makes sense, and ultimately I've had my problems solved with a reasonable minimum of fuss.
Applecare is not exactly cheap, but given that I tend to keep my Macs for 4-5 years, it's not an outrageous expense. I think other computer manufacturers could win some points with their customers by being more like Apple.
There was an Episode II ?
Be careful about just comparing number of pixels, though. You need to look at stuff like noise, color rendition, etc. as well.
I'll second this - I used it to bridge my Internet connection from my townhouse to my neighbor's townhouse (so I could have Internet when I was visiting her). Security is probably just as bad as Wireless (anyone on the same transformer can tap in) but it's cheap and easy. I got a Ethernet->Powerline, and then a Powerline->Wireless so that I can use my laptop via wireless, but I didn't have to run any wires, or do a pringles point-to-point link between the townhouses.
Not only that, but it wasn't even a hack of the DRM, it was an exploit of the analog hole - "Hey dude, I stuck a microphone next to your computer and recorded your music... I HACKED your DRM!"
Sigh. The article was one of the most biased, least informative I've seen to date. Somebody feels threatened.
"slowly" ?????
You said: No I don't. I just didn't address that topic.
I think you did address it. You said:
In that paragraph, you are clearly linking open/closed source to being able to pay people, and thus "crack the whip" to meet deadlines, or not. I don't believe open/closed dictates anything about how the software gets written. It dictates who owns the software once it is written. I think it should be owned by the taxpayers who paid for it, and open to scrutiny by them.>Tell me, who should pay?
The same people who are paying for proprietary solutions now! The government bodies who are paying Diebold and other companies to create propriatary e-vote solutions.
Again, I have to disagree with you. The issue is not just quality assurance. It is a question of trust. In a closed source situation, we all have to trust the manufacturer that they did not put in trogan horses, etc. With open source, anyone, including the common citizen, can verify the operation of the software. He can scrutinize it, test it, and even verify that the binary that HE builds is the same as the binary running on the actual voting machine.That's my issue - I don't want to have to trust that a corporation, or some employee of that corporation, doesn't have an agenda that includes tampering with votes.
There are still ways to spoof with open source, so I still want a paper trail. But I think open source makes it that much more difficult to launch an attack on an election. Right now, it appears to be laughably easy, and that makes me very very nervous.
I think that closed software/hardware solutions need to be declared unconstitutional. If we can't trust that the results are accurate, we risk our democracy.
I also agree with the people that say you need to have a paper trail...
Super computers cost more than 5 million dollars
Mainframes cost more than 1 million dollars
Mini-Super computers cost more than 1/4 million dollars
Everything else is by definition a Plain Jane (TM) computer
btw, I've worked on all 4 kinds ;-)
That's not true in my experience. My 266Mhz G3 with 192 meg runs OS X just fine... It's not as fast as my 800Mhz TiBook, but it runs pretty much as fast under OS X as it ever did under OS 9... I still use it for Photoshop work...
Horror story at my last job: company IT policy: only PC hardware connected to the network (unless you were buddies with someone, then you could attach anything you wanted). User's can't install software - only IT can do that. All sorts of Nazi controls in place to prevent users from hosing the network. Except, hey! Several times a year my snazzy PC would hose the disk, the IT department would "reinstall" the OS, and I'd lose all my data, applications, and about 3 days of work. So, all their Nazi techniques made me unproductive, and the computer system STILL wasn't reliable. Their response? *SHRUG*.
Switched to a Mac running OS-X. No more problems with file systems suddenly going away. No more crashes. No more lost productivity. Typical uptime? 6 months - mostly reboots to load new versions of the operating system.
Predictably, IT was pissed that I was connecting a Mac, even tho a) it worked perfectly and b) it didn't need any attention from them at all (I'm sure that's what they were scared of).
The IT staff there was SO USED to the problem being the flakey Microsoft OS, that they would always just reload the OS, even when anyone with half a brain could see the problem was elsewhere. I had one PC that would reboot itself occasionally. I get grilled: no, I didn't install any funky software. No, no funky hardware. No, I didn't reboot it myself. No, nobody else walked up and rebooted it. Did you guys think about power? Maybe you should put a AC monitor in my office in case the power is flakey? No, they don't have anything like that.... Let's reload the OS again! Guess what it was... Power! Duh! The point is that they are so used to it being the lousy OS, they just always suspect that any problem is caused by the lousy OS. And... running this lousy OS benefits me how?
I'm a pilot, and in the aircraft I fly with NiCd batteries, there are many warnings systems, etc., because the batteries can thermally run away. If you're lucky, the battery just melts through the aircraft and fall out the bottom. If you're unlucky, it explodes and takes the nose off the aircraft. Seriously.
The other one is Enea Data Systems, now known as OSE. I've used their OS, and it is designed for use in safety critical applications, including avionics, chemical plants, health, etc. It's a great little OS, btw.