UAV's (drones) have a horrible saftey record, true, but it is because the automation on them is quite poor. At this point, very few of them are "autonomous". The problem is not too much automation, but automotion that is not sufficiently rigorous, testable, nor capable to handle somewhat common contingencies.
I have multiple thousands of hours in heavy aircraft, and I can tell you that 'hand flying more' is not the solution to the problem either.
A well placed, rigorously programmed, redundantly powered backup system that could auto-land when normal power goes out is a much better soln than trying to land a heavy with no electrics (think: at night, in fog, heavy winds, ice, nuns/babies on board, etc).
Rockwell better make sure my retro encabulator is secure from those hackers!
If someone got access to the modial interaction of magneto reluctance and capacitive duractance, we could all be in a world of hurt.
this is great...
I mean, I was thinking this whole conversation yesterday in my head
gawd help us (elec/cmptr engrs) when the public no longer 'needs' our little trinkets
there is an awful bunch of crap (3D TV, 4K projectors, etc) that really have no REAL use
I just hope we all don't go back to burlap sacks and huts for homes
I really like getting wifi internet in my jacuzzi while watching my 60" flat screen
ok
go back to flaming each other
mod that sh!t up.
I mean, we are all just trying to make a living, I don't think anyone really wants world domination anymore - takes too much time.
meh, I could be wrong tho.
ok, that's one.
what about all those Millions of people in other sinking bowls ( sacramento river valley).
New Orleans is not unique, and you can't blame people who live there for being in the path of disaster.
besides, the old city of new orleans is above sea level, its the 'burbs that get flooded (definition of 'burbs in new orleans can get sticky though, but basically, any neighborhoods that existed a loooong time ago are well-tested with flood history)
there are many places [1] [2] in the world below sea level
this picture is a little exaggerated, but shows that the main threat is the mighty mississippi, not the sea. and the army corps of engineers has a divert-the-mississippi spillway upriver that virtually guarantees the river flood threat to mitigated.
ask anyone from new orleans (or others) and they will say that it was engineering that failed the city: intracoastal canals, notably MRGO, created for commerce gave intrusion paths to storm surge from the lake and the gulf. it was those levees that failed and spilled into the city.
from http://en.wikipedia.org/wiki/Drainage_in_New_Orleans we have:
On August 31, flood levels started to subside. The water level in the city had reached that of Lake Pontchartrain, and as the lake started to drain back into the Gulf, some water in the city started to flow into the lake via the same levee breeches they had entered through. In 19th century lake floods, the water soon flowed back into the lake as there were no levees on that side.
as humans, what makes us special is not just our ability to adapt, but to adapt the environment around us. If we never lived anywhere there was a threat of disaster, I am not sure where you could live(definitely not Texas, or a few other places. And those maps don't even include floods (the most common natural disaster), for that threats see this map of flood hazards for the US.
miami, ft lauderdale, houston, galveston, wilmington, lower manhattan, long island, etc.
I could go on and on.
New Orleans didn't invent "living below sea level"
now living in a spillway, the atchafalaya, which is flooded (by man, by God indirectly) fairly often, and is within the path to the Gulf that the mississippi really wants to go (very much so, in fact) - now that is a different story.
those folks know the risk and use buoyant foundations (trailers with styrofoam underneath) to maintain their homes.
they love living out there (it is real nice), and they understand they could lose everything one day.
other parts of the mississippi flood plain include MANY areas up north that flood MUCH more regularly, so lay off of new orleans
exactly,
people complain about new orleans, but manhattan is just as vulnerable to sea level rise/surge/flood. of course, they have the advantage of building very tall 'houses'.
well,
if google goes vertical in the tablet market (I think they will -> there is no other way), samsung will be left out in the cold (I also believe microsoft will go vertical in the tablet space as well). There is really no way to make money in tablets without having an appstore, and media to make addl money off of. We already see that amazon is going in this direction.
yea,
I am in this area, I'll see if I can a look at the manuscript, thanks!
I am looking to make constructs on FPGA's that the EDA crapware won't allow, at least with some elegance / automation. Ive been forced to make my undergrad student use xilinx's "fpga editor" - although it works for what we are doing, it is tedious and not very repeatable or scriptable.
why are these companies so fearful to release the specs on the bitstream / architecture?
trying to find morality in war is quite futile: so to say that dropping bombs on "innocents" is bad, sorta seems like saying dropping bombs on non-"innocents" is ok. none of it is good. but that is why we call it war and avoid it at all costs (if we can).
unfortunately, _some_ people won't listen, and they need the motivation of bombs to get them to the negotiating table.
you could also modify the bitstream and release malicious code into a STB.
another thing: sometimes these STB's are more 'trusted' because the engrs assume that the bitstream/designs in the FPGA are secure. Its a great place to put a trojan, monitor packets, etc.
this is not a good thing. It will mean more expensive hardware in the future.
ok, so obscurity isn't working... time for something different. Use a username that is a slight modification of a VERY common person.
bradpitt, obama, billgates, sjobs, stevejobs, ibm, microsoft, etc etc.
then, when some marketing puke googles that : the s/n ratio blows their little analytics apart.
--
john smith
the last time my gps didn't work was in the middle east - it is by far not reliable enough to be used without an INS.
And is suscetiple to 'interference' from many sources.
Instrument flying is based on rules determined by blood: almost every rule is because someone bit the big one for relying on something they should not have.
Basics of flight: pitch/roll heading/altitude, everything else (except the landing) is gravy
Back on topic: True headings ARE used, I believe above the artic circle - as mag gets confused up close to the poles.
j
better yet: get some HF and etch it away, nitric acid to rid yourself of the encapsulate, then scratch the surface of the die (and the bondwires) to disable the chip. oh hell, just grind the thing up, I am pretty sure they blend.
j
don't worry about secure erasing a flash drive - it can't be done.
why? - many reasons, but the chief one is the wear limiting circuitry that swaps out huge sections of the memory, and the fact that the reported memory of an ssd is much smaller than the actual memory. - thus: you don't have access to all of the memory cells so that you can erase them.
some researcher somewhere was able to pull "erased" data off of an SSD, by removing the chips and checking them with a logic analyzer / test bench - an actually easy thing to do (easier than platter manipulation i bet)
j
yea,
well - sort of: these advanced approach ( & autoland) systems are really complex: some planes have 2 or even 3 autopilot systems - they check with each other for agreement, etc. the air data sensors and all kinds of other systems must agree - anyway: I say all that to say that we must fly the approaches - it can be done in the cockpit or the sim, if there are faults - you report them (radar altimeters disagree etc) and the system is checked.
a little history: Aviation had been thoroughly plagued by equipment failure in the past - many more pilots died due to bad design, manufacture, or assembly than to human error (or even combat).
so inherently, we don't trust computers, engines, fuel ourselves, or controllers. That is why we have 2 computers, 2-4 engines, many fuel bladders, co-pilots, and radar.
anywho - this issue of co-pilots goes deep into human factors: a wide and vast field - some folks who are into HCI may have studied some parts of this - and from this we have learned that having at least two in the cockpit is way better than one (three is even better). Just please understand that a co-pilot is NOT a backup pilot in case the pilot has heart failure. he/she is an integral part of a team that keeps the plane going where its intended to be.
j
Commercial airlines are already required by law to do a certain percentage of their landings automatically. They just don't tell you...
what you mean is that pilots must remain proficient in Cat 3 and 3a approaches - so they must maintain currency with those procedures by performing one every once in a while. This currency can also be accomplished in a simulator.
Trying to do one of these without a co-pilot is ill-advised (1 set eyes on instruments another looking out for the runway environment) - don't forget about radio calls, communication with home base / fuel management / emergencies / etc.
I flew very complex, very large planes - and I can tell you that there is a real good reason for at least two in the cockpit.
j
I think this is a good topic: there is alot of misconception with attempting to do this, and a huge variance in what is being attempted by all:
The big takeaway: serving high-bitrate video (home movies/ blu ray) is REALLY HARD. I haven't seen a media player do this a rock solidly as a BD player / or HTPC. It would be hard to imagine a home network handling more than one high bit rate movie (40Mbps) being streamed at one time.
The other big takeaway: there are tons of folks who do this who do not care at all about video quality: transcoding is a nasty way of sharing video - but then again it may suit a multi tv household.
The other other big takeaway: you'll see lots of people saying that this device and that device can do 1080p. That is actually nonsensical: as any faithful handbrake user knows: its high bit-rates that bring quality: not resolution.
that being said, this is a quest to balance quality and convenience - and I wish it was discussed more thoroughly than what we see right now. ie: max bitrates of media players on the market right now / suitable off the shelf HTPC's, etc.
Slashdot Mirror
UAV's (drones) have a horrible saftey record, true, but it is because the automation on them is quite poor. At this point, very few of them are "autonomous". The problem is not too much automation, but automotion that is not sufficiently rigorous, testable, nor capable to handle somewhat common contingencies.
I have multiple thousands of hours in heavy aircraft, and I can tell you that 'hand flying more' is not the solution to the problem either.
A well placed, rigorously programmed, redundantly powered backup system that could auto-land when normal power goes out is a much better soln than trying to land a heavy with no electrics (think: at night, in fog, heavy winds, ice, nuns/babies on board, etc).
Rockwell better make sure my retro encabulator is secure from those hackers!
If someone got access to the modial interaction of magneto reluctance and capacitive duractance, we could all be in a world of hurt.
this is great ...
I mean, I was thinking this whole conversation yesterday in my head
gawd help us (elec/cmptr engrs) when the public no longer 'needs' our little trinkets
there is an awful bunch of crap (3D TV, 4K projectors, etc) that really have no REAL use
I just hope we all don't go back to burlap sacks and huts for homes
I really like getting wifi internet in my jacuzzi while watching my 60" flat screen
ok
go back to flaming each other
mod that sh!t up. I mean, we are all just trying to make a living, I don't think anyone really wants world domination anymore - takes too much time. meh, I could be wrong tho.
what about all those Millions of people in other sinking bowls ( sacramento river valley).
New Orleans is not unique, and you can't blame people who live there for being in the path of disaster.
besides, the old city of new orleans is above sea level, its the 'burbs that get flooded (definition of 'burbs in new orleans can get sticky though, but basically, any neighborhoods that existed a loooong time ago are well-tested with flood history)
there are many places [1] [2] in the world below sea level
this picture is a little exaggerated, but shows that the main threat is the mighty mississippi, not the sea. and the army corps of engineers has a divert-the-mississippi spillway upriver that virtually guarantees the river flood threat to mitigated.
ask anyone from new orleans (or others) and they will say that it was engineering that failed the city: intracoastal canals, notably MRGO, created for commerce gave intrusion paths to storm surge from the lake and the gulf. it was those levees that failed and spilled into the city.
from http://en.wikipedia.org/wiki/Drainage_in_New_Orleans we have:
as humans, what makes us special is not just our ability to adapt, but to adapt the environment around us. If we never lived anywhere there was a threat of disaster, I am not sure where you could live(definitely not Texas, or a few other places. And those maps don't even include floods (the most common natural disaster), for that threats see this map of flood hazards for the US.
they do: http://www.youtube.com/watch?v=97DYuAdJtRY http://issuu.com/lizfenuta/docs/amphibious_architectures_thesis http://www.buoyantfoundation.org/
miami, ft lauderdale, houston, galveston, wilmington, lower manhattan, long island, etc.
I could go on and on.
New Orleans didn't invent "living below sea level"
now living in a spillway, the atchafalaya, which is flooded (by man, by God indirectly) fairly often, and is within the path to the Gulf that the mississippi really wants to go (very much so, in fact) - now that is a different story.
those folks know the risk and use buoyant foundations (trailers with styrofoam underneath) to maintain their homes.
they love living out there (it is real nice), and they understand they could lose everything one day.
other parts of the mississippi flood plain include MANY areas up north that flood MUCH more regularly, so lay off of new orleans
exactly, people complain about new orleans, but manhattan is just as vulnerable to sea level rise/surge/flood. of course, they have the advantage of building very tall 'houses'.
yea, we have seen this stuff before:
http://www.coastalcontractor.net/article/189.html
http://www.npr.org/templates/story/story.php?storyId=113513752
http://www.buoyantfoundation.org/
I guess "buoyant house" would be a better description
well, if google goes vertical in the tablet market (I think they will -> there is no other way), samsung will be left out in the cold (I also believe microsoft will go vertical in the tablet space as well). There is really no way to make money in tablets without having an appstore, and media to make addl money off of. We already see that amazon is going in this direction.
yea, I am in this area, I'll see if I can a look at the manuscript, thanks! I am looking to make constructs on FPGA's that the EDA crapware won't allow, at least with some elegance / automation. Ive been forced to make my undergrad student use xilinx's "fpga editor" - although it works for what we are doing, it is tedious and not very repeatable or scriptable. why are these companies so fearful to release the specs on the bitstream / architecture?
and my advisor says slashdot is a waste of time. thanks for the links, they are immensely useful.
trying to find morality in war is quite futile: so to say that dropping bombs on "innocents" is bad, sorta seems like saying dropping bombs on non-"innocents" is ok. none of it is good. but that is why we call it war and avoid it at all costs (if we can). unfortunately, _some_ people won't listen, and they need the motivation of bombs to get them to the negotiating table.
help me out with that link to the fpga bitstream library at Berkeley, its giving me a 404-like response, surely it's not slash dotted.
you could also modify the bitstream and release malicious code into a STB. another thing: sometimes these STB's are more 'trusted' because the engrs assume that the bitstream/designs in the FPGA are secure. Its a great place to put a trojan, monitor packets, etc. this is not a good thing. It will mean more expensive hardware in the future.
Wait...what??? So they(SSD's) lack security, so truecryot reccomends AGAINST encryption? Shouldn't they brcrdccomending the opposite?
ok, so obscurity isn't working... time for something different.
Use a username that is a slight modification of a VERY common person. bradpitt, obama, billgates, sjobs, stevejobs, ibm, microsoft, etc etc.
then, when some marketing puke googles that : the s/n ratio blows their little analytics apart.
-- john smith
the last time my gps didn't work was in the middle east - it is by far not reliable enough to be used without an INS. And is suscetiple to 'interference' from many sources.
Instrument flying is based on rules determined by blood: almost every rule is because someone bit the big one for relying on something they should not have.
Basics of flight: pitch/roll heading/altitude, everything else (except the landing) is gravy
Back on topic: True headings ARE used, I believe above the artic circle - as mag gets confused up close to the poles.
j
sorry that was degrees C not F.
j
better yet: get some HF and etch it away, nitric acid to rid yourself of the encapsulate, then scratch the surface of the die (and the bondwires) to disable the chip. oh hell, just grind the thing up, I am pretty sure they blend.
j
you must have a pretty good oven, melting point of Si is 1414 degrees F. :)
must be a kenmore elite
j
don't worry about secure erasing a flash drive - it can't be done. why? - many reasons, but the chief one is the wear limiting circuitry that swaps out huge sections of the memory, and the fact that the reported memory of an ssd is much smaller than the actual memory. - thus: you don't have access to all of the memory cells so that you can erase them. some researcher somewhere was able to pull "erased" data off of an SSD, by removing the chips and checking them with a logic analyzer / test bench - an actually easy thing to do (easier than platter manipulation i bet) j
yea, well - sort of: these advanced approach ( & autoland) systems are really complex: some planes have 2 or even 3 autopilot systems - they check with each other for agreement, etc. the air data sensors and all kinds of other systems must agree - anyway: I say all that to say that we must fly the approaches - it can be done in the cockpit or the sim, if there are faults - you report them (radar altimeters disagree etc) and the system is checked.
a little history: Aviation had been thoroughly plagued by equipment failure in the past - many more pilots died due to bad design, manufacture, or assembly than to human error (or even combat).
so inherently, we don't trust computers, engines, fuel ourselves, or controllers. That is why we have 2 computers, 2-4 engines, many fuel bladders, co-pilots, and radar.
anywho - this issue of co-pilots goes deep into human factors: a wide and vast field - some folks who are into HCI may have studied some parts of this - and from this we have learned that having at least two in the cockpit is way better than one (three is even better). Just please understand that a co-pilot is NOT a backup pilot in case the pilot has heart failure. he/she is an integral part of a team that keeps the plane going where its intended to be.
j
Commercial airlines are already required by law to do a certain percentage of their landings automatically. They just don't tell you...
what you mean is that pilots must remain proficient in Cat 3 and 3a approaches - so they must maintain currency with those procedures by performing one every once in a while. This currency can also be accomplished in a simulator.
Cat 3 and 3a autoland has been around for a long time. (1965)
http://en.wikipedia.org/wiki/Autoland
Trying to do one of these without a co-pilot is ill-advised (1 set eyes on instruments another looking out for the runway environment) - don't forget about radio calls, communication with home base / fuel management / emergencies / etc. I flew very complex, very large planes - and I can tell you that there is a real good reason for at least two in the cockpit. j
I think this is a good topic: there is alot of misconception with attempting to do this, and a huge variance in what is being attempted by all: The big takeaway: serving high-bitrate video (home movies/ blu ray) is REALLY HARD. I haven't seen a media player do this a rock solidly as a BD player / or HTPC. It would be hard to imagine a home network handling more than one high bit rate movie (40Mbps) being streamed at one time. The other big takeaway: there are tons of folks who do this who do not care at all about video quality: transcoding is a nasty way of sharing video - but then again it may suit a multi tv household. The other other big takeaway: you'll see lots of people saying that this device and that device can do 1080p. That is actually nonsensical: as any faithful handbrake user knows: its high bit-rates that bring quality: not resolution. that being said, this is a quest to balance quality and convenience - and I wish it was discussed more thoroughly than what we see right now. ie: max bitrates of media players on the market right now / suitable off the shelf HTPC's, etc.