Considering that after all this testing this is the first jaywalking fatality one would assume the self driving cars have avoided plenty of jaywalkers, probably more than your average human driver manages. Eventually you will have a situation where its mechanically impossible to avoid a crash. The thing with software tho, is that you can take it to limit to what kind of accidents you can avoid, humans however will forever be as lousy drivers as they have always been. There is nothing to it, analyse the accident, improve what can be improved and continue testing.
[quote]will not be using computers in their daily lives for the foreseeable future[/quote]
That is very "I think there is a world market for maybe five computers" type of thinking, Ghana it may be, but these kids have entire lifetime ahead of them. Changes are to be expected over decades. And if Africa does indeed continue to lag behind for the next century, well it's a bit of a catch 22 situations isn't it? Can't progress because majority of population has never even touched a computer, can't get population familiar with modern tech because they are too poor. Knowing your way around a computer is fast becoming nearly as important as basic literacy, what kind of jobs you can't do if you are illiterate or unable to use a computer overlaps nearly completely.
Science is the art of figuring out what is real, in detail. Science can tell you how and why climate is changing, from that you can extrapolate what you need to do in order to make it change differently, but that already falls under geoengineering, not science per se.
ITER will not use latest and greatest, it will use what was available when initial design was locked down. That kind of dictates the torus dimensions so yeah, a newer design could get it done with much smaller reactor because there have indeed been significant advances in superconductors meanwhile.
To predict where it will crash you have to predict exactly when it will crash, as the thing is going 7.5km/s tiny error in timing will result in large error in location. If you can't predict the crash time more accurately than one orbit, that is 90mins, then you essentially have no idea where it will crash. Predictions will get more accurate as time passes, but few weeks before the event the margin of error is just too large. To make it more difficult it's tumbling, meaning that it's ballistic coefficient is changing all the time, couple that with lack of prediction for near space weather and there you have it, all you can do is shrug when someone asks where it will crash.
Machinist workshop or more modernly, machine park and it's been a standard since before industrial revolution really got going. Many businesses do nothing else but manufacture custom parts as ordered, some machine parks are part of a larger business but still take outside orders. If you have entire product design there are companies willing to source the parts and manufacture that too.
Jeez, the guy talks as if he came up with something new, when in fact the entire bloody world has operated like that since forever.
Vocations die out all the time, but work never ends, such has been the case for ever and such will be the case for foreseeable future, automation changes nothing here the same way it has changed nothing in this respect since the start of industrial revolution. Do the pants sanders have to find a new job now, yes, is it a bad thing in any way, no. It just means that there is one more stupid task humanities collective manpower need not be wasted on, in this case sanding pants. There is no limit to human greed, we always want more stuff, now that sanded jeans are less labor intensive that labor can put into use on making some other equally inane but nevertheless desirable product. To put it into market terms, manpower is limited, greed is not, ergo work will never run out because demand is limitless.
If you found a bunch of third world illiterates who never learned to write, with pencil or any other way, would they have reduced fine motor skills compared to people with pretty handwriting?
Sapphire is right out and getting optical glass in such a shape is a costly pain in the rear, I am not at all surprised they just make it out of plastic. It's not a design mistake, it's a sensible engineering compromise.
http://home.roboticlab.eu/_med...
I am thinking lidar is probably not the future in autonomous cars, because of cost and complexity. Human can drive a car using just two eyes, therefore clever enough vision software should be able to manage things with just cameras.
Ireland load of trees does however filter out a bunch of smog, which is kind of the main point. Also more than 1 part in 10000 of carbon in plantmass gets captured in soil. Does it solve the CO2 problem, no, not really. But it does improve air quality, enrich soil and is altogether better use of land than leaving it as wasteland. Go plant a tree, it doesn't fix the world, but it does do quite a bit of good, effort to payoff ratio is pretty damn good.
With a delay of year or two there will be an excess of cheap used GPU-s, overall the alien searchers can get more computed even if it means delay of 1-2 years. For a long term project like that, such a delay doesn't really matter.
As long as you know the time precisely you can calculate your location using observations of solar system objects. It works a lot like marine navigation of olde, development of marine chronometer was a major advancement then too.
Most large payloads have volume issues, massive ones at that, GEO sats are built like origami these days. https://en.wikipedia.org/wiki/...
Falcon is not at a disadvantage here tho, all the heavy lift vehicles have about the same size fairing.
With this FH launch they also demonstrated relight of second stage after 6h coast phase, that pretty much covers direct geostationary insertion, geostationary transfer orbit has period of 10.5h, you need a coast of half that for direct insertion or 5h15min.
Also Falcon fairing is same size as Delta IV Heavy fairing, so no advantage there. There is also that tiny issue of price tag, 90M$ vs 400M$.
Only way ULA gets any of the 5 upcoming Air Force launch bids, is as a charity case.
None of these things would have any use where that rocket is going. It's not going to orbit Mars, it's going to fly by and then orbit Sun in perpetuity.
For small missiles a carrier plane does make a significant difference, small rockets suffer from atmospheric drag losses much more than big ones. That is why you can't scale an orbital rocket down to estes model rocket size to launch post stamps to orbit.
Yeah right, the trick in chemistry is not in mixing stuff together and getting a reaction going, its in separating out the good stuff from the resulting gunk. In a long series of reactions the problem is that you need pure products from previous reactions, but each separation and purification brings a yield penalty with it, so if your processes are not up to snuff then your cumulative yield over the entire process is going to be pretty much zero. And every step of the way you have to make sure you actually got what you wanted, not some crap you didn't. 3D printing reaction vessels doesn't help with any of that.
Most industrial systems are built once and then left well alone to not muck anything up, periodic Windows updates and accompanying mandatory restarts are pretty much a non starter in most cases. If it works, don't touch it, is the wisdom when it comes to industrial systems.
Says the guy who has never created industrial anything. Not that it wouldn't be nice to use Linux, but reality is that Linux might as well not exist on industrial scene, all the software for pretty much anything is Windows only, so good luck working with your Linux based control system.
Slashdot Mirror
"for making false statements when obtaining unemployment benefits and attempted armed robbery" Not really relevant to driving ability
Considering that after all this testing this is the first jaywalking fatality one would assume the self driving cars have avoided plenty of jaywalkers, probably more than your average human driver manages. Eventually you will have a situation where its mechanically impossible to avoid a crash. The thing with software tho, is that you can take it to limit to what kind of accidents you can avoid, humans however will forever be as lousy drivers as they have always been. There is nothing to it, analyse the accident, improve what can be improved and continue testing.
[quote]will not be using computers in their daily lives for the foreseeable future[/quote] That is very "I think there is a world market for maybe five computers" type of thinking, Ghana it may be, but these kids have entire lifetime ahead of them. Changes are to be expected over decades. And if Africa does indeed continue to lag behind for the next century, well it's a bit of a catch 22 situations isn't it? Can't progress because majority of population has never even touched a computer, can't get population familiar with modern tech because they are too poor. Knowing your way around a computer is fast becoming nearly as important as basic literacy, what kind of jobs you can't do if you are illiterate or unable to use a computer overlaps nearly completely.
Science is the art of figuring out what is real, in detail. Science can tell you how and why climate is changing, from that you can extrapolate what you need to do in order to make it change differently, but that already falls under geoengineering, not science per se.
In no way is this a good idea, this social credit thing is some next level 1984 shite if you look a little deeper into it.
ITER will not use latest and greatest, it will use what was available when initial design was locked down. That kind of dictates the torus dimensions so yeah, a newer design could get it done with much smaller reactor because there have indeed been significant advances in superconductors meanwhile.
To predict where it will crash you have to predict exactly when it will crash, as the thing is going 7.5km/s tiny error in timing will result in large error in location. If you can't predict the crash time more accurately than one orbit, that is 90mins, then you essentially have no idea where it will crash. Predictions will get more accurate as time passes, but few weeks before the event the margin of error is just too large. To make it more difficult it's tumbling, meaning that it's ballistic coefficient is changing all the time, couple that with lack of prediction for near space weather and there you have it, all you can do is shrug when someone asks where it will crash.
Machinist workshop or more modernly, machine park and it's been a standard since before industrial revolution really got going. Many businesses do nothing else but manufacture custom parts as ordered, some machine parks are part of a larger business but still take outside orders. If you have entire product design there are companies willing to source the parts and manufacture that too.
Jeez, the guy talks as if he came up with something new, when in fact the entire bloody world has operated like that since forever.
Vocations die out all the time, but work never ends, such has been the case for ever and such will be the case for foreseeable future, automation changes nothing here the same way it has changed nothing in this respect since the start of industrial revolution. Do the pants sanders have to find a new job now, yes, is it a bad thing in any way, no. It just means that there is one more stupid task humanities collective manpower need not be wasted on, in this case sanding pants. There is no limit to human greed, we always want more stuff, now that sanded jeans are less labor intensive that labor can put into use on making some other equally inane but nevertheless desirable product. To put it into market terms, manpower is limited, greed is not, ergo work will never run out because demand is limitless.
If you found a bunch of third world illiterates who never learned to write, with pencil or any other way, would they have reduced fine motor skills compared to people with pretty handwriting?
Shouldn't it be hp(E) in this case or 746W? But yes, horsepower is a bloody stupid unit best forgotten about.
Musk to engineers: "Are you still an effective team?"
Sapphire is right out and getting optical glass in such a shape is a costly pain in the rear, I am not at all surprised they just make it out of plastic. It's not a design mistake, it's a sensible engineering compromise. http://home.roboticlab.eu/_med...
I am thinking lidar is probably not the future in autonomous cars, because of cost and complexity. Human can drive a car using just two eyes, therefore clever enough vision software should be able to manage things with just cameras.
Fertilizer for the forest
Ireland load of trees does however filter out a bunch of smog, which is kind of the main point. Also more than 1 part in 10000 of carbon in plantmass gets captured in soil. Does it solve the CO2 problem, no, not really. But it does improve air quality, enrich soil and is altogether better use of land than leaving it as wasteland. Go plant a tree, it doesn't fix the world, but it does do quite a bit of good, effort to payoff ratio is pretty damn good.
With a delay of year or two there will be an excess of cheap used GPU-s, overall the alien searchers can get more computed even if it means delay of 1-2 years. For a long term project like that, such a delay doesn't really matter.
As long as you know the time precisely you can calculate your location using observations of solar system objects. It works a lot like marine navigation of olde, development of marine chronometer was a major advancement then too.
Most large payloads have volume issues, massive ones at that, GEO sats are built like origami these days.
https://en.wikipedia.org/wiki/...
Falcon is not at a disadvantage here tho, all the heavy lift vehicles have about the same size fairing.
With this FH launch they also demonstrated relight of second stage after 6h coast phase, that pretty much covers direct geostationary insertion, geostationary transfer orbit has period of 10.5h, you need a coast of half that for direct insertion or 5h15min. Also Falcon fairing is same size as Delta IV Heavy fairing, so no advantage there. There is also that tiny issue of price tag, 90M$ vs 400M$. Only way ULA gets any of the 5 upcoming Air Force launch bids, is as a charity case.
None of these things would have any use where that rocket is going. It's not going to orbit Mars, it's going to fly by and then orbit Sun in perpetuity.
For small missiles a carrier plane does make a significant difference, small rockets suffer from atmospheric drag losses much more than big ones. That is why you can't scale an orbital rocket down to estes model rocket size to launch post stamps to orbit.
I wonder how they managed to make that statement with a straight face? https://www.youtube.com/watch?...
Yeah right, the trick in chemistry is not in mixing stuff together and getting a reaction going, its in separating out the good stuff from the resulting gunk. In a long series of reactions the problem is that you need pure products from previous reactions, but each separation and purification brings a yield penalty with it, so if your processes are not up to snuff then your cumulative yield over the entire process is going to be pretty much zero. And every step of the way you have to make sure you actually got what you wanted, not some crap you didn't. 3D printing reaction vessels doesn't help with any of that.
Most industrial systems are built once and then left well alone to not muck anything up, periodic Windows updates and accompanying mandatory restarts are pretty much a non starter in most cases. If it works, don't touch it, is the wisdom when it comes to industrial systems.
Says the guy who has never created industrial anything. Not that it wouldn't be nice to use Linux, but reality is that Linux might as well not exist on industrial scene, all the software for pretty much anything is Windows only, so good luck working with your Linux based control system.