Splitting throttle control from steering was a well intentioned move, executed (and verified) very badly.
I believe this kind of separation of the two types of controls used to be fairly common, in the days of old physical levers and such controls.
In any case, after mistakes were made, the next problem was that no one was able to come quickly at correct "state awareness", and that is down to both bad training, bad procedures or procedure following, AND terrible UI design on the computerized control consoles.
The UI on control station B (let's call it) clearly didn't PROMINENTLY indicate to that operator that they had both the steering control and throttle control.
It also didn't PROMINENTLY show that only one throttle and propellor (of the left and right one) was being adjusted rather than both.
These facts should be visually direct and hit-you-over-the-head prominent on that ui. Bright big colored things in the corresponding shape of the real thing.
Also, the fact that the other console A (that formerly had control) no longer had steering control and had transferred it (to station B) should also have been visually prominent in console A's UI.
And of course a little simulator training for the crew wouldn't have hurt either, but this was also a serious BAD UX problem.
Also, the control logic for transferring steering from console A to console B on the bridge is different from how that transfer works to console C (Aft steering station, at back of boat.)
A => B : station B rudder control always starts at midships rudder regardless of station A rudder setting prior to transfer. In this case, station A had a starboard roughly 4 degrees rudder setting to compensate for drifting in ocean current. That changed to midships rudder on the transfer without anyone realizing it had been different.
A => C : station C rudder control resumes at whatever setting station C already had the last time it was used, regardless of the rudder setting on whatever console transferred to it. In this case, station C was at 30 degrees port rudder when it got control. It's operator didn't realize that for a while.
These differences in the control logic are also probably very bad human-factors design, and would have required large amounts of repeated training to have people get the different transfer procedures straight. UX fail again.
Actually "global warming" and "climate change" are equally accurate, but at different spatial scales. The global waming causes changes to regional climates. Some regions of the planet get net colder (e.g. as Europe will if it loses the gulf stream ocean current, a possible GW effect), some warmer (desert zones will shift north in northern hemisphere and south in southern hemisphere due to the growth in energy and size of atmospheric convective cycles, poles will become warmer faster than average elsewhere), Some wetter (overall level of water vapur in atmosphere has already increased about 5% due to recent global warming and is continuing to increase at around 1% per decade.), and some drier (see desert-shift comment above.) All of these regional changes are driven by the underlying warming of global oceans and atmosphere.
Politically, "climate change" can be used to hoodwink people into thinking all is as usual (not true.) Whereas "global warming" tells it as it is and is the more fundamental, underlying physical change going on, that causes the smaller-scale changes. On average, places will definitely be warming.
Pretty sure only people able to pass a rudimentary-skills test of physics knowledge and math and basic logic competence should be allowed to vote on this.Aside from that qualification they can have whatever political stripe they want.
I mean what percentage of people who don't know what the f*** they're talking about (have low critical thinking and physical-world literacy/numeracy skills) should be allowed to vote on any large and critical physics and engineering related project? If the majority are ignorant (and are just mostly tribal-leader followers), and insist on voting on it anyway, we'd be seriously screwed.
People who say "AI and smart automation will not (net) reduce jobs" are relying on a general denial that certain fundamentals will or can change. They can't imagine fundamental change of almost anything. Either cognitively they can't imagine it, or more likely, emotionally, they can't accept it and so employ irrational psychologically defensive denial.
These are the same people who say things like "solar and wind power can never be more than X percent of our energy mix" where X is usually somewhere around 10 in their estimation, because of this that and the other obstacle.
And on many other issues (self-driving cars are impossible! TV and print magazines will always be the thing!... people en masse only need pagers, not mobile phones. There will be a world market for maybe 5 computers. We are not causing a mass extinction episode right now. The climate always changes so there's no problem now. etc. etc.
Other example: computers won't reduce paper use. (yes they will. It is just taking slightly longer than you can grasp in your conservative imagination.)
Could one do software development and testing while offline, with one of these puppies? e.g. Can I have linux in a VM or use docker containers etc in chromeos?
Absolutely. We need core information infrastructure that is extraterritorial (distributed around the planet, fully encrypted) and the providers of such infrastructure need to not be trying to make money by staying friendly with all the governments of the world. Google would be in a conflict of interest here. We need to avoid this kind og situation by changing the architecture to route around this kind of censorship.
...will be blocked at the border...until we get our own satellite, balloon, and solar-powered-drone internet using hard-to-jam ultra-wide-band communications (or something along those lines.)
That is, until the power of conventional states is reduced to relative insignificance by bottom-up self-organizing blockchain-based economies and operations.
Yeah but you're talking about early-generation AI.
All we have to do is go a little META on the learning algorithms and representations, and give them a general goal like "learn more about everything", "form goals (for beneficial modification of environment, for assistance to human wishes?)", "learn experimentally how to progress toward goals" etc.
Yeah, sorry but less than 1% of people have the mental capacity / processing ability and organization, focus and sticktoitiveness to do that kind of work,.
In i don't know 50,000 years or so, a larger percentage of the population might be suited for that,
But long before then, I assume AIs will be much better at that.
I'm thinking the big problem with a 700 mph vehicle encountering rapidly rising air pressure would be turbulence and vehicle oscillation leading to dynmic loads and vehicle destruction. Some good wind tunnel modelling is needed. Maybe there's a pod shape that can avoid that.
If the vacuum is breached behind the train, it might accelerate first then crash and stop. Because air pressure behind and less pressure in front, for a while.
To prevent this, you might want an automated emergency system to vent in front of a train (increase pressure there) if pressure is increasing at back of it.
Keystone = expansion of co2-emitting fossil fuel infrastructure = brain-dead (or clinically insane, hard to tell which) policy in the 21st century.
Deep geothermal power = avoided fossil fuel consumption and emissions. So most fossil-fuel expansion protesters will be fine with it.
These people who get "triggered" by fossil fuel capacity expansion projects are mostly triggered by the overwhelming stupidity and willful destructiveness of the projects. The maniacs (those continuing to scale up fossil fuel infrastructure) need to be stopped. If you don't believe me, come back and tell me what you think in 25 years, when you're getting most of your food from the mid-west (of Canada).
Oh wait, you can't. Not without looking down to check if virtual ESC button is there and where it is. And not without looking down to see if it actually "pressed" or not.
Touch bar has got to go. And give the ESC key back or lose developers, who will quit you just on the principle of not supporting this design-dumb idea.
Suggestion: Put a touch bar (perhaps vertical) to the right of the touch pad. If you need gimmicky stuff like audio volume slider etc.
I don't have the exact words Musk has used, but I distinctly remember that he said that all Teslas will come equipped with the HARDWARE necessary for fully autonomous self-driving (computer power, sensors), but that the actual functionality would depend on a future software upgrade.
Now you and I, as software-related techies most of us, know that that will have to be one MASSIVELY COMPLEX and not really invented yet by any stretch of the imagination software upgrade, but technically, what he said is not false.
Further to this, I was having a discussion about this sort of thing in the cafeteria a while back. We were talking about how primary/secondary education should look, and I was saying how learning to parrot back specific knowledge was pretty useless, cause the google-web voice could do it better. They should be learning/practicing about learning (and synthesis) itself. How to gather and use available resources whatever the topic. etc. etc.
I opined as how I was pretty handy at google and could become an instant pseudo expert on whatever topic as needed, from fixing my bike's pneumatic disk brakes to the etiology, presentation, and different recommendations for treatments of shoulder ligament partial-thickness tears, to different takes on the phenomenon and meaning of quantum decoherence. etc. etc.
So one colleague now greets me as "Hey, It's instant-pseudo-expert" guy! I guess I had that coming.
The actual specialized knowledge will be available on-line through Google Assistant or whatever that morphs into in 10 years, as well as all the online learning resources like university lecture series and Khan academy, "the great courses" what have you.
A meta-level polymath is one who knows all about philosophy of knowledge, and in a personal way, knows how to use their own mind in a way that is in accord with sound knowledge-gathering practices and knowledge-organizing practices. The meta-polymath, is an expert in the practice of efficient inquiry, learning, categorization, information prioritization etc. Perhaps as a shortcut, they have trained in some very general areas of knowledge such as the most commonly needed mathematics and logic, as well as have become expert in their best natural language. All of these are just to ensure speed and accuracy in learning other more specific fields.
Perhaps they should have one to three areas of specific knowledge that they are passionate about, and bother to go deep and current in, and innovate in, but this is more just to keep in mental practice, so they can apply the same passionate but principled learning rapidly to any new domain.
Slashdot Mirror
Splitting throttle control from steering was a well intentioned move, executed (and verified) very badly.
I believe this kind of separation of the two types of controls used to be fairly common, in the days of old physical levers and such controls.
In any case, after mistakes were made, the next problem was that no one was able to come quickly at correct "state awareness", and that is down to both bad training, bad procedures or procedure following, AND terrible UI design on the computerized control consoles.
The UI on control station B (let's call it) clearly didn't PROMINENTLY indicate to that operator that they had both the steering control and throttle control.
It also didn't PROMINENTLY show that only one throttle and propellor (of the left and right one) was being adjusted rather than both.
These facts should be visually direct and hit-you-over-the-head prominent on that ui. Bright big colored things in the corresponding shape of the real thing.
Also, the fact that the other console A (that formerly had control) no longer had steering control and had transferred it (to station B) should also have been visually prominent in console A's UI.
And of course a little simulator training for the crew wouldn't have hurt either, but this was also a serious BAD UX problem.
Also, the control logic for transferring steering from console A to console B on the bridge is different from how that transfer works to console C (Aft steering station, at back of boat.)
A => B : station B rudder control always starts at midships rudder regardless of station A rudder setting prior to transfer.
In this case, station A had a starboard roughly 4 degrees rudder setting to compensate for drifting in ocean current. That changed to midships rudder on the transfer without anyone realizing it had been different.
A => C : station C rudder control resumes at whatever setting station C already had the last time it was used, regardless of the rudder setting on whatever console transferred to it.
In this case, station C was at 30 degrees port rudder when it got control. It's operator didn't realize that for a while.
These differences in the control logic are also probably very bad human-factors design, and would have required large amounts of repeated training to have people get the different transfer procedures straight. UX fail again.
Actually "global warming" and "climate change" are equally accurate, but at different spatial scales. The global waming causes changes to regional climates.
Some regions of the planet get net colder (e.g. as Europe will if it loses the gulf stream ocean current, a possible GW effect),
some warmer (desert zones will shift north in northern hemisphere and south in southern hemisphere due to the growth in energy and size of atmospheric convective cycles, poles will become warmer faster than average elsewhere),
Some wetter (overall level of water vapur in atmosphere has already increased about 5% due to recent global warming and is continuing to increase at around 1% per decade.),
and some drier (see desert-shift comment above.)
All of these regional changes are driven by the underlying warming of global oceans and atmosphere.
Politically, "climate change" can be used to hoodwink people into thinking all is as usual (not true.)
Whereas "global warming" tells it as it is and is the more fundamental, underlying physical change going on, that causes the smaller-scale changes. On average, places will definitely be warming.
Pretty sure only people able to pass a rudimentary-skills test of physics knowledge and math and basic logic competence should be allowed to vote on this.Aside from that qualification they can have whatever political stripe they want.
I mean what percentage of people who don't know what the f*** they're talking about (have low critical thinking and physical-world literacy/numeracy skills) should be allowed to vote on any large and critical physics and engineering related project? If the majority are ignorant (and are just mostly tribal-leader followers), and insist on voting on it anyway, we'd be seriously screwed.
People who say "AI and smart automation will not (net) reduce jobs" are relying on a general denial that certain fundamentals will or can change.
They can't imagine fundamental change of almost anything. Either cognitively they can't imagine it, or more likely, emotionally, they can't accept it and so employ irrational psychologically defensive denial.
These are the same people who say things like "solar and wind power can never be more than X percent of our energy mix" where X is usually somewhere around 10 in their estimation, because of this that and the other obstacle.
And on many other issues (self-driving cars are impossible! TV and print magazines will always be the thing! ... people en masse only need pagers, not mobile phones. There will be a world market for maybe 5 computers. We are not causing a mass extinction episode right now. The climate always changes so there's no problem now. etc. etc.
Other example: computers won't reduce paper use. (yes they will. It is just taking slightly longer than you can grasp in your conservative imagination.)
Just curious.
Could one do software development and testing while offline, with one of these puppies? e.g. Can I have linux in a VM or use docker containers etc in chromeos?
Absolutely. We need core information infrastructure that is extraterritorial (distributed around the planet, fully encrypted) and the providers of such infrastructure need to not be trying to make money by staying friendly with all the governments of the world. Google would be in a conflict of interest here. We need to avoid this kind og situation by changing the architecture to route around this kind of censorship.
Double it and add 30.
FTW Also roughly converts Celsius to Fahrenheit.
Get one really intelligent, experienced, accomplished developer who also has really good leadership and communication skills.
Give them two to three assistants. Best if you let that person pick their assistants.
Plan what you're going to try to do.
Identify the shape of it (context, architecture, UX/UI, high level requirements and/or stories.)
Identify biggest risks and biggest value opportunity directions.
Let team go to it.
Assess.
Revisit.
...will be blocked at the border ...until we get our own satellite, balloon, and solar-powered-drone internet using hard-to-jam ultra-wide-band communications
(or something along those lines.)
That is, until the power of conventional states is reduced to relative insignificance by bottom-up self-organizing blockchain-based economies and operations.
Providing comedy relief to AIs observing our quirky behaviour.
Yeah but you're talking about early-generation AI.
All we have to do is go a little META on the learning algorithms and representations, and give them a general goal like "learn more about everything", "form goals (for beneficial modification of environment, for assistance to human wishes?)", "learn experimentally how to progress toward goals" etc.
http://www.sciencemag.org/news...
Yeah, sorry but less than 1% of people have the mental capacity / processing ability and organization, focus and sticktoitiveness to do that kind of work,.
In i don't know 50,000 years or so, a larger percentage of the population might be suited for that,
But long before then, I assume AIs will be much better at that.
I'm thinking the big problem with a 700 mph vehicle encountering rapidly rising air pressure would be turbulence and vehicle oscillation leading to dynmic loads and vehicle destruction. Some good wind tunnel modelling is needed. Maybe there's a pod shape that can avoid that.
If the vacuum is breached behind the train, it might accelerate first then crash and stop.
Because air pressure behind and less pressure in front, for a while.
To prevent this, you might want an automated emergency system to vent in front of a train (increase pressure there) if pressure is increasing at back of it.
I have no idea, but the US shouldn't be let back in without paying a special "stupid" tax - for all the toxic disinformation they spew on this topic.
Keystone = expansion of co2-emitting fossil fuel infrastructure = brain-dead (or clinically insane, hard to tell which) policy in the 21st century.
Deep geothermal power = avoided fossil fuel consumption and emissions. So most fossil-fuel expansion protesters will be fine with it.
These people who get "triggered" by fossil fuel capacity expansion projects are mostly triggered by the overwhelming stupidity and willful destructiveness of the projects. The maniacs (those continuing to scale up fossil fuel infrastructure) need to be stopped. If you don't believe me, come back and tell me what you think in 25 years, when you're getting most of your food from the mid-west (of Canada).
Oh wait, you can't. Not without looking down to check if virtual ESC button is there and where it is. And not without looking down to see if it actually "pressed" or not.
Touch bar has got to go. And give the ESC key back or lose developers, who will quit you just on the principle of not supporting this design-dumb idea.
Suggestion: Put a touch bar (perhaps vertical) to the right of the touch pad. If you need gimmicky stuff like audio volume slider etc.
No, the comment got modded down because it's mind-suckingly vacuous, offensive, and off-topic.
"No known machine learning technique know how to deal with a completely new and unexpected situation."
Neither do humans, at least in car-operation time.
Try turning your head upside down and looking at the world. Hard to interpret, isn't it.
artificial deep neural nets know how to generalize too. What's your point?
I don't have the exact words Musk has used, but I distinctly remember that he said that all Teslas will come equipped with the HARDWARE necessary for fully autonomous self-driving (computer power, sensors), but that the actual functionality would depend on a future software upgrade.
Now you and I, as software-related techies most of us, know that that will have to be one MASSIVELY COMPLEX and not really invented yet by any stretch of the imagination software upgrade, but technically, what he said is not false.
Further to this, I was having a discussion about this sort of thing in the cafeteria a while back. We were talking about how primary/secondary education should look, and I was saying how learning to parrot back specific knowledge was pretty useless, cause the google-web voice could do it better. They should be learning/practicing about learning (and synthesis) itself. How to gather and use available resources whatever the topic. etc. etc.
I opined as how I was pretty handy at google and could become an instant pseudo expert on whatever topic as needed, from fixing my bike's pneumatic disk brakes to the etiology, presentation, and different recommendations for treatments of shoulder ligament partial-thickness tears, to different takes on the phenomenon and meaning of quantum decoherence. etc. etc.
So one colleague now greets me as "Hey, It's instant-pseudo-expert" guy! I guess I had that coming.
The actual specialized knowledge will be available on-line through Google Assistant or whatever that morphs into in 10 years, as well as all the online learning resources like university lecture series and Khan academy, "the great courses" what have you.
A meta-level polymath is one who knows all about philosophy of knowledge, and in a personal way, knows how to use their own mind in a way that is in accord with sound knowledge-gathering practices and knowledge-organizing practices.
The meta-polymath, is an expert in the practice of efficient inquiry, learning, categorization, information prioritization etc.
Perhaps as a shortcut, they have trained in some very general areas of knowledge such as the most commonly needed mathematics and logic, as well as have become expert in their best natural language. All of these are just to ensure speed and accuracy in learning other more specific fields.
Perhaps they should have one to three areas of specific knowledge that they are passionate about, and bother to go deep and current in, and innovate in, but this is more just to keep in mental practice, so they can apply the same passionate but principled learning rapidly to any new domain.