Sadly this is true, but it shouldn't be. Technical people should have the professionalism to analyse requirements and check that the requirements fit the purpose.
I have a friend who bids government contracts (highways,schools,sewage plants,etc). He says that there is no advantage to fix the contract before the bid because then all the other bidders get those same cost savings. Also, you also can't have multiple people bidding and making suggestions on what to change as then you have no way of comparing the resulting bids. Likewise, after the bid, you can tell them how to fix it but then you're fighting an uphill battle because you're basically trying to change the contract at that point. So it isn't really about professionalism and taking the higher ground but more an effect of the entire bidding process.
Trolls are a different matter; they only do it for the lulz. Their whole purpose is to create discord. It's a pointless, unproductive waste of time, and the fact that people get jollies out of deliberately aggravating other people bespeaks of a certain level of sociopathy.
I'm not sure you can make any generalization about trolls. I think it would be a more interesting study if they attempted to study why trolls exists. My guess is that alot of the trolls are infrequent. They've had a bad day, drunk, etc.. There was even a slashdot article recently about someone being shocked that their vandalism they did when they were drunk in college was still there 10 years later. There are also people who are using it to vent some non-politically correct opinion that they are unable to talk about in real life and probably a few who are out to watch the world burn but my guess is that the latter category is actually the minority not the majority. On slashdot, for instance, what percentage of the AC are probably people with accounts that just click the "post anonymously" button because they are stating a controversial position or otherwise don't want their nickname associated with that one particular comment.
Outside matte still isn't the greatest in direct sunlight but I found that the glossy was even hard to use sometimes while inside if I was too close to a window.
Matte definitely helps. I made the mistake on my last laptop purchase of accepting a "free upgrade" to a glossy screen without realizing what I was doing. I hated it. All my other laptops before and since have been matte finish and they do decently ok with glare. A more specialized screen would probably be better if outside visibility was your primary objective but if you have additional selection criterias then requiring a specialized screen really narrows your selection. I find that for my purposes a glossy screen is completely unusable while a matte screen has acceptable performance in sunlight and gives me a much larger selection of laptops to chose from at more reasonable prices than a specialized screen would.
Not only are the difficult problems more fun but how will you ever advance if you are only ever given easy problems. Yes, someone might make more mistakes when they are doing a harder problem but that is how a person learns. Taking away the hard problems isn't the answer but maybe requiring additional peer review for hard problems but you don't need an invasive brain monitor to find this out. Just ask the programmer. As a programmer I can easily tell you after it's done which parts were easy, which parts were hard and which parts are more likely to have bugs.
And the leagues banned this type of ownership (corporate, both public and private) to keep it from happening again. A team has to be owned by a maximum of 32 people (not corporations), one of which must own at least 30% of the team.
I would love to hear their "politically correct" reason for doing that. That basically means that they intentionally made a rule that made it a rich person only game. I'm not sure what the non politically correct reasoning would be either. This just reeks of cronyism. I'm glad I'm not a sports fan and I think everyone else would be better off if they also stopped contributing to professional sports. Watch college sports or better yet join a local league and play yourself. Stop contributing to the corruption that is professional sports.
I wouldn't consider a combine, a washing machine, or turbotax a "business method patent". Those all automate tedious tasks. If you were the only one who had a washing machine or tax software then you could charge the same as everyone else yet do work multiple times faster. Historically this has been tried many times. Some times the person takes the secret to their grave and noone ever discovers it but most of the time the competitive advantage doesn't last too long as someone else figures out a way to copy it.
I agree with this. 3-5 is probably the sweet spot. Any more and developers are going to start slacking and not test all the browsers. Unfortunately we have 3-5 browsers and then a dozen different versions of each. This move by microsoft is a move in the right direction. Web developers shouldn't be expected to support 6 different versions of IE and 6 different versions of firefox. After 3 years you should probably get a security warning every time you open your browser telling you that your web browser is obsolete and making you click an "i understand that the web will be broken if I continue" button before continuing.
I can't count the number of times I've seen companies with scripts or apps that perform some simple operation, but it only saves a few minutes each day. Yet at some point something with the automation breaks or needs to be changed, but the original developers are long gone, and now some other developer has to investigate.
This poor developer will end up needing hours, days, weeks, or even months in some cases in order to find out where the fuck the script or app is running, where the hell the most recent version of the source code is, how to get it running on a development system, and how it works, all before being able to make the fix or the change. Then it takes time to fix it or make the change, plus some time for testing, and then it needs to be redeployed, and finally it needs to be monitored for some time.
So the automation saved maybe a few dollars a day. Yet just a single fix or change to the automation can end up costing hundreds, thousands, or even tens of thousands of dollars once all is said and done. Merely one small fix or change can literally wipe out any cost savings that the automation has ever brought in the past, and then wipe it out for the next few years!
If it's only saving a few minutes per day then when/if it breaks then you scrap it. Why would you spend "days, weeks, or even months" fixing a script that is only saving you a few minutes per day?
This xkcd comic tells you when to automate: http://xkcd.com/1205/ but it applies equally to how long you can afford to spend fixing a script that breaks. This is across 5 years so IMHO you should probably cut the number in half so that 1) you are actually coming out ahead and 2) you have some room for updating/repairing the script when it breaks.
No. That would be wrong. A worker should maximize efficiency by discovering the best way to achieve maximum pay with minimal work. That is what economists say the company should be doing and since companies are now people and workers are people that's what workers should do. In fact doing it any other way flies in the face of the "Free Market" and therefore maximizing efficiency is both an ethical and moral imperative.
This "free market" theory of yours works for a while until someone else discovers that they can automate the task you've already automated and you're out of a job. That's how the "free market" really works. If you were the only farmer with a combine then you would make a ton of money but eventually everyone else gets one too and your competitive advantage goes away. If you really do come up with a "time-saving" device, the best way to get rich with it is to either hide it so noone knows about it or license it so that every pays you for it.
While you're at it, we should go to a mesh network. The internet was originally conceived tob e able to withstand a nuclear attack by routing around damage and "finding" the path from point A to point B by whatever path possible. Unfortunately we've discovered that it's faster to have internet backbones than it is to have to have 50 hops to get to your destination. Encryption might help a little but what we really need to do is figure out a way to have a more peer to peer system so there aren't bottlenecks where everyone's traffic automatically has to go through and can be tracked.
Don't worry about CO2, the plants need it to live, the more there is, the more they grow.
I've heard this argument before and I know plants need co2 but is co2 really the bottleneck and does increasing co2 cause plants to really grow faster to compensate? If co2 is the bottleneck and an increase in co2 causes plants to respond in step and keep co2 stable them that's fine but that doesn't appear to be what is happening. Co2 levels appear to be increasing so obviously this feedback loop is either not working or not working fast enough. If a 5% increase in co2 causes plants to use 1% more co2 then we still have an increase of 4% so yes plants might help a little but they aren't really a solution.
Any difference looks a lot smaller than the markup I've ended up paying for things like going through an energy co-op instead of straight from the generating company.
Those numbers are almost meaningless. The nuclear numbers for the most part don't include the cost of cleanup operations like what happened in Japan or Chernobyl. They might include a little bit paid to the government for disaster recovery but that would quickly get used up in a real disaster. Likewise coal doesn't include environmental damage and oil doesn't include all the military needed to keep oil stable. Even solar and wind have some negative affects. We do need to talk about cost but we need to talk about ALL the costs not just the operating costs but all the externalized costs as well.
Indeed. I'm curious why is not "closed source, with a strong industry support" an option?
Because both "open source" and "strong industry support" when put together like that pretty much means that they don't want to get stuck holding the bag if the company goes out of business. With "strong industry support" the odds of a company going out of business is minimized and with "open source" even if it does go out of business then you can still continue to use the software indefinitely while you look for a replacement.
Black swan theory would disagree with your first statement. If that one creature was freakishly significant it could change things dramatically. Granted some (most / all?) evolutionary steps and system changes will occur anyway just later, but cleanly predicting the impact isn't trivial.
Except that it's rarely just one creature. It a group of similiar creatures and each generation is only slightly different than the previous one so any one individual in that group is not very important to the group's survival. That being said I read somewhere that looking at genetics it appears that at one point the human race dropped down to under 1000 individuals. At that low number then a significant illness could potentially wipe the whole species out which obviously would have changed history.
I think this is exactly what is going on. It's the reason that some people who aren't suppose to succeed in college still do. Over the years our brains have come up with hundreds of survival strategies and hundreds of different types of intelligence. It doesn't really matter which one we use as long as we get to the next generation. If we make it to the next generation then we select for whatever intelligence got us there.
Only if you redefine the "problem" to be "find people like these people".
But that's exactly what they're doing. The only way they are going to be able to "see" intelligence with a brainwave scan is to look at people that they "think" are intelligent and compare other people to that group. Whether that means more brain activity, less brain activity, certain areas active, etc... that's the best they can do as there is no truly objective measure of intelligence. Intelligence is an abstract concept that we attempt to judge based on tests and situations and not only varies from person to person but varies from situation to situation. One person might be really good at crossword puzzles while another person might be really good at manipulation. Who is more intelligent?
Now find me people who: a. will agree with me b. will agree on who the scapegoat is for when it fails c. will not argue with me d. we will call those people the "smartest" ones
If you pick your "benchmark group" well enough and find people with similiar brainwaves/traits then this still solves their problem nicely. You might not have actually picked the "smartest" people but you picked the people that are most likely to do what you want and succeed where you want them to succeed so I don't see this as being a problem if you can really predict future performance. It might actually be easier to detect "people who are good at following orders" than it is to detect "people who are smart"
There are dozens of attributes that I missed and many overlap (like resourcefulness and certain types of intelligence). I wasn't trying to be all inclusive. I think you could probably come up with 50 or more tests of random traits and attributes. Many like if you tested for eye color or hair color probably won't have any correlation at all. The point is to have enough tests that then you can look at the good candidates and see where they all clump then you could eliminate all the useless tests and only keep the half dozen tests that were shown to be good at predicting. Even if you wanted to do a brain scan to eliminate the need for a dozen different tests, knowing which attributes were most important might help you narrow down the regions of the brain to focus on. Fear for instance is know to be associated with a certain area of the brain so if fearlessness was found to be an important trait then you could possibly focus on that area of the brain.
Obviously intelligence varies from person to person and we have tests like IQ tests that can measure this but IQ tests are not super good at measuring people who are successful at accomplishing tasks because it takes more than raw intelligence. Things like willpower, dedication, creativity, work ethic, etc... all play into whether someone is successful at accomplishing tasks. I don't see how a brain scanner is going to accomplish this or how it would be any better than existing testing methods. If I wanted to know this I would be more inclined to give a group of people a ton of different types of tests and then watch their career and decide which of the tests more closely correlated with what I was seeking then I could narrow it down to a combination of traits for instance maybe the results would be high IQ, high creativity, and high level of willpower or some other combination of 3 or 4 attributes then you could test for only those 3-4 attributes instead of dozen of attributes. If you didn't want to wait, you could instead give the same battery of tests to the people in your company that you considered most successful and see if there are any patterns.
What the car needs to do is find the nearest exit, pull off, and then pull onto the shoulder.
Then once stopped, automatically deploy the artificial arm and slap the driver firmly about the head and shoulders.
I was thinking something similiar but I was thinking it should lock all the doors and wait for the police to arrive. I'm pretty sure leaving your seat when the car is on glorified cruise control is grounds for reckless endangerment which in the US is just as bad if not worse than a DUI/DWI.
That probably is one of the tests they perform NOW but that's kindof the point of the OP that car manufacturers are held to insanely high reliability and testing metrics compared to the average computer software. The only other industry that is probably anywhere close is implantable medical devices or space missions.
I'm not saying that they shouldn't be. Automobile accidents are a leading cause of death but it does mean you better know what you're doing before attempting to compete in that space.
It's getting that last 1% right that will take the most time.
That last 1% might not be as big of a hurdle depending on how we define it. An automated car doesn't need to be 99.9% safe on all roads everywhere. If an automated car was 99.9% safe on interstates and didn't engage on non-interstate roads then it would still be extremely useful. To me this should be the first short term goal. Make an automated car that only engages above 30mph on interstates and knows how to safely navigate and/or stop if it encounters a problem. This would be highly useful for semis and anyone travelling long distances which is where you want to use it the most.
But this "almost automated" crap is not only extremely stupid and dangerous, it threatens to set the whole industry back 10 years the first time one of these "automated" cars kills someone.
Slashdot Mirror
Sadly this is true, but it shouldn't be. Technical people should have the professionalism to analyse requirements and check that the requirements fit the purpose.
I have a friend who bids government contracts (highways,schools,sewage plants,etc).
He says that there is no advantage to fix the contract before the bid because then all the other bidders get those same cost savings.
Also, you also can't have multiple people bidding and making suggestions on what to change as then you have no way of comparing the resulting bids.
Likewise, after the bid, you can tell them how to fix it but then you're fighting an uphill battle because you're basically trying to change the contract at that point.
So it isn't really about professionalism and taking the higher ground but more an effect of the entire bidding process.
Trolls are a different matter; they only do it for the lulz. Their whole purpose is to create discord. It's a pointless, unproductive waste of time, and the fact that people get jollies out of deliberately aggravating other people bespeaks of a certain level of sociopathy.
I'm not sure you can make any generalization about trolls. I think it would be a more interesting study if they attempted to study
why trolls exists. My guess is that alot of the trolls are infrequent. They've had a bad day, drunk, etc.. There was even
a slashdot article recently about someone being shocked that their vandalism they did when they were drunk in college was
still there 10 years later. There are also people who are using it to vent some non-politically correct opinion that they are
unable to talk about in real life and probably a few who are out to watch the world burn but my guess is that the latter category
is actually the minority not the majority. On slashdot, for instance, what percentage of the AC are probably people with
accounts that just click the "post anonymously" button because they are stating a controversial position or otherwise don't
want their nickname associated with that one particular comment.
Outside matte still isn't the greatest in direct sunlight but I found that the glossy was even
hard to use sometimes while inside if I was too close to a window.
Matte definitely helps. I made the mistake on my last laptop purchase of accepting a "free upgrade" to a glossy screen
without realizing what I was doing. I hated it. All my other laptops before and since have been matte finish and they
do decently ok with glare. A more specialized screen would probably be better if outside visibility was your primary
objective but if you have additional selection criterias then requiring a specialized screen really narrows your selection.
I find that for my purposes a glossy screen is completely unusable while a matte screen has acceptable performance in
sunlight and gives me a much larger selection of laptops to chose from at more reasonable prices than a specialized screen
would.
Not only are the difficult problems more fun but how will you ever advance if you are only ever given easy problems.
Yes, someone might make more mistakes when they are doing a harder problem but that is how a person learns.
Taking away the hard problems isn't the answer but maybe requiring additional peer review for hard problems but
you don't need an invasive brain monitor to find this out. Just ask the programmer. As a programmer I can easily
tell you after it's done which parts were easy, which parts were hard and which parts are more likely to have bugs.
And the leagues banned this type of ownership (corporate, both public and private) to keep it from happening again. A team has to be owned by a maximum of 32 people (not corporations), one of which must own at least 30% of the team.
I would love to hear their "politically correct" reason for doing that. That basically means that they intentionally made a rule that
made it a rich person only game. I'm not sure what the non politically correct reasoning would be either. This just reeks of cronyism.
I'm glad I'm not a sports fan and I think everyone else would be better off if they also stopped contributing to professional sports.
Watch college sports or better yet join a local league and play yourself. Stop contributing to the corruption that is professional sports.
This is a good example of Poe's Law: http://en.wikipedia.org/wiki/P...
I wouldn't consider a combine, a washing machine, or turbotax a "business method patent".
Those all automate tedious tasks. If you were the only one who had a washing machine
or tax software then you could charge the same as everyone else yet do work multiple
times faster. Historically this has been tried many times. Some times the person takes
the secret to their grave and noone ever discovers it but most of the time the competitive
advantage doesn't last too long as someone else figures out a way to copy it.
I agree with this. 3-5 is probably the sweet spot. Any more and developers are going to start slacking and not test all the
browsers. Unfortunately we have 3-5 browsers and then a dozen different versions of each. This move by microsoft
is a move in the right direction. Web developers shouldn't be expected to support 6 different versions of IE and 6 different
versions of firefox. After 3 years you should probably get a security warning every time you open your browser
telling you that your web browser is obsolete and making you click an "i understand that the web will be broken if I continue"
button before continuing.
I can't count the number of times I've seen companies with scripts or apps that perform some simple operation, but it only saves a few minutes each day. Yet at some point something with the automation breaks or needs to be changed, but the original developers are long gone, and now some other developer has to investigate.
This poor developer will end up needing hours, days, weeks, or even months in some cases in order to find out where the fuck the script or app is running, where the hell the most recent version of the source code is, how to get it running on a development system, and how it works, all before being able to make the fix or the change. Then it takes time to fix it or make the change, plus some time for testing, and then it needs to be redeployed, and finally it needs to be monitored for some time.
So the automation saved maybe a few dollars a day. Yet just a single fix or change to the automation can end up costing hundreds, thousands, or even tens of thousands of dollars once all is said and done. Merely one small fix or change can literally wipe out any cost savings that the automation has ever brought in the past, and then wipe it out for the next few years!
If it's only saving a few minutes per day then when/if it breaks then you scrap it. Why would you spend "days, weeks, or even months" fixing a script
that is only saving you a few minutes per day?
This xkcd comic tells you when to automate: http://xkcd.com/1205/ but it applies equally to how long you can afford to spend fixing a script that breaks.
This is across 5 years so IMHO you should probably cut the number in half so that 1) you are actually coming out ahead and 2) you have some room
for updating/repairing the script when it breaks.
No. That would be wrong. A worker should maximize efficiency by discovering the best way to achieve maximum pay with minimal work. That is what economists say the company should be doing and since companies are now people and workers are people that's what workers should do. In fact doing it any other way flies in the face of the "Free Market" and therefore maximizing efficiency is both an ethical and moral imperative.
This "free market" theory of yours works for a while until someone else discovers that they can automate the task
you've already automated and you're out of a job. That's how the "free market" really works. If you were the only
farmer with a combine then you would make a ton of money but eventually everyone else gets one too and your
competitive advantage goes away. If you really do come up with a "time-saving" device, the best way to get rich
with it is to either hide it so noone knows about it or license it so that every pays you for it.
While you're at it, we should go to a mesh network. The internet was originally conceived tob e able to
withstand a nuclear attack by routing around damage and "finding" the path from point A to point B
by whatever path possible. Unfortunately we've discovered that it's faster to have internet backbones
than it is to have to have 50 hops to get to your destination. Encryption might help a little but what we
really need to do is figure out a way to have a more peer to peer system so there aren't bottlenecks
where everyone's traffic automatically has to go through and can be tracked.
Don't worry about CO2, the plants need it to live, the more there is, the more they grow.
I've heard this argument before and I know plants need co2 but is co2 really the bottleneck and does increasing
co2 cause plants to really grow faster to compensate? If co2 is the bottleneck and an increase in co2 causes
plants to respond in step and keep co2 stable them that's fine but that doesn't appear to be what is happening.
Co2 levels appear to be increasing so obviously this feedback loop is either not working or not working fast enough.
If a 5% increase in co2 causes plants to use 1% more co2 then we still have an increase of 4% so yes plants
might help a little but they aren't really a solution.
eh.
Any difference looks a lot smaller than the markup I've ended up paying for things like going through an energy co-op instead of straight from the generating company.
Those numbers are almost meaningless. The nuclear numbers for the most part don't include the cost of cleanup operations
like what happened in Japan or Chernobyl. They might include a little bit paid to the government for disaster recovery but that
would quickly get used up in a real disaster. Likewise coal doesn't include environmental damage and oil doesn't include all the
military needed to keep oil stable. Even solar and wind have some negative affects. We do need to talk about cost but we
need to talk about ALL the costs not just the operating costs but all the externalized costs as well.
Indeed. I'm curious why is not "closed source, with a strong industry support" an option?
Because both "open source" and "strong industry support" when put together like that pretty
much means that they don't want to get stuck holding the bag if the company goes out of business.
With "strong industry support" the odds of a company going out of business is minimized and
with "open source" even if it does go out of business then you can still continue to use the
software indefinitely while you look for a replacement.
Black swan theory would disagree with your first statement. If that one creature was freakishly significant it could change things dramatically. Granted some (most / all?) evolutionary steps and system changes will occur anyway just later, but cleanly predicting the impact isn't trivial.
Except that it's rarely just one creature. It a group of similiar creatures and each generation is only slightly different than the previous one
so any one individual in that group is not very important to the group's survival.
That being said I read somewhere that looking at genetics it appears that at one point the human race dropped down to under 1000 individuals.
At that low number then a significant illness could potentially wipe the whole species out which obviously would have changed history.
Isn't that the plot of Dune?
Much closer to the plot of the book "Wyrms"
I think this is exactly what is going on. It's the reason that some people who aren't suppose to succeed in college still do.
Over the years our brains have come up with hundreds of survival strategies and hundreds of different types of intelligence.
It doesn't really matter which one we use as long as we get to the next generation. If we make it to the next generation then
we select for whatever intelligence got us there.
Only if you redefine the "problem" to be "find people like these people".
But that's exactly what they're doing. The only way they are going to be able to "see" intelligence with a brainwave scan
is to look at people that they "think" are intelligent and compare other people to that group.
Whether that means more brain activity, less brain activity, certain areas active, etc... that's the best they can do as
there is no truly objective measure of intelligence. Intelligence is an abstract concept that we attempt to judge based
on tests and situations and not only varies from person to person but varies from situation to situation. One person
might be really good at crossword puzzles while another person might be really good at manipulation. Who is more
intelligent?
Now find me people who:
a. will agree with me
b. will agree on who the scapegoat is for when it fails
c. will not argue with me
d. we will call those people the "smartest" ones
If you pick your "benchmark group" well enough and find people with similiar brainwaves/traits then this still solves their problem nicely.
You might not have actually picked the "smartest" people but you picked the people that are most likely to do what you want and
succeed where you want them to succeed so I don't see this as being a problem if you can really predict future performance.
It might actually be easier to detect "people who are good at following orders" than it is to detect "people who are smart"
There are dozens of attributes that I missed and many overlap (like resourcefulness and certain types of intelligence).
I wasn't trying to be all inclusive. I think you could probably come up with 50 or more tests of random traits and attributes.
Many like if you tested for eye color or hair color probably won't have any correlation at all. The point is to have enough
tests that then you can look at the good candidates and see where they all clump then you could eliminate all the useless
tests and only keep the half dozen tests that were shown to be good at predicting. Even if you wanted to do a brain scan
to eliminate the need for a dozen different tests, knowing which attributes were most important might help you narrow down
the regions of the brain to focus on. Fear for instance is know to be associated with a certain area of the brain so if
fearlessness was found to be an important trait then you could possibly focus on that area of the brain.
Obviously intelligence varies from person to person and we have tests like IQ tests that can measure this
but IQ tests are not super good at measuring people who are successful at accomplishing tasks because
it takes more than raw intelligence. Things like willpower, dedication, creativity, work ethic, etc... all play
into whether someone is successful at accomplishing tasks. I don't see how a brain scanner is going to
accomplish this or how it would be any better than existing testing methods. If I wanted to know this I
would be more inclined to give a group of people a ton of different types of tests and then watch their
career and decide which of the tests more closely correlated with what I was seeking then I could narrow
it down to a combination of traits for instance maybe the results would be high IQ, high creativity, and
high level of willpower or some other combination of 3 or 4 attributes then you could test for only those
3-4 attributes instead of dozen of attributes. If you didn't want to wait, you could instead give the same
battery of tests to the people in your company that you considered most successful and see if there are
any patterns.
What the car needs to do is find the nearest exit, pull off, and then pull onto the shoulder.
Then once stopped, automatically deploy the artificial arm and slap the driver firmly about the head and shoulders.
I was thinking something similiar but I was thinking it should lock all the doors and wait for the police to arrive.
I'm pretty sure leaving your seat when the car is on glorified cruise control is grounds for reckless endangerment
which in the US is just as bad if not worse than a DUI/DWI.
That probably is one of the tests they perform NOW but that's kindof the point of the OP that car manufacturers
are held to insanely high reliability and testing metrics compared to the average computer software.
The only other industry that is probably anywhere close is implantable medical devices or space missions.
I'm not saying that they shouldn't be. Automobile accidents are a leading cause of death but it does mean
you better know what you're doing before attempting to compete in that space.
It's getting that last 1% right that will take the most time.
That last 1% might not be as big of a hurdle depending on how we define it. An automated car doesn't need to be 99.9%
safe on all roads everywhere. If an automated car was 99.9% safe on interstates and didn't engage on non-interstate
roads then it would still be extremely useful. To me this should be the first short term goal. Make an automated car
that only engages above 30mph on interstates and knows how to safely navigate and/or stop if it encounters a problem.
This would be highly useful for semis and anyone travelling long distances which is where you want to use it the most.
But this "almost automated" crap is not only extremely stupid and dangerous, it threatens to set the whole industry back 10 years
the first time one of these "automated" cars kills someone.