The resignation to complete technologization has been termed by a favourite blogger of mine as the "Borg Complex." It is just faith, pure and simple. Companies have always spent billions of dollars with increasingly marginal return, and it doesn't mean that they will inevitably someday be successful.
Artificial intelligence, for example, is decades, maybe centuries away, from any meaningful advance.
In regards to automation, we have only advanced the capacity to which computers can process information (even this has a cap, when we get transistors down to the atomic level, Moore's Law just stops working). We haven't solved software problems, economies of scale issues, or the growing birth dearth. All of these are issues that will incredibly impede, slow, or make impossible the progress you think is inevitable.
The truth is, all technological innovation is contingent, over-hyped, and always exposed to the human component.
maybe the gate size has shrunk, but how about other materials? GaAs? how about using photons instead of electrons? Moore's law no longer isn't going to be the limit in optics.
why do you consider Moore's law to be a limit, when it comes to AI?
what software problems do you see as the road block (instead of speed bump)
EDIT : in case i wasn't clear, the point wasn't to nitpick and show that you aren't a tech person, but the point is to show tech doesn't hit a limit (nor does it take a quantum leap) as much as the world thinks. Sometimes they take big leaps, sometimes they get slowed to a crawl over what seems trivial but can catch you treading water.
__________________
"With a coach and a player, sometimes there's just so much respect there that it's boils over"
-Taylor Hall
Last edited by Phanuthier; 11-03-2015 at 10:31 AM.
Yeah I'm not arguing any of that at all. I totally agree.
I do think though, that in order for something like this to work, being implemented at the same time, 0-60, into full-automation is the best (albeit most unrealistic) way to go. Not unrealistic because it can't be done, unrealistic because the coordination that would need to be involved would be like nothing ever seen in human history, and we have proven again and again that working together is just too damn difficult to get useful things into practice in the proper way.
i'm pretty sure this is well known, but its already been implemented on real roads already. (not fully automated yet). I believe first instance of this was 2011 or 2012?
__________________
"With a coach and a player, sometimes there's just so much respect there that it's boils over"
-Taylor Hall
maybe the gate size has shrunk, but how about other materials? GaAs? how about using photons instead of electrons? Moore's law no longer isn't going to be the limit in optics.
why do you consider Moore's law to be a limit, when it comes to AI?
what software problems do you see as the road block (instead of speed bump)
As far as I have read (I am not an engineer), you max out at the atomic level, even with advanced materials, and electrons start bleeding all over the place. If you had a quantum computer, then yeah, things keep rolling forward. Only problem is that a quantum computer seems to be a long way off.
Software is still written by human beings. Computers are pretty dumb outside of the tasks written for them by humans. At this point, they seem pretty good simulating, and then executing responses based on the increasing ability to recall, and reorganize lots of information. This is why we all love algorithms so much.
I guess it remains to see if we have picked all of the low-hanging fruit. A lot of the new "profitable" software platforms are just hyped-up sharing platforms. They have always existed in one shape or form, but are now being revived under this not so new idea of the sharing economy, which is really just Silicon Valley VCs figuring out a new way to become rentiers.
As I keep saying, the implementation of technology has a lot of barriers - political and economic. When we start to understand these barriers, we see that even the idea that technology is inevitable is in some way a political way of viewing technology. Look at the graveyard of Google products. Look at something like Google Glass - a few years ago, it was supposed to revolutionize the way we experience reality. Humans rejected it, almost en-masse. Oculus Rift is the same thing, basically an iPad strapped to your face.
I am reeling way off-topic here. There is a lot to say, but to tie it back to autonomous cars, yeah, we have had the technology for like 20 years. Didn't Carnegie Mellon's robotic department drive a Pontiac across the state on autopilot in like 1993? Pretty much robot cars can do anything in a straight line. That is why they have had, for a long time, limited use in industrial settings. A robot truck can drive from point A to point B and back again day and night. But when we get into the real issue of widespread adoption of such technology, the questions become way more complicated.
As far as I have read (I am not an engineer), you max out at the atomic level, even with advanced materials, and electrons start bleeding all over the place. If you had a quantum computer, then yeah, things keep rolling forward. Only problem is that a quantum computer seems to be a long way off.
.
Well I'm an engineer, I work in the field (semiconductors/tech). Again not trying to make you feel bad, but its hard for people not in the industry to understand the grand scope of how much it takes (and thus how much time it takes) to really get the (technology) infrastructure to implement a lot of these things. I have probably a few dozen friends that work for the aforementioned companies and do these projects (note: they dont tell me what they do, its guarded like Fort Knox, but its pretty easy to tell knowing their background and the job descriptions that recruiters send to people in the industry like me to figure out whats going on)'
I wasn't really talking about quantum computing, but look at things like what Intel has been trying with silicon photonics. I don't work for that team ... but to me, Moore's law is not the problem. Lots of other problems, and resolutions are much bigger.
I don't object that it takes a lot of factors to implement, but what I disagree with is that you speak of it like its a small chance of ever taking over or we're going to have a terminator like society or there won't be jobs because everything is automated. Google can probably tell you this, but 2 of the 3 of these are not true.
Quote:
Originally Posted by peter12
Software is still written by human beings. Computers are pretty dumb outside of the tasks written for them by humans. At this point, they seem pretty good simulating, and then executing responses based on the increasing ability to recall, and reorganize lots of information. This is why we all love algorithms so much.
Yes to the first part, no to the 2nd. On your point about economics driving things... money is driving where innovation is going and the money for good software engineers are going into a lot of dumb fields. Besides Uber, Apple and Google which are able to recruit expensive talent.
Quote:
Originally Posted by peter12
I guess it remains to see if we have picked all of the low-hanging fruit. A lot of the new "profitable" software platforms are just hyped-up sharing platforms. They have always existed in one shape or form, but are now being revived under this not so new idea of the sharing economy, which is really just Silicon Valley VCs figuring out a new way to become rentiers.
pretty much. more money is going into stupid startup ideas more then useful ones in society.
most of the limits of technology, however, aren't low hanging fruit. they are more a maze of unknowns that you aren't really sure what you are looking at (yet)
Quote:
Originally Posted by peter12
As I keep saying, the implementation of technology has a lot of barriers - political and economic. When we start to understand these barriers, we see that even the idea that technology is inevitable is in some way a political way of viewing technology. Look at the graveyard of Google products. Look at something like Google Glass - a few years ago, it was supposed to revolutionize the way we experience reality. Humans rejected it, almost en-masse. Oculus Rift is the same thing, basically an iPad strapped to your face. .
those aren't really political or economic reasons for the failure of products. Google can tell you a better reason of the graveyard of Google products, but those aren't the reasons.
political/economic can be a barrier to some, yes. But if there is enough push, barriers can be overcome. and money can do a lot of talking (we see this a lot in American politics). I can't guarantee success as you said there are many factors, but some of these companies (and not some others) likely wouldn't put so much money if they didn't think it had a realistic chance.
Quote:
Originally Posted by peter12
I am reeling way off-topic here. There is a lot to say, but to tie it back to autonomous cars, yeah, we have had the technology for like 20 years. Didn't Carnegie Mellon's robotic department drive a Pontiac across the state on autopilot in like 1993? Pretty much robot cars can do anything in a straight line. That is why they have had, for a long time, limited use in industrial settings. A robot truck can drive from point A to point B and back again day and night. But when we get into the real issue of widespread adoption of such technology, the questions become way more complicated.
not really, no. They had a proof of concept. lots of layers underneath still to develop. I'm a small fish in a big pond, but from some of the customer requests I've seen, they end product lines still don't really know what they need still but they know they don't have it, yet. (hopefully they do get it)
also, as dumb as you think technology is (which I don't disagree with you) tell me again when I see someone reading a book, cutting their toe nails, doing their makeup or smoking a bong while driving on the Deerfoot going 90 with their left turn signal still on.
__________________
"With a coach and a player, sometimes there's just so much respect there that it's boils over"
-Taylor Hall
Last edited by Phanuthier; 11-03-2015 at 11:03 AM.
For those still unaware, I joined Carl and Deena last week to set two new EV and autonomous Cannonball Records, driving from Los Angeles to New York City in 57 hours and 48 minutes.
Most of the time, their 2015 Tesla Model S was driving itself. The experience utterly transformed my views on the future of EVs, autonomous cars and the future of driving.
...
Approximately 96 percent of our drive was on Autopilot. Yes. Ninety-six percent, which translates to approximately 44 hours of driving without human intervention. I did about 1/3rd of the human driving, which translates to about half an hour, most of it in NYC traffic.
yeah, I think the first pass (California? county of San Mateo I think it was?) gave it its first political pass in 2008, the first (but not pilot-less) automated car started rolling around on city streets (in designated counties) in 2010 or something, I think it started to pass around California in 2011 or something... then I stopped paying attention so I'm not really sure what has happened the past few years other then they are picking up steam, there's tons of job postings and there's a mad rush for talent in those fields right now. i kind of wish i was smarter to get in on the gold rush but c'est la vie.
i don't think any are pilot-less, yet....
__________________
"With a coach and a player, sometimes there's just so much respect there that it's boils over"
-Taylor Hall
Last edited by Phanuthier; 11-03-2015 at 11:02 AM.
Well I'm an engineer, I work in the field (semiconductors/tech). Again not trying to make you feel bad, but its hard for people not in the industry to understand the grand scope of how much it takes (and thus how much time it takes) to really get the (technology) infrastructure to implement a lot of these things. I have probably a few dozen friends that work for the aforementioned companies and do these projects (note: they dont tell me what they do, its guarded like Fort Knox, but its pretty easy to tell knowing their background and the job descriptions that recruiters send to people in the industry like me to figure out whats going on)'
So, okay. What is your point? I know, this is the cutting-edge of basically everything productivity-related, and the possibility of real breakthroughs is small, but would obviously be super lucrative.
Quote:
I wasn't really talking about quantum computing, but look at things like what Intel has been trying with silicon photonics. I don't work for that team ... but to me, Moore's law is not the problem. Lots of other problems, and resolutions are much bigger.
I would be interested in hearing your perspective. Honestly. I was only focusing on the futurist mode of thinking that guarantees technological progress based on exponentially processing speeds.
Quote:
I don't object that it takes a lot of factors to implement, but what I disagree with is that you speak of it like its a small chance of ever taking over or we're going to have a terminator like society or there won't be jobs because everything is automated. Google can probably tell you this, but 2 of the 3 of these are not true.
I think these are the wrong questions to ask. There is no "it." There are people working on many different things, that yes, are seriously complicated, but won't necessarily or absolutely coalesce into "automation." Whatever that is.
Quote:
Yes to the first part, no to the 2nd. On your point about economics driving things... money is driving where innovation is going and the money for good software engineers are going into a lot of dumb fields. Besides Uber, Apple and Google which are able to recruit expensive talent.
Difference between VC investment. Lots of it because in these short-run low interest rate times there isn't a lot of places for your money to go. When I say economics, I mean the overall limits on human behavior.
Quote:
most of the limits of technology, however, aren't low hanging fruit. they are more a maze of unknowns that you aren't really sure what you are looking at (yet)
So how is that a limit? It seems like people don't even know or recognize that there are limits.
Quote:
those aren't really political or economic reasons for the failure of products. Google can tell you a better reason of the graveyard of Google products, but those aren't the reasons.
Well, they weren't profitable. That is always the main reason consumer products fail.
Quote:
political/economic can be a barrier to some, yes. But if there is enough push, barriers can be overcome. and money can do a lot of talking (we see this a lot in American politics). I can't guarantee success as you said there are many factors, but some of these companies (and not some others) likely wouldn't put so much money if they didn't think it had a realistic chance.
It is this inevitability that I really reject. It is not that I don't think technology is good. It is the gnostic belief that there is some hidden truth that is being uncovered through technological progress. Or that success is just based upon the will that wealthy people or corporations exert on a particular progress that bets on a certain outcome.
So, okay. What is your point? I know, this is the cutting-edge of basically everything productivity-related, and the possibility of real breakthroughs is small, but would obviously be super lucrative.
I would disagree its small, because its not a all or nothing deal. Parts of this technology are/have been implemented into cars for the past decade. (Lexus, Ford, Chevy, etc)
where it goes? again I'm not sure, I'm not sure if we'll ever have a pilot-less vehicle. But automation isn't a future thing, its a process that began a long time ago, is already implanted and in some of your cars already. its it a matter of "will it reach the real world" as the underlying infrastructure has already been implemented into car that has 60,000km on it.
Quote:
I would be interested in hearing your perspective. Honestly. I was only focusing on the futurist mode of thinking that guarantees technological progress based on exponentially processing speeds.
i guess a little off topic, but lots of things... heat and power dissipation, limits on power supplies, QA of such small gate size due to the microfabrication process (things like electromigration, gate breakdown)... maybe power sequencing and startups, especially if you don't know quit how to deal with race conditions in your system... solutions like other materials, moving more towards optics as a solution to these rather then using slow, drift electrons. Things like the analog behind things, noise/precision/offset of the analog, similarly noise and how its mitigated/filtered in the analog, its effect on power supply and how that powers your entire system level design... and this is still the tip of the ice berg, still... again a lot of it is a maze that is really confusing (its still really confusing to me) and most people even working in this field are walking the line on barely knowing what the hell is going on, and they aren't dumb.
lots of factors change the direction of things like supply and demand (since fabs are so expensive, see my posts regarding TowerJazz in the stocks thread; or see the economic supply/demand of Micron and how it mirrors natural gas)... not to mention just the general economics of the world and companies trying to meet their quarterly earnings and hedging their bets
all of these factor in the advancement of things you identify to be "success" or "failure" for a generalized "technology".... which you cannot do. thats like using advanced stats in hockey.
Quote:
Or that success is just based upon the will that wealthy people or corporations exert on a particular progress that bets on a certain outcome.
rich/powerful people usually have the most sway
__________________
"With a coach and a player, sometimes there's just so much respect there that it's boils over"
-Taylor Hall
Spoiler Alert:
- Weather (not just ice and snow, even just rain)
- Potholes/Manholes
- Construction
- People
All big problems still.
That is a horrible article.
1) says buses are safer because number of deaths. Doesn't extrapolate for number of people who drive cars comparably. Really stupid suggestion
2) hadn't been tested in heavy rain doesn't mean it can't do it. And other manufacturers have tested theirs, but it's really hard to find days as they're not sharing.
3) they can sense construction and humans. Even little ones
4) detecting an open manhole or large pothole would be an extremely easy programming solution
This is a far more interesting article. In fact, FT's Alphaville has a bunch of excellent criticism of autonomous vehicles, and the significant economic barriers that would inhibit full-on implementation. It is free to register. http://ftalphaville.ft.com/2015/10/2...-and-john-law/
It's not detecting humans that they were inferring is the problem. It's detecting what a human is doing that is.
I.E. A human waving their hands to stop cause of something ahead.
Big problem if there is no driver.
Well look at the problems made during the 14 people killed last weekend - are we better or worse off with the issues of automation vs human error? On one side you have someone weaving their hand to stop and a automated car doesn't. On the other, you have someone weaving their hand to stop but the driver is texting while driving and doesn't stop either. (I realize this is a little one sided, but there are way too many human errors - due to low standards as mentioned) that at least for me, want to see eliminated. Despite being pro-tech-advancement in this thread, I'm generally more conventional in my life style (no twitter, instagram, never owned google glass or oculus or fitbit, still driving a simple Honda and not a Tesla, my wife doesn't even have a smart phone and uses a pay as you go phone!)
__________________
"With a coach and a player, sometimes there's just so much respect there that it's boils over"
-Taylor Hall
That is a legitimate concern. Once AI actually takes off, the natural course is that machines will self-improve at a faster rate than humans can. As time goes on, that rate will rise exponentially thereby making humans pretty useless in the majority of jobs. A learning computer can do a better job than a human can.
In this thread it is driving a car. But artificial intelligence will have the power to take over all sorts of other jobs.
This concern has been raised since automation has existed and feared to diminish human usefulness. The tech will not create and maintain itself, maybe some day but for the foreseeable future not at all. There were some predictions that computers would put millions permanently out of work.
This concern has been raised since automation has existed and feared to diminish human usefulness. The tech will not create and maintain itself, maybe some day but for the foreseeable future not at all. There were some predictions that computers would put millions permanently out of work.
I would never say never, but I honestly can't see the process ever automating itself. Lots of jobs needed for the hardware and software. Designers, layout, device modelling, QA, EDA tools to develop these products, test engineers, some sort of system diagnostics, process engineers, lab operations, product engineeres, marketing, finance, HR, system support, mid level management, IT support, etc etc etc... I didn't even include the companies that support these, like measurement equipment, fabrication equipment... the only thing in that list that may start to get automated out is maybe some of the finance.. the rest needs a person behind it (yes including lab ops)... the world may evolve and specialize but lots of jobs are created for this stuff too.
__________________
"With a coach and a player, sometimes there's just so much respect there that it's boils over"
-Taylor Hall
Last edited by Phanuthier; 11-03-2015 at 12:25 PM.
I would never say never, but I honestly can't see the process ever automating itself. Lots of jobs needed for the hardware and software. Designers, layout, device modelling, QA, EDA tools to develop these products, test engineers, some sort of system diagnostics, process engineers, lab operations, product engineeres, marketing, finance, HR, system support, mid level management, IT support, etc etc etc... I didn't even include the companies that support these, like measurement equipment, fabrication equipment... the only thing in that list that may start to get automated out is maybe some of the finance.. the rest needs a person behind it (yes including lab ops)... the world may evolve and specialize but lots of jobs are created for this stuff too.
IQ requirements would rule out about 80% of the population.
This concern has been raised since automation has existed and feared to diminish human usefulness. The tech will not create and maintain itself, maybe some day but for the foreseeable future not at all. There were some predictions that computers would put millions permanently out of work.
Yes, but there's reason to believe this time it's different. New technologies and new businesses are not labour-intensive. You only need so many software developers to keep them afloat. Kodak once employed 150,00 people. When Instragram was bought out, it employed less than 20. Google has a valuation bigger than IBM's ever was, and employs a tiny fraction of the number of people IBM employed at its peak. Even in professions like legal, accounting, and medicine, grunt-work is being automated. Five developers can create a product that puts thousands or tens of thousands of out of work.
So what do people transition to? Services? Are our grandchildren all going to be personal trainers and sommeliers? The question there is who's going to have the money for those services if 70 per cent of people are unnecessary to the economy? Maybe each mega-rich innovator and venture capitalist will employ hundreds of lackeys to serve his every need. Doesn't sound like an especially appealing future.
__________________
Quote:
Originally Posted by fotze
If this day gets you riled up, you obviously aren't numb to the disappointment yet to be a real fan.
The Following User Says Thank You to CliffFletcher For This Useful Post:
In addition to the processing power that goes into individual cars' "brains," the other huge factor available to advance auto-pilot is the ubiquity of wireless communications for information sharing. What we'll see is that rather than relying on a static map (or GIS database) of road infrastructure, the "database" will be dynamically updated and maintained by every single vehicle that uses the road. Potholes...pedestrians leaving football games...ice-covered roads -- in 10 years or less, our cars will know far more about the current state of our roads than we do now. AND, our cars will know about problems on the road minutes before encountering them, rather than seconds before (relying on an attentive driver watching the road).
While I don't think human driving will be obsolete within 10 years, I would bet that among children born today, fewer than half will ever hold a driver's license as we now know it.
This concern has been raised since automation has existed and feared to diminish human usefulness. The tech will not create and maintain itself, maybe some day but for the foreseeable future not at all. There were some predictions that computers would put millions permanently out of work.
Automation is much different than artificial intelligence. Automation is simply replacing human labor with machine labor; i.e.: robotics. This did put a lot of people out of work, but not on a catastrophic level.
Artificial intelligence is a whole other ball game. The theory is that a computer can learn faster than a human can, and given enough time, the computers will be learning at such a rate that they will become smarter than humans. This is a real possibility. This video does a great job explaining it. And this guy is anything but a nutjob/conspiracy theorist.
