Most companies above a certain size run a type of software called "Enterprise Resource Management" or "ERP". The functionality is a bit nebulous, but it can include everything from purchasing to HR, inventory, ordering, fulfillment, etc. It's the software that essentially runs the business. There are lots of ERP systems out there, but SAP is a very very big (probably the biggest) one. There's probably some statistic about X% of fortune 500 companies use SAP as their ERP system. It's kind of notorious for being 1) expensive to license, 2) expensive to customize, 3) expensive for users to be trained on, and 4) generally sold more on the pretty graphs management gets to see rather than on the usefulness it brings to the company. Good developers who know SAP customization are paid a lot of money. Typical SAP implementations for a large business will run into the millions of dollars easily.
There is no way for this to be trustworthy. The system must be both comprehensible *and verifiable* by the vast majority of citizens. That means less technology. The future lies in simpler laws and rules. That's supposed to be the big draw of a minimum income - significantly reduce the complexity of government by making the rules extremely simple: everyone gets $X stipend. No welfare, old age pension, foodstamps, etc.
I work in the automation industry. PC-based control is very common now, and is increasing in popularity, and yes you have to firewall those systems off from the network, or air-gap them, depending on the threat model. However, even an air-gapped control system needs to have maintenance people move files on and off of it. In the typical PLC-based system there's typically a laptop with the programming software on it which you have to hook up to the PLC to program, debug, troubleshoot, etc. The fact is, a PC-based control system sometimes has advantages because the PC has the programming software on it and doesn't leave the controlled area. Still, people want to copy files, so you have to defend air-gapped systems anyway. It's a tough problem, and one that the major control system manufacturers aren't providing any assistance to help us solve either. Remember, most controls people have electrical/mechanical engineering degrees. In a large plant it should be IT's job to come up with security procedures as the automation people just aren't qualified.
The amount of pushing they do of Prime is what really turns me off of it. When someone comes to my house to sell me something, I know it's likely not in my best interest because they've invested considerable time and energy to come and solicit me, so they're likely to take a big cut, and I'm better off finding the product on my own if I can. It's the same with Amazon Prime - they seem to want to push it on me so bad that it must be a really valuable sale for them, which likely means it's not a good deal for me.
I guess I should mention that I don't use the paper filters for them either - just dump the grinds into the cup and brew it. The holes seem to big for fine grind, but I hardly ever get grinds in the coffee. If you're worried, get medium grind.
The trick is to get the right kind. When my wife got a Keurig, I really didn't want to contribute to the k-cup trash pile so I looked around for a re-usable filter. I found the ekobrew one (which was stainless steel instead of plastic). I fill it to 3/4 full of Kirkland brand fine grind that I pick up at Costco and it works great. Been using it for over a year now with no problems. I've heard that the plastic ones don't work as well.
This is about worker productivity, not household productivity. PCs are still the workhorse in industry. You don't see a vast majority of people at work doing things on smartphones. I also doubt a person obsessively checking a work email acct on their Blackberry is actually being "productive." Yes, there's lots of productivity on PCs: AutoCAD, Solidworks, graphics packages, ERP systems, inventory systems, MES systems, and now much industrial automation is moving to PC-based control.
While companies are busy measuring smartphone sales as some proxy for how well the industry is doing (and calling the PC market dead), I see a difference between PCs as machines used to do things, and smartphones as ways to waste time. Obviously this isn't exactly true, but in general it is. This is why smartphones have replaced PCs in popularity - people would rather waste time than do work. The media is so focused on getting our "attention" rather than helping us get things done, and we're so connected to that media now, that in my opinion it's obvious why productivity is falling. People aren't really working when they're "working" anymore. They're just distracted.
Also, don't discount the importance of air conditioning in US productivity, especially in the southern US.
That said, there could be a new jump in productivity as better technologies are developed. What if we counteracted smartphones with a drug or a widget that could make you focus?
Only thing I've seen so far is that the scores and user IDs are appearing in binary. Better than the flood of stupid april fools stories, in my opinion.
Are the side-sensor things multi-touch? Does this contraption have NFC integrated into the software? Is the surface of it resilient enough for you to lean on? I highly doubt this is an apples-to-apples comparison.
The vehicles, at least, are already including this call-home technology (think Ford) and it doesn't require you to actively do anything for it to call home. It probably uses the on-star-like system over cellular data, and is working to send data even if you don't subscribe to their service. As these technologies get less and less expensive, expect to see them in more and more products.
But everyone *knows* that if you want to do something alone it means you have a mental disorder! I went on a trip to the Caribbean once (there were several couples and we all went together). First day we were headed down to the beach, I brought a book plopped down in a chair and started reading while everyone else went in the water. At least two others came by and asked me if I was OK, and if I was feeling well. I'm like, "yeah, this is awesome!" The silly thing is they were mostly psychologists, but they've been steeped in the cult of "positive energy" or something, I guess. My wife understood and didn't bother me, but it was a little off-putting to have people assume something was wrong with me.
Besides, most of the "normal" people you talk to just repeat the same old stuff about magnets healing their sore elbows, or some device they put on their car that splits water into hydrogen and oxygen and gives them better fuel mileage, or about whatever they saw on Fox News or CNN last night. Giving them serious looks and nodding your approval all day does get really tiring. You can call it introversion if you want, but I'm pretty sure an actor would be tired if they had to give a 12 to 16-hour performance very day too.
No, nobody thinks it's creepy to pick up over-the-air radio signals wherever they are in the world. This thing doesn't automatically decrypt or anything like that. It's just a receiver. The cool part is that it's receiving and storing every single frequency simultaneously, and then you use a software bandpass filter to get just the station or frequency you want. No different than having a DVR that could simultaneously record every TV channel. Installing a rootkit on everyone's phone to remotely activate the camera and/or microphone - that's creepy. Connecting IoT sensors in your house and having them upload everything to the "cloud," - that's creepy.
Yes, I agree. Add to that the fact that the more repetitive the job, the more likely the employee is to have a repetitive strain injury. Depends on your jurisdiction, but here it means that employee is at home collecting disability and the company has to 1) pay a portion of it and 2) if you have too many such claims your rates go up and you get investigated, and 3) now you have to go hire someone else. The average cost for a basic automation cell is going to be in the $150,000 range (obviously it varies widely, but that's a reasonable average with all the safety systems, etc.). The total cost of a laborer per year working one shift is probably in the $35,000 to $40,000 range, and a difference in pay of a buck or two doesn't make a huge difference in that. A lot of it is the overhead of employing someone such as benefits, parking lots, protective equipment, lunch rooms, etc. So if your industrial robot cell is fast enough to replace two workers and you run two shifts, then the payback is around a year. If it can only replace one worker and you run 3 shifts, your payback is in the 16 month range. Assuming the economy is humming along and your company is growing so there's capital to invest, these are reasonable payback periods and the company will likely invest in automation. Add to this the fact that the cost of automation is continually dropping. That $150,000 number I used was from a few years ago, and recent experience tells me that number is a bit high now.
I work in automation. It isn't so much that minimum wage matters... sure if you have really, really low minimum wage and people willing to work for it then you might just throw labor at a problem, but typically we automate for a variety of reasons: improved accuracy/quality, better throughput (a robot loading a machine can often keep up better than a human, which means I get more throughput out of my expensive machine), more consistent process. We *want* to automate everything, and when we look at what we *can* automate, it's always the boring repetitive jobs anyway. So it doesn't matter that much whether someone's making $6 or $8 or $10 an hour, if we can automate it we will. Certainly we are growing more concerned with the fact that a growing percentage of the population isn't going to be able to find the easy put-nut-A-on-bolt-B type of work anymore, and there's definitely a portion of those people who may not be able to be retrained to do something that a robot can't do. That's a societal problem, not an engineering problem. First is understanding that this isn't the same thing we saw in the industrial revolution. If I gave a laborer a steam shovel I made them a lot more productive. If I just say "stand aside while this robot does the job" that's different. And no, you're not going to take someone who works on an assembly line and retrain them to be a robot programmer. That's absurd. They won't get a job assembling robots either, as Fanuc apparently has a "lights out" manufacturing facility for their robots - it's a completely automated line. Minimum wage is doing a good thing: encouraging factories to automate by making the payback look better. Automated factories are better. Automated restaurants are probably better too. The fact that we have a very low skilled portion of the populace is a separate issue that needs addressing... maybe a guaranteed minimum income, I don't know. But coming up with make-work jobs for them is no better than putting them in prison and having them dig holes and fill them in. Also relevant to this discussion - has everyone seen the short story, "Manna"?
I'm going from memory here. The basic "pyramid" had 4 levels, where the higher the level, the higher the quality:
Intellectual thought/logic/ideas (highest)
Society
Biology
Physics (lowest)
The lower levels of the pyramid have value in that they support the higher levels. So the physical laws of the universe have "quality" in that they support the higher levels of quality. Also it means that building a house would be good (physics) if it sheltered a person (biology) who had the capacity to form ideas about the universe (logic/thought). Also, a society was good if it promotes free exchange of ideas, but is bad if it inverts the relationship of quality (so a fascist dictatorship is bad in that it suppresses free thought). Killing living things is bad, because life (biology) has more quality than dead stuff (physics) but curing a virus (biology) by killing it has quality because a virus can kill a person, and a person can have think about the universe.
Then there's the differentiation of static vs. dynamic quality. Easiest example is evolution - the static quality is encoded in DNA and replicates (remembers) with each generation. Mutations are dynamic quality. Dynamic quality has to be tested (natural selection) and if it's found to be better, then needs to be remembered, but if it's worse, then discarded. His point is that this static/dynamic thing exists at all levels. There needs to be a mechanism of remembering good ideas, testing new (dynamic) ideas, and adding that to the (static) body of knowledge. He somewhat illustrates it as an ongoing tension between static and dynamic quality, a little like conservatism (static) vs. liberalism (dynamic).
That's quite simple. The behavior you'd expect a moral human being to take in the same situation. In the case of Asimov's stories, the failure is usually quite obvious. My point is that Pirsig's framework actually gives you a good way to determine what that "correct" behavior should be.
According to TFS, it learned by running simulations of a situation and then being rewarded or punished based on its actions in the simulation. They just happened to setup the simulation, reward, and punishment based on a story they selected. I'd hardly call that learning by reading a story.
I remember reading Zen and the Art of Motorcycle Maintenance and the "sequel" Lila and thinking that what Pirsig had done wasn't inventing some new philosophy, but he did a really good job of expressing western values in a rule-based way. For instance, it explains why killing is wrong, but why a moral individual might find themselves in a situation where killing is justified. It explains how some forms of government are better than others, and why. As I said, it's all been done before, but what impressed me was that it was very clearly defined and rule-based. Everyone talks about encoding Asimov's 3 laws into robots, but Asimov's stories were all about how those 3 laws failed to produce correct behavior. If I were trying to program morals into a robot, I'd start with Pirsig's books and his ideas of static and dynamic quality.
The right thing to do here is to cheer on this research. Everyone now just has this idea: it's a nice thought but it won't work. However, the only major experiment done so far (in Manitoba in the 70's) indicated the net result seems to be positive. Now with Finland planning a big experiment and 4 Dutch cities trying it, and now Y Combinator, we might finally get some data to see how people really react. No matter whether you think this will work or not, at least there are experiments on the go to actually try it. That's excellent.
I'm in automation, so I do see stuff like this happening. However, the counter-point is invariably that people claimed the industrial revolution would do the same, and it did the opposite. People making this point about AI putting so many people out of work need to focus on showing how this revolution is so fundamentally different than the industrial revolution, or people won't buy the argument.
Just as only wealthy ancient Romans could afford lead pipes, only wealthy individuals now can afford these silly expensive "connected/smart" appliances, so we're in an odd situation where being well off affords you much more risk of being hacked. Everyone else will be just fine with their hardwired home controls.
Slashdot Mirror
Most companies above a certain size run a type of software called "Enterprise Resource Management" or "ERP". The functionality is a bit nebulous, but it can include everything from purchasing to HR, inventory, ordering, fulfillment, etc. It's the software that essentially runs the business. There are lots of ERP systems out there, but SAP is a very very big (probably the biggest) one. There's probably some statistic about X% of fortune 500 companies use SAP as their ERP system. It's kind of notorious for being 1) expensive to license, 2) expensive to customize, 3) expensive for users to be trained on, and 4) generally sold more on the pretty graphs management gets to see rather than on the usefulness it brings to the company. Good developers who know SAP customization are paid a lot of money. Typical SAP implementations for a large business will run into the millions of dollars easily.
Can I write the software? Oh you forgot about that little detail, eh? I guess you'll insist on it being open course, of course. Sure, nobody could every fool you if you could see the source code.
There is no way for this to be trustworthy. The system must be both comprehensible *and verifiable* by the vast majority of citizens. That means less technology. The future lies in simpler laws and rules. That's supposed to be the big draw of a minimum income - significantly reduce the complexity of government by making the rules extremely simple: everyone gets $X stipend. No welfare, old age pension, foodstamps, etc.
I work in the automation industry. PC-based control is very common now, and is increasing in popularity, and yes you have to firewall those systems off from the network, or air-gap them, depending on the threat model. However, even an air-gapped control system needs to have maintenance people move files on and off of it. In the typical PLC-based system there's typically a laptop with the programming software on it which you have to hook up to the PLC to program, debug, troubleshoot, etc. The fact is, a PC-based control system sometimes has advantages because the PC has the programming software on it and doesn't leave the controlled area. Still, people want to copy files, so you have to defend air-gapped systems anyway. It's a tough problem, and one that the major control system manufacturers aren't providing any assistance to help us solve either. Remember, most controls people have electrical/mechanical engineering degrees. In a large plant it should be IT's job to come up with security procedures as the automation people just aren't qualified.
The amount of pushing they do of Prime is what really turns me off of it. When someone comes to my house to sell me something, I know it's likely not in my best interest because they've invested considerable time and energy to come and solicit me, so they're likely to take a big cut, and I'm better off finding the product on my own if I can. It's the same with Amazon Prime - they seem to want to push it on me so bad that it must be a really valuable sale for them, which likely means it's not a good deal for me.
I guess I should mention that I don't use the paper filters for them either - just dump the grinds into the cup and brew it. The holes seem to big for fine grind, but I hardly ever get grinds in the coffee. If you're worried, get medium grind.
The trick is to get the right kind. When my wife got a Keurig, I really didn't want to contribute to the k-cup trash pile so I looked around for a re-usable filter. I found the ekobrew one (which was stainless steel instead of plastic). I fill it to 3/4 full of Kirkland brand fine grind that I pick up at Costco and it works great. Been using it for over a year now with no problems. I've heard that the plastic ones don't work as well.
This is about worker productivity, not household productivity. PCs are still the workhorse in industry. You don't see a vast majority of people at work doing things on smartphones. I also doubt a person obsessively checking a work email acct on their Blackberry is actually being "productive." Yes, there's lots of productivity on PCs: AutoCAD, Solidworks, graphics packages, ERP systems, inventory systems, MES systems, and now much industrial automation is moving to PC-based control.
While companies are busy measuring smartphone sales as some proxy for how well the industry is doing (and calling the PC market dead), I see a difference between PCs as machines used to do things, and smartphones as ways to waste time. Obviously this isn't exactly true, but in general it is. This is why smartphones have replaced PCs in popularity - people would rather waste time than do work. The media is so focused on getting our "attention" rather than helping us get things done, and we're so connected to that media now, that in my opinion it's obvious why productivity is falling. People aren't really working when they're "working" anymore. They're just distracted.
Also, don't discount the importance of air conditioning in US productivity, especially in the southern US.
That said, there could be a new jump in productivity as better technologies are developed. What if we counteracted smartphones with a drug or a widget that could make you focus?
Throwing your customers under the bus is always a good PR move. Way to go.
"... but some are more equal than others."
Only thing I've seen so far is that the scores and user IDs are appearing in binary. Better than the flood of stupid april fools stories, in my opinion.
Are the side-sensor things multi-touch? Does this contraption have NFC integrated into the software? Is the surface of it resilient enough for you to lean on? I highly doubt this is an apples-to-apples comparison.
The vehicles, at least, are already including this call-home technology (think Ford) and it doesn't require you to actively do anything for it to call home. It probably uses the on-star-like system over cellular data, and is working to send data even if you don't subscribe to their service. As these technologies get less and less expensive, expect to see them in more and more products.
You won't know about all the ones that come in the appliances and vehicles you buy. They have no incentive to tell you.
But everyone *knows* that if you want to do something alone it means you have a mental disorder! I went on a trip to the Caribbean once (there were several couples and we all went together). First day we were headed down to the beach, I brought a book plopped down in a chair and started reading while everyone else went in the water. At least two others came by and asked me if I was OK, and if I was feeling well. I'm like, "yeah, this is awesome!" The silly thing is they were mostly psychologists, but they've been steeped in the cult of "positive energy" or something, I guess. My wife understood and didn't bother me, but it was a little off-putting to have people assume something was wrong with me.
Besides, most of the "normal" people you talk to just repeat the same old stuff about magnets healing their sore elbows, or some device they put on their car that splits water into hydrogen and oxygen and gives them better fuel mileage, or about whatever they saw on Fox News or CNN last night. Giving them serious looks and nodding your approval all day does get really tiring. You can call it introversion if you want, but I'm pretty sure an actor would be tired if they had to give a 12 to 16-hour performance very day too.
No, nobody thinks it's creepy to pick up over-the-air radio signals wherever they are in the world. This thing doesn't automatically decrypt or anything like that. It's just a receiver. The cool part is that it's receiving and storing every single frequency simultaneously, and then you use a software bandpass filter to get just the station or frequency you want. No different than having a DVR that could simultaneously record every TV channel. Installing a rootkit on everyone's phone to remotely activate the camera and/or microphone - that's creepy. Connecting IoT sensors in your house and having them upload everything to the "cloud," - that's creepy.
Yes, I agree. Add to that the fact that the more repetitive the job, the more likely the employee is to have a repetitive strain injury. Depends on your jurisdiction, but here it means that employee is at home collecting disability and the company has to 1) pay a portion of it and 2) if you have too many such claims your rates go up and you get investigated, and 3) now you have to go hire someone else. The average cost for a basic automation cell is going to be in the $150,000 range (obviously it varies widely, but that's a reasonable average with all the safety systems, etc.). The total cost of a laborer per year working one shift is probably in the $35,000 to $40,000 range, and a difference in pay of a buck or two doesn't make a huge difference in that. A lot of it is the overhead of employing someone such as benefits, parking lots, protective equipment, lunch rooms, etc. So if your industrial robot cell is fast enough to replace two workers and you run two shifts, then the payback is around a year. If it can only replace one worker and you run 3 shifts, your payback is in the 16 month range. Assuming the economy is humming along and your company is growing so there's capital to invest, these are reasonable payback periods and the company will likely invest in automation. Add to this the fact that the cost of automation is continually dropping. That $150,000 number I used was from a few years ago, and recent experience tells me that number is a bit high now.
I work in automation. It isn't so much that minimum wage matters... sure if you have really, really low minimum wage and people willing to work for it then you might just throw labor at a problem, but typically we automate for a variety of reasons: improved accuracy/quality, better throughput (a robot loading a machine can often keep up better than a human, which means I get more throughput out of my expensive machine), more consistent process. We *want* to automate everything, and when we look at what we *can* automate, it's always the boring repetitive jobs anyway. So it doesn't matter that much whether someone's making $6 or $8 or $10 an hour, if we can automate it we will. Certainly we are growing more concerned with the fact that a growing percentage of the population isn't going to be able to find the easy put-nut-A-on-bolt-B type of work anymore, and there's definitely a portion of those people who may not be able to be retrained to do something that a robot can't do. That's a societal problem, not an engineering problem. First is understanding that this isn't the same thing we saw in the industrial revolution. If I gave a laborer a steam shovel I made them a lot more productive. If I just say "stand aside while this robot does the job" that's different. And no, you're not going to take someone who works on an assembly line and retrain them to be a robot programmer. That's absurd. They won't get a job assembling robots either, as Fanuc apparently has a "lights out" manufacturing facility for their robots - it's a completely automated line. Minimum wage is doing a good thing: encouraging factories to automate by making the payback look better. Automated factories are better. Automated restaurants are probably better too. The fact that we have a very low skilled portion of the populace is a separate issue that needs addressing... maybe a guaranteed minimum income, I don't know. But coming up with make-work jobs for them is no better than putting them in prison and having them dig holes and fill them in. Also relevant to this discussion - has everyone seen the short story, "Manna"?
I know it was a military drone in the movie, but suddenly that scene in Interstellar doesn't seem quite so crazy. :)
I'm going from memory here. The basic "pyramid" had 4 levels, where the higher the level, the higher the quality:
The lower levels of the pyramid have value in that they support the higher levels. So the physical laws of the universe have "quality" in that they support the higher levels of quality. Also it means that building a house would be good (physics) if it sheltered a person (biology) who had the capacity to form ideas about the universe (logic/thought). Also, a society was good if it promotes free exchange of ideas, but is bad if it inverts the relationship of quality (so a fascist dictatorship is bad in that it suppresses free thought). Killing living things is bad, because life (biology) has more quality than dead stuff (physics) but curing a virus (biology) by killing it has quality because a virus can kill a person, and a person can have think about the universe.
Then there's the differentiation of static vs. dynamic quality. Easiest example is evolution - the static quality is encoded in DNA and replicates (remembers) with each generation. Mutations are dynamic quality. Dynamic quality has to be tested (natural selection) and if it's found to be better, then needs to be remembered, but if it's worse, then discarded. His point is that this static/dynamic thing exists at all levels. There needs to be a mechanism of remembering good ideas, testing new (dynamic) ideas, and adding that to the (static) body of knowledge. He somewhat illustrates it as an ongoing tension between static and dynamic quality, a little like conservatism (static) vs. liberalism (dynamic).
That's quite simple. The behavior you'd expect a moral human being to take in the same situation. In the case of Asimov's stories, the failure is usually quite obvious. My point is that Pirsig's framework actually gives you a good way to determine what that "correct" behavior should be.
According to TFS, it learned by running simulations of a situation and then being rewarded or punished based on its actions in the simulation. They just happened to setup the simulation, reward, and punishment based on a story they selected. I'd hardly call that learning by reading a story.
I remember reading Zen and the Art of Motorcycle Maintenance and the "sequel" Lila and thinking that what Pirsig had done wasn't inventing some new philosophy, but he did a really good job of expressing western values in a rule-based way. For instance, it explains why killing is wrong, but why a moral individual might find themselves in a situation where killing is justified. It explains how some forms of government are better than others, and why. As I said, it's all been done before, but what impressed me was that it was very clearly defined and rule-based. Everyone talks about encoding Asimov's 3 laws into robots, but Asimov's stories were all about how those 3 laws failed to produce correct behavior. If I were trying to program morals into a robot, I'd start with Pirsig's books and his ideas of static and dynamic quality.
The right thing to do here is to cheer on this research. Everyone now just has this idea: it's a nice thought but it won't work. However, the only major experiment done so far (in Manitoba in the 70's) indicated the net result seems to be positive. Now with Finland planning a big experiment and 4 Dutch cities trying it, and now Y Combinator, we might finally get some data to see how people really react. No matter whether you think this will work or not, at least there are experiments on the go to actually try it. That's excellent.
I'm in automation, so I do see stuff like this happening. However, the counter-point is invariably that people claimed the industrial revolution would do the same, and it did the opposite. People making this point about AI putting so many people out of work need to focus on showing how this revolution is so fundamentally different than the industrial revolution, or people won't buy the argument.
Just as only wealthy ancient Romans could afford lead pipes, only wealthy individuals now can afford these silly expensive "connected/smart" appliances, so we're in an odd situation where being well off affords you much more risk of being hacked. Everyone else will be just fine with their hardwired home controls.