Disney already tried this with licensing out their characters to other companies to produce video games. They decided to stop that practice entirely and use an in-house game studio instead. Their games went to shit. Then a couple years later, they started licensing out again.
I have a feeling that history will repeat itself with this news of licensing streaming content.
So, Disney is bad at making video games. They seem to be pretty successful at making movies; have been for a century. Streaming licensing has nothing to do with the making of movies, it's just one distribution channel, and not even the most lucrative.
This 'technology' is being rushed way too quickly to market.
I'd like to agree with you, particularly with respect to the semi-autonomous systems presently deployed. I argued for years that having a system that worked most of the time but expected the user to take over when necessary was extremely dangerous. But the thing is that human drivers are extremely dangerous. Tesla has very compelling data showing that, as half-baked as their system is, it's actually better than the human drivers that it's replacing. The same will be even more true of the first fully-autonomous vehicles.
The systems don't have to be perfect, they just have to be better, and the bar is not very high.
Some societies have more women studying CS than men, e.g. Iran. The women there moved into that relatively new field before it became male dominated, and view it as liberating.
Programming and CS used to be "woman's work" in the US as well. Indeed, the first professional software engineers I met in the 70s were all women. Over the decades women left and/or were forced out. I wonder if there might be a similar dynamic in Iran.
Other examples include Iceland and New Zealand, where girls now slightly out-perform boys in maths at school.
Statements like this actually contain a lot less information than they appear to. The scientifically-demonstrated gender differences are complex, and they produce different statistical distributions of abilities across the populations. Merely noting that the mean of test performance leans one way or the other doesn't actually say much. It could be that there is no biological difference in mean math ability (which is actually a synthesis of multiple cognitive abilities, each likely with its own per-gender statistical distribution), which would mean that the difference in mean scores is entirely sociologically-determined. Or perhaps girls even have a slight biological advantage, which could mean that Iceland and New Zealand have successfully erased sociologically-driven gender bias in mathematics. Or maybe girls have a large biological advantage, and as we eliminate sociological bias, girls will come out far ahead in mean mathematical ability, and Iceland and New Zealand are ahead of the curve but haven't got the job done yet.
But that doesn't necessarily mean that, for example, we'll start seeing women dominating the leading edges of mathematical research, or filling an equivalent share of posts in prestigious mathematical faculties, or winning math's highest awards, like the Fields Medal, because the people who do those things aren't average. The extremes depend less on the location of the mean,and more on the value of the variance.
In many, many areas of physical, cognitive and behavioral characteristics, it has been found that men have a larger variance than women do, and therefore that extreme outliers are almost always men. This could be part of the reason that only one woman has ever won the Fields Medal, because extreme mathematical talent is rarer among women than among men, even if average ability is the same... or even favors women. It's very likely part of the reason why there are so many more men in prison than women, because criminality is correlated with cognitive deficiencies in logic and abstract thinking, and men dominate the outliers on that end of the spectrum as well.
(Note that I am not making the claim that men do have greater levels of variance in mathematical ability than women. That's an empirically-testable question to which neither of us know the answer. Men have greater levels of variance in many areas, so it's a plausible hypothesis that they do in math -- or programming -- as well. I'm accepting the hypothesis as a hypothetical, and exploring the implications if it turned out to be true. But the science needs to be done to find out if it is true.)
With respect to Google, while Google engineers are not the sort of extreme outliers that Fields Medal winners are, they are hardly average people. Programming talent is not the norm in the human population (most people find it very hard to think that way), and it may well be that competent programmers must come from the upper tails of the bell curve in several cognitive skills (and possibly the lower tails in other areas). It so, we may never see gender equity in the field without extreme affirmative action, because if, say, 5% of men have the characteristics needed for the job, it may be that only 1% of women do, even there is no difference in mean ability. And when you narrow that field to, say, the best 1% of programmers, the gender ratio would skew further since you're reaching farther from the mean.
I wonder how this compares to Google's approach to speeding up ML, the Tensor Processing Unit, and whether the ideas can be combined for even faster learning.
It's all old, tired arguments that have been comprehensively refuted before.
I don't think this is true. Note that I'm strongly pro-diversity and think that affirmative action makes sense, so I'm not arguing in favor of the doc's author's conclusions, but I note that all of the responses keep saying that his science was wrong... and as far as I can tell it's not. The doc itself has pretty extensive citations to current research, and the only responses I can find from working scientists in the field all state that he got the science right, with minor errors at most.
The conclusions he drew from the science are bad, because the gender and racial differences science has found are probably too small to produce the results he attributes to them.
I am curious: does that include discrimination against those protected classes in the job interview process? Like, say, for example, ageism?
Ageism isn't really an issue at Google. Most of the employee population is young-ish, because the company hasn't been around long and does most of its hiring straight out of school. Not all, though, and there are many successful older employees. I'm nearly 50 and work for Google, and although my current team trends young (though I'm not the oldest), my previous team included several guys in their 60s and one in his 70s (who didn't need to work, but liked it).
That said, I would agree that the interview process does not necessarily value the things that older engineers bring to the table that youngsters do not. In addition, older engineers may need to brush up on their CS fundamentals before going into the interview, while recent grads will be fresh. Assuming equivalent CS fundamentals knowledge, I'd say that the playing field is quite level, but it is a playing field that is aimed at testing the things that both can do and specifically ignores the value of experience, which younger candidates cannot match.
The value of experience is factored in when it comes time to set salary, of course. And it's definitely taken into account when putting people on teams and handing out tasks.
Didn't really matter because they all cost about the same regardless. I suppose you could say they were price fixing, or colluding or whatever.
In most places their prices were set by the taxi authority, by regulation, mostly to make sure you didn't find yourself getting ripped off. This was necessary precisely because there was no good way (unlike Uber) for you to know in advance what your price would be. In the modified approach I suggest, in which drivers set their own prices, you'd still be able to know what price you'd pay before getting in, so there's no need for central control, either by Uber or by government.
What makes Uber stand out as well is their policy on how old the car can be, so you don't end up getting a lift from some ancient beat up car.
That is nice. Uber could maintain that requirement while allowing drivers to set their own prices. That policy wouldn't be incompatible with true independent contractor status. Alternatively, the app could give you vehicle quality rating information as well as price, and you could choose whether you wanted to get in a beater for a lower fare. That might make the user interface unusably complicated, though. It would require testing.
It's all old, tired arguments that have been comprehensively refuted before. For example, he states that women are more neurotic and less able to deal with stress. We know that isn't true, because we have studied it in great detail.
Of course, as the first of those abstracts points out, variation among individuals is far higher than the measured difference between genders, so it's hard to see how this can have a large effect.
You're forgetting the program in which applicants who were declined by hiring committee could be sent back through for another chance. And this was mainly used for applicants from underrepresented minorities. Created a huge stir when someone suggested it was lowering the bar (even though it obviously is, even if only slightly).
I didn't hear about that one. I wouldn't be surprised if it also applies to candidates from small schools though. The sense I get from talking to the recruiters is that "diversity" is a boolean label that is applied to a candidate, and then a whole set of additional options kick in.
everything else in the paper are legitimate concerns about "how" diversity is being enforced.
One of his claims about how diversity is being "enforced" is factually erroneous. He says that Google offers programs which make it easier for "diversity" candidates to get hired, and claims that those programs are only open to women and minority races.
As it happens, I work for Google, and do university outreach to my alma mater, which is a small four-year commuter university quite different from the big-name schools that lots of Googlers attended. Because the school is one that Google typically does not get many applicants from, or many many hires from, all students at my university are considered "diversity" hires, regardless of their race or gender, and are eligible for the diversity programs. I believe the same is true of almost any small college or university in the country that doesn't traditionally have a strong CS program.
In addition, it should be pointed out that Google does not "lower the bar" for these diversity candidates. Instead, the programs provide additional mentoring and development through a summer internship (for students) or a one-year contract (for graduates). The programs include education as well as opportunities to work with regular employees on real products. At the end of the contract, the participants who want to attempt to convert to full-time employees go through the regular interview process and are scrutinized according to the same standards as anyone else. I understand the programs are fairly successful, and a high percentage of participants convert to regular employees.
First, I'll mention that I'm old enough that I used to navigate all the time without GPS. I traveled a lot for business from the mid 90s until about 2010, and didn't start using GPS navigation until ~2005. Almost every week for a decade I was finding my way around a different city, getting from the airport car rental location to the hotel, to the customer site (or sites) to restaurants, etc., all with nothing more than paper maps and road signs. So I know that not only is it possible, it's not even hard. It does require having the paper maps (rental car agencies still hand them out on demand) and it requires more preparation and more time -- and if you don't want to be late also generally involves arriving at destinations fairly early, because although you can estimate distances and times from the map, you know nothing about local traffic and road conditions.
Now, I use Google Maps even when I know exactly where I'm going, around my own home town, because I like to see the computed ETA. It's way, way better than I am at guessing travel time, especially since it factors in real-time traffic. On occasion doing this has saved me a bunch of time, too, because it has routed me around accidents and road closures and whatnot.
All in all, I greatly prefer GPS navigation over paper maps. About the only thing I think I've lost is that with paper maps you always end up with a sense of the relative locations of things. With GPS navigation you can travel a city for weeks without ever learning your way around.
I have yet to see a good way on a small-screen phone or GPS to answer questions like "how far away is the coast,"
If you just want a rough idea, pinch zoom out to where you can see the coast, then judge the distance from the little distance scale. If you want a more precise idea, it's probably because you would like to know how long it would take to get there. Pick a reasonable point and ask the phone to navigate there, and you'll get a very precise answer, both distance and time. Better than you're going to get from a paper map.
"how much out of our way would it be to go to that town"
Now that Google Maps supports searching along your route, this is easy, and the answer is very precise. I just say "Okay Google <town name>". Maps shows me the city center, with a notation saying how much time it will add to my trip to go there. If the town is big enough that I don't want the city center, I have it search for gas stations in the city. Maps will find a list of gas stations in the town and near my route, and show me all of them marked on the map, with a notation indicating how much time it will add to my trip if I go to each. This works fine even if the town in question is off-route.
what's the next sizable town within 2 miles of the road we are on
This one is more difficult, but can be done with a little zooming and scrolling to look along the road/route. The trick of searching for gas stations can also be used (without specifying town name) or modified to find something else that indicates a sizable town. Odds are you don't want a "sizable" town anyway, but instead are looking for some specific amenity: restaurant, hotel, golf course, etc., which you're not likely to find in a town that's nothing more than a wide spot in the road. So search for the amenity and see what comes up.
Unless you can then resell the token for real world currency
You can.
Have you not heard of gold farmers?
However, farmed gold sells at a lower exchange rate than WoW token dollars do going the other way, due to the risk created by the fact that selling gold for cash is a violation of the terms of service, so the buyer is taking a risk buying farmed gold.
On the gripping hand, though, at the rate the Bolivar is going it'll drop below the value of a farmed gold piece in a few months.
Based on what I've been reading Tesla currently only puts regenerative brakes in the rear. Since most of your brake loading is on the front wheels, I doubt this reduces pad wear by very much.
Your doubt is misplaced.
EV drivers tend to use nothing but regeneration for the vast majority of braking. This requires driving a little less aggressively and "coasting" most of the way to the stoplight, engaging the brake pads only at the end. Yes, this means that most braking is done only with the rear wheels (on a single-motor Tesla), but we're talking about normal, gentle braking. Aggressive braking is where you need to make sure the braking force is appropriately distributed, and that works the same as an ICEV.
None of that matters, he wasn't hailing the virtues of Visa's transaction handling, he was pointing out their transaction rate as a guide to what is required to meet the needs of commerce today. BTC is nowhere near where it needs to be if it is to take over.
I wasn't commenting on BTC, just correcting a common misunderstanding of how credit card transactions work.
If you want a comment on BTC, though, I'll say this: The only way to scale to very high transaction volumes is by decentralizing. BTC's problem is that it is too centralized, which seems an odd thing to say about a fully decentralized transaction ledger, but it's true. The problem is that there's only one ledger, and all copies of it must be synchronized. I can think of a few different ways this could be solved, so the problem isn't inherent in blockchain-based cryptocurrencies, but only represents a flaw in the design of this one.
Yeah, like that fucking worked with metered taxi's.
Completely different situation. With metered taxis riders couldn't choose from all of the taxis within a large radius of them, and couldn't know the price they'd pay for their trip before they selected the cab to take.
Yep, it's risky. I think they'll make it, and I expect a handsome reward when they do. Or I may lose a few thousand dollars if they don't. That's the game. The bulk of my retirement portfolio is in safer, more consistent bets, of course.
It's explained reasonably well by analogy with the Parable of the Ox
No, it isn't. At least, it isn't if we're talking about shares in real companies, or quantities of real commodities.
The parable starts out pretty good, but where it breaks down is in its assumption that the scale goes away completely. In the parable, the weight of the ox becomes entirely divorced from any sort of reality, but that doesn't happen in securities, because there is an underlying value in the security: it's future earning potential. The parable really needs to add some notion that the ox, or parts of it, will eventually be weighed, actually weighed, not just acceptance of an average value of guesses, and incorrect guesses will lose at that point.
For example, for the last couple of years I've been investing money now and then in shares of TSLA. At this point, I own a few thousand dollars worth of the company, based on the current consensus view of what the whole company is worth, divided by the number of shares in total. Now this consensus total value is many, many times the value of Tesla's actual assets. And it's also significantly higher than most other common methods of estimating value, based on revenue, income, etc.
Why is it high? Because lots of people guess it should be? Sort of... but not really. And we will eventually find out whether that high valuation was or was not correct. The ox will get weighed. How?
It's high because there is a possibility that Tesla will experience explosive success, generating massive revenues and profits that will replay the relatively high cost of investing now. Pulling the price down, though, is the possibility that Tesla will fail to achieve that level of success, but just continue being a niche carmaker. Pulling it down some more is the possibility that the EV and solar home markets will never really pan out, or that Tesla will fail to execute well on delivering its promises to its customers. In either case, it could crash and burn, becoming worth a tiny fraction of its current value, or even going bankrupt. I owned one stock that followed this path, becoming completely worthless.
There's an underlying business that will succeed or fail, and in doing so will justify some current stock price. We don't know what that currently-justifiable price is, which is where the uncertainty and the guessing come in. But, as I said before, the ox will be weighed. There is a core reality underlying the security whose value we're guessing; it's not pure guess-averaging.
There is no objective measurement of perceived value, which is what the economy runs on.
Nonsense.
You absolutely can objectively measure value, in both individual and aggregate cases. The measurement will not be a single number, it will be a probability distribution, and the distribution will vary over time, but that makes it no less a real, objective measurement. What you're measuring is human demand and desire, and that is a real, objective thing, even if it's variable across the population and across time, and even if it is hard to measure accurately.
If you measured the value of a slice of Wonder bread, you'd find the distribution has a mean on the order of a few pennies, and the vast majority of potential buyers would be willing to buy it for that. There is a non-zero probability you can find someone to pay $10K for it, but it's extremely small unless there's something unusual about your particular piece of Wonder bread. If it's fungible with all of the other pieces of Wonder bread, then the probability approaches zero.
Basically, you're confusing "objective" with "fixed and universal". Objective measurements need not have single values, nor do they have to remain fixed over time or space.
In the case of business valuations, unlike goods there's very little variation (after risk adjustment) across the population, because people rarely have individualistic reasons for wanting to own shares of one company over another. People buy stocks because they hope to make money. On the other hand, business valuations are inherently rooted in predictions of future business prospects. That doesn't make them so much "subjective" as it does "uncertain". Different analysts will assign different probabilities to different outcomes, because forecasting is an unpredictable business. Also, new facts will change the probabilities.
However, aggregate, broad-scale predictions, made by large groups of people tend to produce fairly good results... and that's what stock prices are. Aggregated predictions.
What this study found is that the aggregated predictions of small groups of private investors do not match the aggregated predictions of the public markets, and that the private investors systematically overvalue the companies, as compared to the public markets. Does this mean the private investors' predictions are wrong and the public markets' are right? In one way, yes, by definition, because the private investors ultimately cash out through the public markets, so the private investors are really trying to predict the valuations that will be assigned by the public markets. In another way, no one knows, because the actual value of the company depends on what happens in the future. On average, though, it's likely that the public markets make better predictions.
You're assuming that people in your area can only sell local. Why do you assume that? Some businesses are inherently local, sure, but you don't need to make a choice to support them, because you can't buy remotely anyway. But many businesses are globalizable, and that applies to both buyers and sellers.
Your numbers are a little out of date. Visa last year alone averaged 4500 transactions a second.
It should be pointed out that "Visa" isn't a transaction clearing system. It's a bank association. Large numbers of transactions are easily handled because there is no single system that all of the transactions flow through. There is a set of card issuing banks, a set of merchant acquiring banks, and a set of clearinghouses that connect them -- for the cases that the acquiring banks and issuing banks don't just connect directly, which they often do.
The Leaf is being heavily discounted right now, which is why I'm even considering one.
If you're not getting the 2018 LEAF with its bigger battery, I suggest looking for a used car. 2012-2014 LEAFs are crazy cheap right now. The main thing to watch out for is the battery capacity. Ideally, you want one that still has 12 bars on the capacity display. If you can do with less capacity, fine, but the price should drop accordingly.
I recommend getting a Bluetooth OBDII adapter and the LeafSpy Pro app on your phone. Plug in the adapter and the app will show you the total battery capacity with a high degree of accuracy. After you've found a car that seems good mechanically and in appearance, and has reasonable battery capacity for its price, take it to a Nissan dealership and have them do the free battery analysis.
Unlike with ICEVs, there's very little to go wrong with the drivetrain. Electric motors last forever, there is no transmission to speak of (just a reversing gear). Take a look at CV joints, etc., just like you would an ICEV. Brake pads wear out, but slowly, and replacing them is the same as an ICEV. So really it's just the battery you need to pay attention to when evaluating a used EV.
The Leaf's lack of a TMS causes their battery to degrade rapidly, losing as much as 40% of its capacity in 2-3 years
This depends on where you live. My 2012 LEAF has 50K miles on it and has lost only about 4% of its battery capacity. (I'd go check it right now with my OBDII interface and LeafSpy Pro, but my wife has it.)
Get a used Volt. They're ass-cheap. Just don't buy a Leaf
Used LEAFs are much cheaper. You can get one about like mine (note: I'm not selling mine; I like it) for around $6K. Assuming you don't need more than 60 or so miles of daily range, and don't live in an area with a very hot climate (which causes rapid battery degradation), for $6K you can get an EV that will be a great commuter and around-town vehicle for several years, and will cost less than a nickel per mile to operate, including electricity and maintenance.
Unless you live in Arizona,or the like, they're great little cars, and very, very cheap right now.
Slashdot Mirror
Bad use of statistics on one area does not imply that they're applied incorrectly everywhere.
Disney already tried this with licensing out their characters to other companies to produce video games. They decided to stop that practice entirely and use an in-house game studio instead. Their games went to shit. Then a couple years later, they started licensing out again.
I have a feeling that history will repeat itself with this news of licensing streaming content.
So, Disney is bad at making video games. They seem to be pretty successful at making movies; have been for a century. Streaming licensing has nothing to do with the making of movies, it's just one distribution channel, and not even the most lucrative.
Completely different situations.
This 'technology' is being rushed way too quickly to market.
I'd like to agree with you, particularly with respect to the semi-autonomous systems presently deployed. I argued for years that having a system that worked most of the time but expected the user to take over when necessary was extremely dangerous. But the thing is that human drivers are extremely dangerous. Tesla has very compelling data showing that, as half-baked as their system is, it's actually better than the human drivers that it's replacing. The same will be even more true of the first fully-autonomous vehicles.
The systems don't have to be perfect, they just have to be better, and the bar is not very high.
Some societies have more women studying CS than men, e.g. Iran. The women there moved into that relatively new field before it became male dominated, and view it as liberating.
Programming and CS used to be "woman's work" in the US as well. Indeed, the first professional software engineers I met in the 70s were all women. Over the decades women left and/or were forced out. I wonder if there might be a similar dynamic in Iran.
Other examples include Iceland and New Zealand, where girls now slightly out-perform boys in maths at school.
Statements like this actually contain a lot less information than they appear to. The scientifically-demonstrated gender differences are complex, and they produce different statistical distributions of abilities across the populations. Merely noting that the mean of test performance leans one way or the other doesn't actually say much. It could be that there is no biological difference in mean math ability (which is actually a synthesis of multiple cognitive abilities, each likely with its own per-gender statistical distribution), which would mean that the difference in mean scores is entirely sociologically-determined. Or perhaps girls even have a slight biological advantage, which could mean that Iceland and New Zealand have successfully erased sociologically-driven gender bias in mathematics. Or maybe girls have a large biological advantage, and as we eliminate sociological bias, girls will come out far ahead in mean mathematical ability, and Iceland and New Zealand are ahead of the curve but haven't got the job done yet.
But that doesn't necessarily mean that, for example, we'll start seeing women dominating the leading edges of mathematical research, or filling an equivalent share of posts in prestigious mathematical faculties, or winning math's highest awards, like the Fields Medal, because the people who do those things aren't average. The extremes depend less on the location of the mean,and more on the value of the variance.
In many, many areas of physical, cognitive and behavioral characteristics, it has been found that men have a larger variance than women do, and therefore that extreme outliers are almost always men. This could be part of the reason that only one woman has ever won the Fields Medal, because extreme mathematical talent is rarer among women than among men, even if average ability is the same... or even favors women. It's very likely part of the reason why there are so many more men in prison than women, because criminality is correlated with cognitive deficiencies in logic and abstract thinking, and men dominate the outliers on that end of the spectrum as well.
(Note that I am not making the claim that men do have greater levels of variance in mathematical ability than women. That's an empirically-testable question to which neither of us know the answer. Men have greater levels of variance in many areas, so it's a plausible hypothesis that they do in math -- or programming -- as well. I'm accepting the hypothesis as a hypothetical, and exploring the implications if it turned out to be true. But the science needs to be done to find out if it is true.)
With respect to Google, while Google engineers are not the sort of extreme outliers that Fields Medal winners are, they are hardly average people. Programming talent is not the norm in the human population (most people find it very hard to think that way), and it may well be that competent programmers must come from the upper tails of the bell curve in several cognitive skills (and possibly the lower tails in other areas). It so, we may never see gender equity in the field without extreme affirmative action, because if, say, 5% of men have the characteristics needed for the job, it may be that only 1% of women do, even there is no difference in mean ability. And when you narrow that field to, say, the best 1% of programmers, the gender ratio would skew further since you're reaching farther from the mean.
I wonder how this compares to Google's approach to speeding up ML, the Tensor Processing Unit, and whether the ideas can be combined for even faster learning.
It's all old, tired arguments that have been comprehensively refuted before.
I don't think this is true. Note that I'm strongly pro-diversity and think that affirmative action makes sense, so I'm not arguing in favor of the doc's author's conclusions, but I note that all of the responses keep saying that his science was wrong... and as far as I can tell it's not. The doc itself has pretty extensive citations to current research, and the only responses I can find from working scientists in the field all state that he got the science right, with minor errors at most.
The conclusions he drew from the science are bad, because the gender and racial differences science has found are probably too small to produce the results he attributes to them.
I am curious: does that include discrimination against those protected classes in the job interview process? Like, say, for example, ageism?
Ageism isn't really an issue at Google. Most of the employee population is young-ish, because the company hasn't been around long and does most of its hiring straight out of school. Not all, though, and there are many successful older employees. I'm nearly 50 and work for Google, and although my current team trends young (though I'm not the oldest), my previous team included several guys in their 60s and one in his 70s (who didn't need to work, but liked it).
That said, I would agree that the interview process does not necessarily value the things that older engineers bring to the table that youngsters do not. In addition, older engineers may need to brush up on their CS fundamentals before going into the interview, while recent grads will be fresh. Assuming equivalent CS fundamentals knowledge, I'd say that the playing field is quite level, but it is a playing field that is aimed at testing the things that both can do and specifically ignores the value of experience, which younger candidates cannot match.
The value of experience is factored in when it comes time to set salary, of course. And it's definitely taken into account when putting people on teams and handing out tasks.
Didn't really matter because they all cost about the same regardless. I suppose you could say they were price fixing, or colluding or whatever.
In most places their prices were set by the taxi authority, by regulation, mostly to make sure you didn't find yourself getting ripped off. This was necessary precisely because there was no good way (unlike Uber) for you to know in advance what your price would be. In the modified approach I suggest, in which drivers set their own prices, you'd still be able to know what price you'd pay before getting in, so there's no need for central control, either by Uber or by government.
What makes Uber stand out as well is their policy on how old the car can be, so you don't end up getting a lift from some ancient beat up car.
That is nice. Uber could maintain that requirement while allowing drivers to set their own prices. That policy wouldn't be incompatible with true independent contractor status. Alternatively, the app could give you vehicle quality rating information as well as price, and you could choose whether you wanted to get in a beater for a lower fare. That might make the user interface unusably complicated, though. It would require testing.
It's all old, tired arguments that have been comprehensively refuted before. For example, he states that women are more neurotic and less able to deal with stress. We know that isn't true, because we have studied it in great detail.
I'm not so sure about that. https://www.ncbi.nlm.nih.gov/p..., https://www.ncbi.nlm.nih.gov/p....
Of course, as the first of those abstracts points out, variation among individuals is far higher than the measured difference between genders, so it's hard to see how this can have a large effect.
You're forgetting the program in which applicants who were declined by hiring committee could be sent back through for another chance. And this was mainly used for applicants from underrepresented minorities. Created a huge stir when someone suggested it was lowering the bar (even though it obviously is, even if only slightly).
I didn't hear about that one. I wouldn't be surprised if it also applies to candidates from small schools though. The sense I get from talking to the recruiters is that "diversity" is a boolean label that is applied to a candidate, and then a whole set of additional options kick in.
everything else in the paper are legitimate concerns about "how" diversity is being enforced.
One of his claims about how diversity is being "enforced" is factually erroneous. He says that Google offers programs which make it easier for "diversity" candidates to get hired, and claims that those programs are only open to women and minority races.
As it happens, I work for Google, and do university outreach to my alma mater, which is a small four-year commuter university quite different from the big-name schools that lots of Googlers attended. Because the school is one that Google typically does not get many applicants from, or many many hires from, all students at my university are considered "diversity" hires, regardless of their race or gender, and are eligible for the diversity programs. I believe the same is true of almost any small college or university in the country that doesn't traditionally have a strong CS program.
In addition, it should be pointed out that Google does not "lower the bar" for these diversity candidates. Instead, the programs provide additional mentoring and development through a summer internship (for students) or a one-year contract (for graduates). The programs include education as well as opportunities to work with regular employees on real products. At the end of the contract, the participants who want to attempt to convert to full-time employees go through the regular interview process and are scrutinized according to the same standards as anyone else. I understand the programs are fairly successful, and a high percentage of participants convert to regular employees.
First, I'll mention that I'm old enough that I used to navigate all the time without GPS. I traveled a lot for business from the mid 90s until about 2010, and didn't start using GPS navigation until ~2005. Almost every week for a decade I was finding my way around a different city, getting from the airport car rental location to the hotel, to the customer site (or sites) to restaurants, etc., all with nothing more than paper maps and road signs. So I know that not only is it possible, it's not even hard. It does require having the paper maps (rental car agencies still hand them out on demand) and it requires more preparation and more time -- and if you don't want to be late also generally involves arriving at destinations fairly early, because although you can estimate distances and times from the map, you know nothing about local traffic and road conditions.
Now, I use Google Maps even when I know exactly where I'm going, around my own home town, because I like to see the computed ETA. It's way, way better than I am at guessing travel time, especially since it factors in real-time traffic. On occasion doing this has saved me a bunch of time, too, because it has routed me around accidents and road closures and whatnot.
All in all, I greatly prefer GPS navigation over paper maps. About the only thing I think I've lost is that with paper maps you always end up with a sense of the relative locations of things. With GPS navigation you can travel a city for weeks without ever learning your way around.
I have yet to see a good way on a small-screen phone or GPS to answer questions like "how far away is the coast,"
If you just want a rough idea, pinch zoom out to where you can see the coast, then judge the distance from the little distance scale. If you want a more precise idea, it's probably because you would like to know how long it would take to get there. Pick a reasonable point and ask the phone to navigate there, and you'll get a very precise answer, both distance and time. Better than you're going to get from a paper map.
"how much out of our way would it be to go to that town"
Now that Google Maps supports searching along your route, this is easy, and the answer is very precise. I just say "Okay Google <town name>". Maps shows me the city center, with a notation saying how much time it will add to my trip to go there. If the town is big enough that I don't want the city center, I have it search for gas stations in the city. Maps will find a list of gas stations in the town and near my route, and show me all of them marked on the map, with a notation indicating how much time it will add to my trip if I go to each. This works fine even if the town in question is off-route.
what's the next sizable town within 2 miles of the road we are on
This one is more difficult, but can be done with a little zooming and scrolling to look along the road/route. The trick of searching for gas stations can also be used (without specifying town name) or modified to find something else that indicates a sizable town. Odds are you don't want a "sizable" town anyway, but instead are looking for some specific amenity: restaurant, hotel, golf course, etc., which you're not likely to find in a town that's nothing more than a wide spot in the road. So search for the amenity and see what comes up.
Unless you can then resell the token for real world currency
You can.
Have you not heard of gold farmers?
However, farmed gold sells at a lower exchange rate than WoW token dollars do going the other way, due to the risk created by the fact that selling gold for cash is a violation of the terms of service, so the buyer is taking a risk buying farmed gold.
On the gripping hand, though, at the rate the Bolivar is going it'll drop below the value of a farmed gold piece in a few months.
Regenerative braking can recover close to 100% of the energy.
True. But even if some of the energy is lost as heat, none of it creates any brake dust.
Based on what I've been reading Tesla currently only puts regenerative brakes in the rear. Since most of your brake loading is on the front wheels, I doubt this reduces pad wear by very much.
Your doubt is misplaced.
EV drivers tend to use nothing but regeneration for the vast majority of braking. This requires driving a little less aggressively and "coasting" most of the way to the stoplight, engaging the brake pads only at the end. Yes, this means that most braking is done only with the rear wheels (on a single-motor Tesla), but we're talking about normal, gentle braking. Aggressive braking is where you need to make sure the braking force is appropriately distributed, and that works the same as an ICEV.
None of that matters, he wasn't hailing the virtues of Visa's transaction handling, he was pointing out their transaction rate as a guide to what is required to meet the needs of commerce today. BTC is nowhere near where it needs to be if it is to take over.
I wasn't commenting on BTC, just correcting a common misunderstanding of how credit card transactions work.
If you want a comment on BTC, though, I'll say this: The only way to scale to very high transaction volumes is by decentralizing. BTC's problem is that it is too centralized, which seems an odd thing to say about a fully decentralized transaction ledger, but it's true. The problem is that there's only one ledger, and all copies of it must be synchronized. I can think of a few different ways this could be solved, so the problem isn't inherent in blockchain-based cryptocurrencies, but only represents a flaw in the design of this one.
Yeah, like that fucking worked with metered taxi's.
Completely different situation. With metered taxis riders couldn't choose from all of the taxis within a large radius of them, and couldn't know the price they'd pay for their trip before they selected the cab to take.
Yep, it's risky. I think they'll make it, and I expect a handsome reward when they do. Or I may lose a few thousand dollars if they don't. That's the game. The bulk of my retirement portfolio is in safer, more consistent bets, of course.
Is that really such a big factor?
Yes.
There's an extremely common point of view among car-buyers, where it simply doesn't matter how awesome it is; $35k is way too much money for a car.
$35K is the average price for a new sedan. It may be above your price point, but it's not above the price point of most buyers of sedans.
It's explained reasonably well by analogy with the Parable of the Ox
No, it isn't. At least, it isn't if we're talking about shares in real companies, or quantities of real commodities.
The parable starts out pretty good, but where it breaks down is in its assumption that the scale goes away completely. In the parable, the weight of the ox becomes entirely divorced from any sort of reality, but that doesn't happen in securities, because there is an underlying value in the security: it's future earning potential. The parable really needs to add some notion that the ox, or parts of it, will eventually be weighed, actually weighed, not just acceptance of an average value of guesses, and incorrect guesses will lose at that point.
For example, for the last couple of years I've been investing money now and then in shares of TSLA. At this point, I own a few thousand dollars worth of the company, based on the current consensus view of what the whole company is worth, divided by the number of shares in total. Now this consensus total value is many, many times the value of Tesla's actual assets. And it's also significantly higher than most other common methods of estimating value, based on revenue, income, etc.
Why is it high? Because lots of people guess it should be? Sort of... but not really. And we will eventually find out whether that high valuation was or was not correct. The ox will get weighed. How?
It's high because there is a possibility that Tesla will experience explosive success, generating massive revenues and profits that will replay the relatively high cost of investing now. Pulling the price down, though, is the possibility that Tesla will fail to achieve that level of success, but just continue being a niche carmaker. Pulling it down some more is the possibility that the EV and solar home markets will never really pan out, or that Tesla will fail to execute well on delivering its promises to its customers. In either case, it could crash and burn, becoming worth a tiny fraction of its current value, or even going bankrupt. I owned one stock that followed this path, becoming completely worthless.
There's an underlying business that will succeed or fail, and in doing so will justify some current stock price. We don't know what that currently-justifiable price is, which is where the uncertainty and the guessing come in. But, as I said before, the ox will be weighed. There is a core reality underlying the security whose value we're guessing; it's not pure guess-averaging.
There is no objective measurement of perceived value, which is what the economy runs on.
Nonsense.
You absolutely can objectively measure value, in both individual and aggregate cases. The measurement will not be a single number, it will be a probability distribution, and the distribution will vary over time, but that makes it no less a real, objective measurement. What you're measuring is human demand and desire, and that is a real, objective thing, even if it's variable across the population and across time, and even if it is hard to measure accurately.
If you measured the value of a slice of Wonder bread, you'd find the distribution has a mean on the order of a few pennies, and the vast majority of potential buyers would be willing to buy it for that. There is a non-zero probability you can find someone to pay $10K for it, but it's extremely small unless there's something unusual about your particular piece of Wonder bread. If it's fungible with all of the other pieces of Wonder bread, then the probability approaches zero.
Basically, you're confusing "objective" with "fixed and universal". Objective measurements need not have single values, nor do they have to remain fixed over time or space.
In the case of business valuations, unlike goods there's very little variation (after risk adjustment) across the population, because people rarely have individualistic reasons for wanting to own shares of one company over another. People buy stocks because they hope to make money. On the other hand, business valuations are inherently rooted in predictions of future business prospects. That doesn't make them so much "subjective" as it does "uncertain". Different analysts will assign different probabilities to different outcomes, because forecasting is an unpredictable business. Also, new facts will change the probabilities.
However, aggregate, broad-scale predictions, made by large groups of people tend to produce fairly good results... and that's what stock prices are. Aggregated predictions.
What this study found is that the aggregated predictions of small groups of private investors do not match the aggregated predictions of the public markets, and that the private investors systematically overvalue the companies, as compared to the public markets. Does this mean the private investors' predictions are wrong and the public markets' are right? In one way, yes, by definition, because the private investors ultimately cash out through the public markets, so the private investors are really trying to predict the valuations that will be assigned by the public markets. In another way, no one knows, because the actual value of the company depends on what happens in the future. On average, though, it's likely that the public markets make better predictions.
You're assuming that people in your area can only sell local. Why do you assume that? Some businesses are inherently local, sure, but you don't need to make a choice to support them, because you can't buy remotely anyway. But many businesses are globalizable, and that applies to both buyers and sellers.
Your numbers are a little out of date. Visa last year alone averaged 4500 transactions a second.
It should be pointed out that "Visa" isn't a transaction clearing system. It's a bank association. Large numbers of transactions are easily handled because there is no single system that all of the transactions flow through. There is a set of card issuing banks, a set of merchant acquiring banks, and a set of clearinghouses that connect them -- for the cases that the acquiring banks and issuing banks don't just connect directly, which they often do.
The Leaf is being heavily discounted right now, which is why I'm even considering one.
If you're not getting the 2018 LEAF with its bigger battery, I suggest looking for a used car. 2012-2014 LEAFs are crazy cheap right now. The main thing to watch out for is the battery capacity. Ideally, you want one that still has 12 bars on the capacity display. If you can do with less capacity, fine, but the price should drop accordingly.
I recommend getting a Bluetooth OBDII adapter and the LeafSpy Pro app on your phone. Plug in the adapter and the app will show you the total battery capacity with a high degree of accuracy. After you've found a car that seems good mechanically and in appearance, and has reasonable battery capacity for its price, take it to a Nissan dealership and have them do the free battery analysis.
Unlike with ICEVs, there's very little to go wrong with the drivetrain. Electric motors last forever, there is no transmission to speak of (just a reversing gear). Take a look at CV joints, etc., just like you would an ICEV. Brake pads wear out, but slowly, and replacing them is the same as an ICEV. So really it's just the battery you need to pay attention to when evaluating a used EV.
The Leaf's lack of a TMS causes their battery to degrade rapidly, losing as much as 40% of its capacity in 2-3 years
This depends on where you live. My 2012 LEAF has 50K miles on it and has lost only about 4% of its battery capacity. (I'd go check it right now with my OBDII interface and LeafSpy Pro, but my wife has it.)
Get a used Volt. They're ass-cheap. Just don't buy a Leaf
Used LEAFs are much cheaper. You can get one about like mine (note: I'm not selling mine; I like it) for around $6K. Assuming you don't need more than 60 or so miles of daily range, and don't live in an area with a very hot climate (which causes rapid battery degradation), for $6K you can get an EV that will be a great commuter and around-town vehicle for several years, and will cost less than a nickel per mile to operate, including electricity and maintenance.
Unless you live in Arizona,or the like, they're great little cars, and very, very cheap right now.