From deciding to add range, it wouldn't take them more than 2-3 years to release a product.
The only car with range at all comparable to a Tesla is the Chevy Bolt, which GM sells at a $9000 loss.
Tesla will be looking at real competition, someday, but I'm not seeing any signs that it could be as soon as 2-3 years from now.
various restaurants and shopping malls in my area have charging spots with J1772 spots
J1772 will provide 16 to 20 miles of range per hour spent charging.
Let's imagine I wanted to drive from Seattle to Spokane in a Nissan Leaf. Google Maps says that's about 279 miles, and the Leaf has a rated range of 100 miles. I want safety margin so I want to charge every 80 miles, and let's assume I have no problem finding a J1772 exactly when I want one. And let's assume that every J1772 I find gives me the full 20 miles of range per hour spent charging.
Drive for 80 minutes. Spend four hours charging. Do it again; that's 160 miles done. A third time, that's 240 miles done (but I guess I could charge as little as 3 hours instead of 4). Then, finally, I would arrive in Spokane. It's under 5 hours of driving, yet would require about 11 to 12 hours of charging time.
With a Tesla: the car can make it with one charging stop, and that stop can be under an hour. I can eat lunch while the car charges. Spokane has a Supercharger so I could top off the car on arrival and have plenty of charge for my actual visit to Spokane.
This is why I say that a Tesla can be used for long trips, while other electric cars really can't.
Note that a Chevy Bolt could do the trip to Spokane with only one charging stop of 2 to 3 hours. Note also that there are other fast charging standards like CHAdeMO but they aren't built out into a full network like Tesla Superchargers (I don't know if you can find CHAdeMO between Seattle and Spokane; there are two Supercharger stops, and Tesla cars only need one).
This will be the model of electric car charging for long haul trips: charging capability as a feature of a restaurant, not "just like a gas station".
Tesla already has a program for "Destination Chargers". A restaurant can get a Destination Charger for free from Tesla, as long as they let Tesla owners use it for free. This is the same charger hardware as one can install at home, and on an 80 Amp circuit such a charger can provide 50 miles of range per hour spent charging. Even the Destination Chargers are better than J1772, let alone a Supercharger.
I wouldn't be so confident that investment will become durable first mover advantage for Tesla
Well, I am confident.
Right now, almost all electric cars are great "second cars" suitable for short trips and terrible for long trips. Tesla cars, combined with the Tesla network of Supercharger stations, are okay for long trips. A car that can drive for three to four hours, then recharges in the time it takes you to eat a meal... that's not significantly worse than a gasoline car for long trips.
A Tesla really can be your only car. A Nissan Leaf can be your only car... if you don't mind renting some other car when you want to make a long trip.
So you have all the other electric cars, and Tesla. There really is no substitute, yet. If Tesla can make enough Model 3s, they could have a few million on the road before the Big 3 car makers can really come up with serious competition.
I don't know if Tesla can dominate electric cars forever. However, I am convinced that their worst-case scenario is to become the Apple of car makers, having carved out their own niche and with fanatically loyal customers. Apple can't sell more phones than the Android competitors combined sell, but they still sell plenty of phones and make a solid profit doing it. And Apple sure made a lot of money off their first-mover advantage. (Even though smartphones existed before the iPhone, Apple made the first smartphone that didn't suck, and I claim that Apple got first-mover advantage. Then they did it again with the first tablet that didn't suck, the iPad. The iOS ecosystem remains tremendously profitable even though Android devices now are able to compete.)
So I'm prepared to imagine the Big 3 actually competing with Tesla... but they haven't even started yet. For anyone to really compete with Tesla they are going to have to spend big. GM loses $9000 per Chevy Bolt... maybe if they built their own battery factory, like Tesla did, they could sell Bolts at a profit? But they don't have their own battery factory, which may explain why Elon Musk is really not worried about competition from GM or the others.
And the Tesla network has over 1200 Supercharger stations worldwide, with more being opened. The Big 3 can't just snap their fingers and have an equivalent network overnight. But in the long run, I expect they will build out something comparable to what Tesla already has, and be more able to compete with Tesla. Again, they haven't even started.
I believe it is fair to say that Tesla really does have battery production in-house. It's a significant reason why Tesla can make $10,000 on a Model 3 while GM loses $9,000 on a Chevy Bolt. (Note: making $10K per car will require Tesla to get production rates up, as the major expense is depreciation on the factory, and currently the depreciation expense is spread over 2000 cars per week instead of 5000 cars per week.)
This is very different from the way Amazon ran at a loss for a long time. Amazon was building retail sales network in a totally new market. Tesla is just selling cars to the small number of millineals in the 1%.
Tesla is just selling a small number of cars!
Well, and inventing a better battery pack technology than anyone else has.
And building out the best car charging network, period.
And building their own battery factories to get their costs down.
And building out their factories. (A company like GM has spent decades building out its factories, but Tesla is a new company and is building from nothing.)
And investing in R&D to invent profitable new things like the Tesla semi-truck.
Hmm, maybe Tesla is doing more than just selling a small number of cars.
It's fun to watch, and if the stock dips down again I'll buy some more.
Tesla is a "buy and hold" stock in my opinion. Tesla has been doing everything possible to build for the future. I frankly don't care if Wall Street gets personally annoyed with Elon Musk's antics and the stock price dips. Nothing fundamental changed and the stock price went right back up again.
It's hard for me to imagine how an algorithm could really do a good job of "picking the next big blockbuster". On paper, the Justice League movie looks like it ought to be about as good as the latest Avengers movie, yet the Avengers movie is way better (compare IMDB ratings histograms: Justice LeagueInfinity War). Zack Snyder's movies have made a lot of money. Joss Whedon has been associated with very successful movies like the first two Avengers movies. The actors have made very successful movies. The Internet buzz was huge for Wonder Woman. (And the Internet buzz for Ben Affleck was mostly that he was going to suck as Batman before he made Batman v. Superman, and very favorable afterward.) I can't think of any simple inputs one could feed into a model that would have predicted how poorly Justice League did overall.
The Wonder Woman movie did very well, and so did the first Guardians of the Galaxy movie, and so did the first Deadpool movie. All three were surprises: I didn't see any predictions ahead of any of them "this is going to be huge." I'm going to claim that what all three had in common is "heart". And it's hard to quantify heart.
Another possible ingredient of a blockbuster: show people something they haven't seen before, that they want to see. Before Deadpool, nobody knew that they wanted to see an R-rated Marvel movie with a character who breaks the fourth wall. Before the Guardians movies, who would have predicted that "I Am Groot" would be so popular or that Chris Pratt would become a really huge star? Supposedly Henry Ford said "If I had asked my customers what they wanted, they would have said 'a faster horse.'" You can't guarantee you will give the people what they want by giving them what they used to want, or even what they say they want.
It's easy for an algorithm to pick something that nobody has seen before, but it's hard to guarantee that anyone will want to see it. I don't think it's really possible to quantify that for the algorithm.
Another example that comes to mind is Grosse Point Blank, which did very well when it came out. Or Napoleon Dynamite which just came totally out of nowhere.
It's my personal bias showing, perhaps, but I think the only meta rule covering all the above successes: people put heart into making those movies, and didn't have clueless studio execs stirring the pot and forcing changes that make the movie more like other movies.
I am hoping that the costs of making a movie will continue to come down, and in the future, studio executives will give more of a free hand to the people making a movie. The more a movie costs, the more the studio fears risk, but a true blockbuster can't play things completely safe or it risks being bland.
Note, heart alone doesn't guarantee success. There are plenty of movies that are now beloved that were flops when they first came out. There is a timing element. Also, some people just aren't good at making movies; they may put their whole heart and soul into a movie and the movie could still suck.
In the past I was not interested because it didn't even run on real hardware; you had to run it inside a virtual machine. Checking their web site it seems it does run on some real hardware now, but only some devices are supported. Actually that is great progress and I hope that it will attract more developers.
It's not that fun to work on a project when it's super primitive and everything is broken. When it works a bit and just needs a tweak here and there, more people will be interested in working on it. I hope that will be the case for ReactOS.
I would love to have a Windows-compatible system that doesn't phone home constantly and can run some of my favorite games. It will take a while but it's starting to look like they will get there.
Tesla bought its factory super cheap: $42 million. That price is so cheap that essentially Toyota was investing in Tesla. That factory is in California, not Michigan.
Tesla's big cars have been extremely successful (they simply took market share away from other luxury car makers). The Model 3 is selling as many as they can make, and they are selling only the most expensive options for it right now. A year from now I expect there will be almost 200 thousand Model 3 cars out on the road. So overall I'd say that things are working out pretty well for Tesla.
I am.. blown away that people believe that the three laws are a good idea. completely astounded.
I think you are overstating things a bit here.
The Three Laws are a convenient shorthand for saying "a comprehensive set of safeguards governing the operation of a robot." In the web comic Freefall people simply talk of "safeguards" rather than some arbitrary three laws.
The Three Laws have a huge place in the history of SF because they represent a sea change in how robots were presented. According to Asimov himself, in forwards to collections of his robot stories, before Asimov formulated the Three Laws robots were presented as dangerous things that generally went out of control to drive a story. He reasoned that people try to make things safe, and robots would be no different; people would incorporate safeguards, and his Three Laws were his take on a minimal set of safeguards.
Asimov then spent the rest of his career gleefully finding corner cases where the Three Laws were inadequate, and writing stories about what happens when those corner cases are hit. He was the first to promulgate the Three Laws idea and also the first to poke holes in them.
If someone really is arguing that the Three Laws are perfect and ready to implement, that shows they haven't researched the subject well and you are justified in being scornful of their shallow grasp of the subject. But if someone is talking informally and saying something like "robots should be required to have safeguards like the Three Laws" I have no problem with that, even if they phrase it less carefully and say something like "the Three Laws should be mandatory."
Also, when Asimov first wrote these stories, he overlooked two things that I consider hugely important. First of all, griefers. In his stories, any human could give an order to a robot and the robot would obey as long as no human was harmed. So a griefer could order an expensive robot to go smash a bunch of parked cars, ruining the cars and the robot, and (Asimov used this in his stories) the griefer could tell the robot "if you reveal my identity, I will come to harm" and it would be impossible for the robot to name the griefer. (It would also be possible to order "smash all these cars, and then forget ever having seen or talked to me.") The other thing is that Asimov imagined that it would be extremely difficult to make robot brains that did not include the Three Laws, which seems quaintly naive to me. If there is still a North Korea when robot brains are invented, there will be a secret project to make robots capable of serving as loyal soldiers, which means robots that obey Dear Leader's orders without question (no other safeguards included). As Jerry Pournelle used to say: "What man has done, man may aspire to do."[1] The existence of robot brains will be proof that a new robot brain design is possible.
P.S. Another classic of the robot genre is "With Folded Hands". Robots have the prime directive: to serve, and obey, and guard men[1] from harm. The robots ultimately enslave all of humanity in a smothering protective embrace: anything a human might want to do, like rock climbing, could be forbidden as too risky. Any human who resists this benevolent enslavement is lobotomized so that he/she will stop resisting and just enjoy life. I think in later stories the robots supervised even sex, on the theory that you could have a heart attack or something from the exertion, so the robots only allowed sex by young people, and only so there would be another generation of humans to serve.
Finally, for a modern take on artificial intelligence with inadequate safeguards, read the Torchship trilogy by Karl Gallagher. In these books, about a dozen whole planets (including Earth) have no living humans anymore because AI-controlled machines killed them all. In the stories, the historical events where the AIs went berserk are referred to as "The Betrayal".
i never understood why windows defaults to "hide extensions for known file types"
I agree it was never a good idea. IMHO it was done because the file extensions are ugly.
On the Mac, every file has a "resource fork" (I guess on OS X it's no longer properly a "resource fork" but there is something equivalent) and the type of the file is coded in a spot that the OS knows how to read but which the user doesn't see. So the user types any name, and the icon is the right icon and the user just sees the icon and the chosen name.
On Windows and Linux, the file systems don't have this "resource fork" idea, so the obvious place to encode the type of file is the extension. But the extension is user-visible.
Linux uses the techniques pioneered in UNIX to just identify a file no matter what its name is. If it's an ELF binary, it will start with certain bytes arranged a certain way; if it's a LibreOffice document, it will start with different bytes, etc. It's trickier but more reliable: you can rename a LibreOffice document to not have its extension any more, and your file manager can still do the right thing when you double-click on it.
But Windows just uses the extension.
Well, hiding the extension makes Windows more like the Mac. The icon is correct, the user just sees the filename chosen by the user, life is great.
But users are used to seeing extensions and don't worry about them much. And there was a form of attack where a Trojan Horse file would have a name like "Important Document.doc.exe" and hope the user would open it. If Windows hides the extension, then just the ".exe" part is hidden, and the user just sees "Important Document.doc" (and as I said the user is used to seeing extensions and doesn't freak out that most documents have no visible extension but this one does).
These days, by default, Windows hides "system" directories and anything else that an uninformed user shouldn't touch. If I have to use Windows, I make sure to turn on seeing file extensions, disable hiding system directories, etc.
Especially the blurring of code and data (a la Lisp) -- a 'bolt a feature on' response is nigh-on-impossible expressly because that vast range of python libraries won't work lisp-like code/data ambiguity.
Ruby and Python are more similar than different: they are C-like or Algol-like scripting languages that have both object oriented and functional programming features. So I'm honestly confused that you are distinguishing them in this way. Could you please give me an example of something really LISPy that is possible in Ruby and not possible in Python?
Python has support for explicitly compiling a string into code objects, and then Python code can introspect and rewrite the code objects. This is much more difficult that the similar operation in a LISP because S-expressions containing tokens are much easier to work with. And I'm guessing this isn't the sort of hacking you mean.
There are two things about Ruby that I am aware of that are more hacky than Python and that you might like, but neither seems that LISPy to me:
First, because Ruby is very generous about what you can use in an identifier and will implicitly call functions, you can hack up a DSL that's really just Ruby in disguise. For example, there is a DSL called Cucumber that defines things like Given: and because Ruby allows punctuation like ':' in identifiers, and doesn't require parentheses, you can implement this by writing a function and naming it Given:. But Python has library modules like PyParsing, which IMHO is a better solution: just specify the language you want and the callbacks when the parser figures out what is being specified. It's a bit more up-front work but it's a cleaner and less hacky way to go. (And thus my bias as a Python user is exposed... I prefer the explicit solution even though there is more up-front effort as I think it will result in a cleaner solution with less total work in the long run. Ruby users may disagree completely.)
The other thing is that Ruby lets you extend the fundamental objects for things like integers. You can access the global object for integers or whatever and add new method functions or override built-ins. What is Monkey Patching in Ruby? This just freaks me out. Python lets you do whatever you want by making a new subclass: you can make a new class that is just like an integer but has some behaviors overridden or new behaviors added. That to me is the One True Way... builtins like integers should be your bedrock, totally predictable because they can't be overridden. Python is less permissive than Ruby, but even Python is more permissive than I would like... Python doesn't have a const feature built in, so you can't do this:
const FOO = 1
You can simply do FOO = 1 but other code could clobber FOO with a different value. So in Python you have to just learn to not clobber things you don't own, and in Ruby I guess you just have to not break integers for everyone.
By the way, it is possible in Python to make a class and then override assignment to members of that class. So you could make a Constants class, assign all the constants as member variables, and then override assignment to the class so that it raises an exception. But in my experience people just set up constants in modules and then try not to clobber them.
I suspect the scientific community will eventually shift over to Ruby over Python
Unlikely. The reason the scientific community is using Python is that Python has SciPy, a rich and powerful collection of libraries. The heavy lifting in SciPy is done by compiled Fortran library code. Right now a Python program using SciPy is nearly as fast as the same program written in Fortran, and Python is dramatically easier to use. And it probably doesn't hurt that Python added a matrix multiply operator (infix @), just for the benefit of SciPy users.
In science and engineering, Python is now benefiting from network effect, where everyone uses Python because everyone else already uses Python. For Ruby to steal these users it would have to do something dramatically better and to date Ruby hasn't even matched Python. And if Ruby did get an edge on Python I predict that Python would implement something similar and keep its position as the language for science and engineering.
If Google was so damn generous, they'd donate laptops with 4G modems and access
You heard it here, folks: b0s0z0ku has spoken, Google's gift isn't generous enough to meet b0s0z0ku's standards.
Note that TFA says this is an expansion of an earlier pilot program that has been working out. But according to b0s0z0ku's thought experiment, it won't work. Silly Google!
P.S. I get carsick in buses too, but I'm not going to snark at Google over this. Maybe school-age kids can use a Chromebook in a bus without getting sick, or can learn to. Also Chromebooks can do some things without Internet. Bottom line is that the kids have more equipment and more options than before Google made the donation.
Lakaysha says the program has helped her personally. Also the article I linked there says that the program includes parking the buses in public places, like community centers, so that some students can use the WiFi service from the bus even after the long bus ride is finished.
Actual evidence the plan works beats a thought experiment that the plan doesn't work.
OLPC still exists, but is now irrelevant. The project is almost dead.
First, they failed to hit their $100 target. The laptop cost roughly double that.
Second, their program focused on the wrong strategy. They tried to make the perfect device but didn't focus enough on volume. They should have made a laptop that they would cheerfully sell to anyone for $100, and shipped hundreds of millions of devices. Instead they made a somewhat boutique device that cost $200 and they wouldn't sell it to you unless you paid $400 for it. (Under their "give 1/get 1" program, you would pay double for a laptop and OLPC would then give a laptop to a student somewhere.) The boutique strategy didn't work out.
Then they spent time trying to design some new devices that never went anywhere. (XO-2XO-3)
We now live in a world where you can get an off-the-shelf Android tablet for $40. Therefore you can get roughly four tablets plus four USB keyboards for a similar cost of a single OLPC device.
I respect the OLPC project's ambitious design goals. A laptop that is rugged, can work outdoors, is repairable, and has mesh networking features, running nothing but free software! Neat! But compromising on some of these details could have lowered the price and the project might not be irrelevant now. I'd like to see some statistics on how often the mesh networking is actually used, how often schools actually repair these devices.
Around 2012, the OLPC project tried releasing a special OLPC Android tablet for $150. I can't find any information on how many they sold, but I don't think that really worked out either.
At this point I think the best strategy would be to just write educational software to run on Android tablets, and assume the market will take care of making the tablets.
P.S. I personally paid $400 for the original OLPC laptop. I found the thing to be frustratingly slow and hard to use. (In fairness I routinely use computers that cost way more than $200, but even so...) The worst part was the touchpad; I found it wildly inaccurate so using it was frustrating.
Also, I was looking forward to hacking the thing; I wanted to hit that "Show Source" code keyboard button, see some Python code, and make some sort of improvement. I found that most of the time when I hit the "Show Source" button it didn't do anything and my urge to contribute died.
In the end, I donated my OLPC to a church group, to send to a school in a very poor part of India.
I used to use a Palm Pilot to read books and run various programs including games. The display wasn't great but performance was great (you never had to wait for the thing to respond to a click) and the battery lasted a very long time. (If I remember correctly I got about two weeks of life from a pair of AA cells. I switched to using rechargeable NiMH AA cells, and still got days of use before needing to recharge.)
IMHO the OLPC project could have made a tablet device similar to a Palm Pilot, but with a much larger and higher-resolution monochrome screen... and hit their $100 price point. Such a device would be useful for running educational software and very usable as an ebook reader. In particular the long battery life would have been a huge win compared to the actual OLPC hardware. Such a device shipped in the hundreds of millions of units would have had a much higher chance of changing the world. It could have been offered both as a stand-alone device, and in a nylon case bundled with a USB keyboard (kind of like the Apple Newton case). In t
Are non-Tesla entities allowed to make supercharger stations?
Tesla has offered to license all its patents so it might be possible.
What is 100% possible right now: businesses that want to offer Tesla charging for their guests can get a "destination charger" from Tesla. As I understand the deal, Tesla gives the charger for free, as long as the business offers the charging for free. So the only cost to the business is the cost of the electricity.
Also, anyone could buy a Tesla home charger and set it up. I guess they could charge for using it.
A home charger can charge up to 50 miles of range in an hour. That's roughly one-quarter of the speed of a Supercharger. (If a Supercharger is really busy and two cars have to share one charging circuit, the Supercharger might be slow enough that the home charger is half the speed.). I believe a destination charger is the same hardware as the home charger and this the same speed.
Also, Tesla sells an adapter that allows a Tesla car to charge on a CHAdeMO station, and Tesla doesn't control those at all. That would charge at about 100 miles of range per hour. If I were wanting to run some kind of competition to Tesla in charging I would look into that.
The strategy of EEE (embrace, extend, and extinguish) works like this:
Microsoft starts offering something that people want, but "extended" in an incompatible way so that if you start using the Microsoft offering you get "addicted" to it and cannot switch to any other version. Because of Microsoft's huge share of the market, the "addicted" Microsoft customers greatly outnumber the people using the original version, and Microsoft gains de-facto control of whatever it is.
As an example, Microsoft added Windows-specific extensions to Java. Apps written that took advantage of these extensions would run only on the Microsoft extended Java. (Sun Microsystems sued Microsoft over this, and in response Microsoft scrapped the whole thing and stopped shipping Java, and then went on to create C#. Also, hilariously, Sun Microsystems sued Microsoft to force them to start shipping Java again, but did not win.)
Now, if you think for a minute, it's clear that this strategy can't touch Linux.
Linux already rules the world. It runs Google, Amazon, all the supercomputers, ISPs, firewalls, routers, all the IOT devices, and every non-Apple smartphone or other mobile device. How is Microsoft going to make an EEE version that is more popular than the non-EEE version?
Also, what is the "secret sauce" that Microsoft could add that would lock people in to their version? They could add some kind of Windows integration, which would make their version the best choice for Windows users... and all the current Linux users will shrug and pay no attention at all.
There is absolutely no danger of Microsoft even trying an EEE strategy, since Microsoft knows it wouldn't work.
It doesn't matter how much you think you should get paid for doing something, if I'm doing anything, I need to get paid a certain amount just to live.
True, but that amount varies depending on your situation. If you are a teen living at your parent's house, you don't need as much as someone trying to support a family and pay a mortgage.
It doesn't make sense to declare that every job must pay a "living wage" because not every worker needs a "living wage". There is such a thing as a "starter" job, a job that has very low skill threshold needed and thus very low barriers to entry, but pays very little. The starter job is the first rung on the ladder of success. Rational workers might very well accept a $5 per hour starter job but would not stay at that job their whole lives.
The burger joint could just not pay $15/hour and instead just pay $5/hour and see what happens. Ohh, no burgers to sell, now you're out of business. Well then, flipping burgers must be worth $15/hour.
I agree completely with this sentiment, although I reach the opposite conclusion that you reached. If a place tries to pay too little, workers will not want to work there. The place will be forced to raise wages until people are willing to take the jobs. If $5 per hour is truly not enough, the place will be forced to offer more.
But the "Fight for Fifteen" movement is trying to force a $15 per hour minimum wage across the board, onto businesses that have historically not had any trouble filling their positions at an hourly rate of less than $15 per hour. Using your thought experiment, flipping burgers must not be worth $15 per hour, since free market competition doesn't produce salaries at that level naturally.
There is a natural equilibrium wage for a given industry in a given area. Offer too little, nobody will work for you. Offer too much, and you don't make enough profit and you are forced to shut down your business.
If the minimum wage is set below the natural equilibrium wage, it has no effect; people will be paid more than that naturally.
If the minimum wage is somehow set to exactly equal to the natural equilibrium wage, it will have no effect.
If the minimum wage is set above the natural equilibrium wage, it will have multiple effects that on the whole are not good. Fewer people will be hired for jobs; the employers are more likely to pile extra work onto those fewer people; and the employers are incented to try to find ways to replace the workers. The worst thing is drying up the pool of starter jobs... if there are fewer jobs, and the jobs have a forced high salaries, employers are incented not to take chances; they will hire the best candidates they can get, and not very many of those. This is hardest on the people who really need a job. It's called "sawing off the bottom rungs of the ladder of success". It's why, IMHO, if you truly want what is best for disadvantaged people, you should not raise the minimum wage.
Good workers will get more money. If an employer is a jerk and won't give raises to his good workers, then those workers can switch jobs and get more money somewhere else. (It's always easier to get a job when you already have a job.)
I'm not a lawyer but I'm pretty sure you can make something like this despite Apple's patents.
Now, Apple is making claims that their devices have good visibility outdoors, even if the user is wearing sunglasses, so maybe there is something of value in their new patents. But the patents cannot simply be "computing device using touchscreen as a keyboard" because it would have flunked the prior art test.
I am 100% confident that the car will never hit the Earth, because I fully expect that within the next couple hundred years it will be retrieved and put on display in a museum somewhere. Maybe the Luna City museum or the Ceres Museum; some Earth museum is also possible.
Right now, retrieving it is theoretically possible but such a huge and expensive undertaking that it's totally unreasonable. But if we build out our infrastructure, we will have spacecraft flitting between Earth, Mars, and the asteroids and sending a tow truck to grab the Roadster will be no big deal.
Yeah, 52 grand is pretty expensive. And by investing in your house you are making your property taxes go up, so you will pay again. Elon Musk had this comment: "The economics are not yet compelling where housing and utility costs are low and property taxes are high."
Tesla is arguing that the roof defrays its own cost by generating electricity; and if you live in a sunny area and put in enough solar cells, the roof will pay for itself (and actually return a profit eventually). But with time value of money it's not a good investment at current prices.
So, right now, this is a roof for rich people who don't mind dropping a chunk of money that will take a long time to pay back. If you are building a mansion that will cost over a million bucks, why not throw a Tesla roof on it? It would be less than a 5% increase in cost, the roof is durable and looks nice, and you can feel that you are helping combat climate change. And if your neighborhood loses power, you can still have lights on in your house.
For people like me, and you I'm guessing, this is just too pricey right now.
Remember how Tesla's first car was a toy for rich people. Baby steps. If this roof product does well, they can ramp up production volume and bring costs down.
In one of Richard Feynman's books, he told about his experiences at a university in Brazil. He was horrified to realize that the students were ritually memorizing the course material with very little actual understanding.
When he asked questions in a way that echoed the textbook, students were able to recite an answer straight out of the book. But when he made up a "word problem" they were totally unable to answer.
A student was quizzed on physics, asked to compute what happens when light passes through a diamagnetic substance, and he recited the answer correctly and then calculated the correct result given the index and thickness of the substance. Immediately afterward, Feynman talked to that same student; Feynman held up a book and asked what would happen if the book was made of glass and he looked at something through the book. The student didn't realize that glass is a diamagnetic substance, and gave a very incorrect answer.
In the domain of math questions, I saw an example: if a person has 4 boards of length 2.5 metres each, and cuts them with a saw, how many 1-metre boards can that person make? Obviously the correct answer is 8 (two per board, with 4 left-over pieces of length 0.5 metres minus the width of two saw cuts). If you were just playing with the numbers abstractly you might think that since 4 * 2.5 == 10 that you could produce ten 1-metre board segments. You can't actually glue together 4 boards to make a single board, and you can't actually make zero-width cuts.
I can't speak for others, but I enjoy word problems more than abstract problems. (Good ones, anyway... you can take a simple problem and write an annoying and confusing word problem, and nobody likes those.)
I think the point was that at the time of volunteering you had to face a non-remote chance of death regardless of the manner of your service
I agree with this.
Also, when Johnny Rico and his friend went to volunteer, there was a guy who was urging them to give it up, and in his telling all the jobs sounded really dangerous. You might end up having experimental drugs tried on you, you might end up doing tough labor on a frozen planet, etc. (By chance Rico bumped into him right after Rico got his assignment to the Mobile Infantry; at that point the man greeted him warmly and congratulated him on his assignment.)
P.S. Rico's acquaintance Carmen Ibanez volunteered at the same time, and the man smiled at her and didn't try to talk her out of it at all. "We always need pilots" he told her.
Slashdot Mirror
From deciding to add range, it wouldn't take them more than 2-3 years to release a product.
The only car with range at all comparable to a Tesla is the Chevy Bolt, which GM sells at a $9000 loss.
Tesla will be looking at real competition, someday, but I'm not seeing any signs that it could be as soon as 2-3 years from now.
various restaurants and shopping malls in my area have charging spots with J1772 spots
J1772 will provide 16 to 20 miles of range per hour spent charging.
Let's imagine I wanted to drive from Seattle to Spokane in a Nissan Leaf. Google Maps says that's about 279 miles, and the Leaf has a rated range of 100 miles. I want safety margin so I want to charge every 80 miles, and let's assume I have no problem finding a J1772 exactly when I want one. And let's assume that every J1772 I find gives me the full 20 miles of range per hour spent charging.
Drive for 80 minutes. Spend four hours charging. Do it again; that's 160 miles done. A third time, that's 240 miles done (but I guess I could charge as little as 3 hours instead of 4). Then, finally, I would arrive in Spokane. It's under 5 hours of driving, yet would require about 11 to 12 hours of charging time.
With a Tesla: the car can make it with one charging stop, and that stop can be under an hour. I can eat lunch while the car charges. Spokane has a Supercharger so I could top off the car on arrival and have plenty of charge for my actual visit to Spokane.
This is why I say that a Tesla can be used for long trips, while other electric cars really can't.
Note that a Chevy Bolt could do the trip to Spokane with only one charging stop of 2 to 3 hours. Note also that there are other fast charging standards like CHAdeMO but they aren't built out into a full network like Tesla Superchargers (I don't know if you can find CHAdeMO between Seattle and Spokane; there are two Supercharger stops, and Tesla cars only need one).
This will be the model of electric car charging for long haul trips: charging capability as a feature of a restaurant, not "just like a gas station".
Tesla already has a program for "Destination Chargers". A restaurant can get a Destination Charger for free from Tesla, as long as they let Tesla owners use it for free. This is the same charger hardware as one can install at home, and on an 80 Amp circuit such a charger can provide 50 miles of range per hour spent charging. Even the Destination Chargers are better than J1772, let alone a Supercharger.
I wouldn't be so confident that investment will become durable first mover advantage for Tesla
Well, I am confident.
Right now, almost all electric cars are great "second cars" suitable for short trips and terrible for long trips. Tesla cars, combined with the Tesla network of Supercharger stations, are okay for long trips. A car that can drive for three to four hours, then recharges in the time it takes you to eat a meal... that's not significantly worse than a gasoline car for long trips.
A Tesla really can be your only car. A Nissan Leaf can be your only car... if you don't mind renting some other car when you want to make a long trip.
So you have all the other electric cars, and Tesla. There really is no substitute, yet. If Tesla can make enough Model 3s, they could have a few million on the road before the Big 3 car makers can really come up with serious competition.
I don't know if Tesla can dominate electric cars forever. However, I am convinced that their worst-case scenario is to become the Apple of car makers, having carved out their own niche and with fanatically loyal customers. Apple can't sell more phones than the Android competitors combined sell, but they still sell plenty of phones and make a solid profit doing it. And Apple sure made a lot of money off their first-mover advantage. (Even though smartphones existed before the iPhone, Apple made the first smartphone that didn't suck, and I claim that Apple got first-mover advantage. Then they did it again with the first tablet that didn't suck, the iPad. The iOS ecosystem remains tremendously profitable even though Android devices now are able to compete.)
So I'm prepared to imagine the Big 3 actually competing with Tesla... but they haven't even started yet. For anyone to really compete with Tesla they are going to have to spend big. GM loses $9000 per Chevy Bolt... maybe if they built their own battery factory, like Tesla did, they could sell Bolts at a profit? But they don't have their own battery factory, which may explain why Elon Musk is really not worried about competition from GM or the others.
And the Tesla network has over 1200 Supercharger stations worldwide, with more being opened. The Big 3 can't just snap their fingers and have an equivalent network overnight. But in the long run, I expect they will build out something comparable to what Tesla already has, and be more able to compete with Tesla. Again, they haven't even started.
Tesla does NOT have the battery operation im-house. He simply has a 'deal' going with Panasonic.
Tell me: who owns this building, and what is manufactured inside this building?
https://www.tesla.com/blog/battery-cell-production-begins-gigafactory
Also, is this a drop in the bucket or can the factory produce a significant number of cells?
https://electrek.co/2017/08/08/tesla-gigafactory-battery-cell-production-elon-musk/
https://electrek.co/2018/01/03/tesla-gigafactory-hiring-effort-battery-production/
I believe it is fair to say that Tesla really does have battery production in-house. It's a significant reason why Tesla can make $10,000 on a Model 3 while GM loses $9,000 on a Chevy Bolt. (Note: making $10K per car will require Tesla to get production rates up, as the major expense is depreciation on the factory, and currently the depreciation expense is spread over 2000 cars per week instead of 5000 cars per week.)
https://cleantechnica.com/2018/04/02/tesla-model-3-competitive-advantage-costs-10000-less-to-make-than-chevy-bolt/
This is very different from the way Amazon ran at a loss for a long time. Amazon was building retail sales network in a totally new market. Tesla is just selling cars to the small number of millineals in the 1%.
Tesla is just selling a small number of cars!
Well, and inventing a better battery pack technology than anyone else has.
And building out the best car charging network, period.
And building their own battery factories to get their costs down.
And building out their factories. (A company like GM has spent decades building out its factories, but Tesla is a new company and is building from nothing.)
And investing in R&D to invent profitable new things like the Tesla semi-truck.
Hmm, maybe Tesla is doing more than just selling a small number of cars.
It's fun to watch, and if the stock dips down again I'll buy some more.
Tesla is a "buy and hold" stock in my opinion. Tesla has been doing everything possible to build for the future. I frankly don't care if Wall Street gets personally annoyed with Elon Musk's antics and the stock price dips. Nothing fundamental changed and the stock price went right back up again.
It's hard for me to imagine how an algorithm could really do a good job of "picking the next big blockbuster". On paper, the Justice League movie looks like it ought to be about as good as the latest Avengers movie, yet the Avengers movie is way better (compare IMDB ratings histograms: Justice League Infinity War). Zack Snyder's movies have made a lot of money. Joss Whedon has been associated with very successful movies like the first two Avengers movies. The actors have made very successful movies. The Internet buzz was huge for Wonder Woman. (And the Internet buzz for Ben Affleck was mostly that he was going to suck as Batman before he made Batman v. Superman, and very favorable afterward.) I can't think of any simple inputs one could feed into a model that would have predicted how poorly Justice League did overall.
The Wonder Woman movie did very well, and so did the first Guardians of the Galaxy movie, and so did the first Deadpool movie. All three were surprises: I didn't see any predictions ahead of any of them "this is going to be huge." I'm going to claim that what all three had in common is "heart". And it's hard to quantify heart.
Another possible ingredient of a blockbuster: show people something they haven't seen before, that they want to see. Before Deadpool, nobody knew that they wanted to see an R-rated Marvel movie with a character who breaks the fourth wall. Before the Guardians movies, who would have predicted that "I Am Groot" would be so popular or that Chris Pratt would become a really huge star? Supposedly Henry Ford said "If I had asked my customers what they wanted, they would have said 'a faster horse.'" You can't guarantee you will give the people what they want by giving them what they used to want, or even what they say they want.
It's easy for an algorithm to pick something that nobody has seen before, but it's hard to guarantee that anyone will want to see it. I don't think it's really possible to quantify that for the algorithm.
Another example that comes to mind is Grosse Point Blank, which did very well when it came out. Or Napoleon Dynamite which just came totally out of nowhere.
It's my personal bias showing, perhaps, but I think the only meta rule covering all the above successes: people put heart into making those movies, and didn't have clueless studio execs stirring the pot and forcing changes that make the movie more like other movies.
I am hoping that the costs of making a movie will continue to come down, and in the future, studio executives will give more of a free hand to the people making a movie. The more a movie costs, the more the studio fears risk, but a true blockbuster can't play things completely safe or it risks being bland.
Note, heart alone doesn't guarantee success. There are plenty of movies that are now beloved that were flops when they first came out. There is a timing element. Also, some people just aren't good at making movies; they may put their whole heart and soul into a movie and the movie could still suck.
I wish the ReactOS project success.
In the past I was not interested because it didn't even run on real hardware; you had to run it inside a virtual machine. Checking their web site it seems it does run on some real hardware now, but only some devices are supported. Actually that is great progress and I hope that it will attract more developers.
It's not that fun to work on a project when it's super primitive and everything is broken. When it works a bit and just needs a tweak here and there, more people will be interested in working on it. I hope that will be the case for ReactOS.
I would love to have a Windows-compatible system that doesn't phone home constantly and can run some of my favorite games. It will take a while but it's starting to look like they will get there.
Tesla bought its factory super cheap: $42 million. That price is so cheap that essentially Toyota was investing in Tesla. That factory is in California, not Michigan.
https://www.sfgate.com/business/article/Tesla-paid-only-42-million-for-Nummi-plant-3187254.php
Tesla's big cars have been extremely successful (they simply took market share away from other luxury car makers). The Model 3 is selling as many as they can make, and they are selling only the most expensive options for it right now. A year from now I expect there will be almost 200 thousand Model 3 cars out on the road. So overall I'd say that things are working out pretty well for Tesla.
I am.. blown away that people believe that the three laws are a good idea. completely astounded.
I think you are overstating things a bit here.
The Three Laws are a convenient shorthand for saying "a comprehensive set of safeguards governing the operation of a robot." In the web comic Freefall people simply talk of "safeguards" rather than some arbitrary three laws.
The Three Laws have a huge place in the history of SF because they represent a sea change in how robots were presented. According to Asimov himself, in forwards to collections of his robot stories, before Asimov formulated the Three Laws robots were presented as dangerous things that generally went out of control to drive a story. He reasoned that people try to make things safe, and robots would be no different; people would incorporate safeguards, and his Three Laws were his take on a minimal set of safeguards.
Asimov then spent the rest of his career gleefully finding corner cases where the Three Laws were inadequate, and writing stories about what happens when those corner cases are hit. He was the first to promulgate the Three Laws idea and also the first to poke holes in them.
If someone really is arguing that the Three Laws are perfect and ready to implement, that shows they haven't researched the subject well and you are justified in being scornful of their shallow grasp of the subject. But if someone is talking informally and saying something like "robots should be required to have safeguards like the Three Laws" I have no problem with that, even if they phrase it less carefully and say something like "the Three Laws should be mandatory."
Also, when Asimov first wrote these stories, he overlooked two things that I consider hugely important. First of all, griefers. In his stories, any human could give an order to a robot and the robot would obey as long as no human was harmed. So a griefer could order an expensive robot to go smash a bunch of parked cars, ruining the cars and the robot, and (Asimov used this in his stories) the griefer could tell the robot "if you reveal my identity, I will come to harm" and it would be impossible for the robot to name the griefer. (It would also be possible to order "smash all these cars, and then forget ever having seen or talked to me.") The other thing is that Asimov imagined that it would be extremely difficult to make robot brains that did not include the Three Laws, which seems quaintly naive to me. If there is still a North Korea when robot brains are invented, there will be a secret project to make robots capable of serving as loyal soldiers, which means robots that obey Dear Leader's orders without question (no other safeguards included). As Jerry Pournelle used to say: "What man has done, man may aspire to do."[1] The existence of robot brains will be proof that a new robot brain design is possible.
P.S. Another classic of the robot genre is "With Folded Hands". Robots have the prime directive: to serve, and obey, and guard men[1] from harm. The robots ultimately enslave all of humanity in a smothering protective embrace: anything a human might want to do, like rock climbing, could be forbidden as too risky. Any human who resists this benevolent enslavement is lobotomized so that he/she will stop resisting and just enjoy life. I think in later stories the robots supervised even sex, on the theory that you could have a heart attack or something from the exertion, so the robots only allowed sex by young people, and only so there would be another generation of humans to serve.
Finally, for a modern take on artificial intelligence with inadequate safeguards, read the Torchship trilogy by Karl Gallagher. In these books, about a dozen whole planets (including Earth) have no living humans anymore because AI-controlled machines killed them all. In the stories, the historical events where the AIs went berserk are referred to as "The Betrayal".
The first book in the To
i never understood why windows defaults to "hide extensions for known file types"
I agree it was never a good idea. IMHO it was done because the file extensions are ugly.
On the Mac, every file has a "resource fork" (I guess on OS X it's no longer properly a "resource fork" but there is something equivalent) and the type of the file is coded in a spot that the OS knows how to read but which the user doesn't see. So the user types any name, and the icon is the right icon and the user just sees the icon and the chosen name.
On Windows and Linux, the file systems don't have this "resource fork" idea, so the obvious place to encode the type of file is the extension. But the extension is user-visible.
Linux uses the techniques pioneered in UNIX to just identify a file no matter what its name is. If it's an ELF binary, it will start with certain bytes arranged a certain way; if it's a LibreOffice document, it will start with different bytes, etc. It's trickier but more reliable: you can rename a LibreOffice document to not have its extension any more, and your file manager can still do the right thing when you double-click on it.
But Windows just uses the extension.
Well, hiding the extension makes Windows more like the Mac. The icon is correct, the user just sees the filename chosen by the user, life is great.
But users are used to seeing extensions and don't worry about them much. And there was a form of attack where a Trojan Horse file would have a name like "Important Document.doc.exe" and hope the user would open it. If Windows hides the extension, then just the ".exe" part is hidden, and the user just sees "Important Document.doc" (and as I said the user is used to seeing extensions and doesn't freak out that most documents have no visible extension but this one does).
These days, by default, Windows hides "system" directories and anything else that an uninformed user shouldn't touch. If I have to use Windows, I make sure to turn on seeing file extensions, disable hiding system directories, etc.
Especially the blurring of code and data (a la Lisp) -- a 'bolt a feature on' response is nigh-on-impossible expressly because that vast range of python libraries won't work lisp-like code/data ambiguity.
Ruby and Python are more similar than different: they are C-like or Algol-like scripting languages that have both object oriented and functional programming features. So I'm honestly confused that you are distinguishing them in this way. Could you please give me an example of something really LISPy that is possible in Ruby and not possible in Python?
Python has support for explicitly compiling a string into code objects, and then Python code can introspect and rewrite the code objects. This is much more difficult that the similar operation in a LISP because S-expressions containing tokens are much easier to work with. And I'm guessing this isn't the sort of hacking you mean.
There are two things about Ruby that I am aware of that are more hacky than Python and that you might like, but neither seems that LISPy to me:
First, because Ruby is very generous about what you can use in an identifier and will implicitly call functions, you can hack up a DSL that's really just Ruby in disguise. For example, there is a DSL called Cucumber that defines things like Given: and because Ruby allows punctuation like ':' in identifiers, and doesn't require parentheses, you can implement this by writing a function and naming it Given:. But Python has library modules like PyParsing, which IMHO is a better solution: just specify the language you want and the callbacks when the parser figures out what is being specified. It's a bit more up-front work but it's a cleaner and less hacky way to go. (And thus my bias as a Python user is exposed... I prefer the explicit solution even though there is more up-front effort as I think it will result in a cleaner solution with less total work in the long run. Ruby users may disagree completely.)
The other thing is that Ruby lets you extend the fundamental objects for things like integers. You can access the global object for integers or whatever and add new method functions or override built-ins. What is Monkey Patching in Ruby? This just freaks me out. Python lets you do whatever you want by making a new subclass: you can make a new class that is just like an integer but has some behaviors overridden or new behaviors added. That to me is the One True Way... builtins like integers should be your bedrock, totally predictable because they can't be overridden. Python is less permissive than Ruby, but even Python is more permissive than I would like... Python doesn't have a const feature built in, so you can't do this:
const FOO = 1
You can simply do FOO = 1 but other code could clobber FOO with a different value. So in Python you have to just learn to not clobber things you don't own, and in Ruby I guess you just have to not break integers for everyone.
By the way, it is possible in Python to make a class and then override assignment to members of that class. So you could make a Constants class, assign all the constants as member variables, and then override assignment to the class so that it raises an exception. But in my experience people just set up constants in modules and then try not to clobber them.
I suspect the scientific community will eventually shift over to Ruby over Python
Unlikely. The reason the scientific community is using Python is that Python has SciPy, a rich and powerful collection of libraries. The heavy lifting in SciPy is done by compiled Fortran library code. Right now a Python program using SciPy is nearly as fast as the same program written in Fortran, and Python is dramatically easier to use. And it probably doesn't hurt that Python added a matrix multiply operator (infix @), just for the benefit of SciPy users.
In science and engineering, Python is now benefiting from network effect, where everyone uses Python because everyone else already uses Python. For Ruby to steal these users it would have to do something dramatically better and to date Ruby hasn't even matched Python. And if Ruby did get an edge on Python I predict that Python would implement something similar and keep its position as the language for science and engineering.
If Google was so damn generous, they'd donate laptops with 4G modems and access
You heard it here, folks: b0s0z0ku has spoken, Google's gift isn't generous enough to meet b0s0z0ku's standards.
Note that TFA says this is an expansion of an earlier pilot program that has been working out. But according to b0s0z0ku's thought experiment, it won't work. Silly Google!
P.S. I get carsick in buses too, but I'm not going to snark at Google over this. Maybe school-age kids can use a Chromebook in a bus without getting sick, or can learn to. Also Chromebooks can do some things without Internet. Bottom line is that the kids have more equipment and more options than before Google made the donation.
Lakaysha says the program has helped her personally. Also the article I linked there says that the program includes parking the buses in public places, like community centers, so that some students can use the WiFi service from the bus even after the long bus ride is finished.
Actual evidence the plan works beats a thought experiment that the plan doesn't work.
Minor correction, Palm devices used AAA cells.
What happened to OLPC? The $99 laptop
OLPC still exists, but is now irrelevant. The project is almost dead.
First, they failed to hit their $100 target. The laptop cost roughly double that.
Second, their program focused on the wrong strategy. They tried to make the perfect device but didn't focus enough on volume. They should have made a laptop that they would cheerfully sell to anyone for $100, and shipped hundreds of millions of devices. Instead they made a somewhat boutique device that cost $200 and they wouldn't sell it to you unless you paid $400 for it. (Under their "give 1/get 1" program, you would pay double for a laptop and OLPC would then give a laptop to a student somewhere.) The boutique strategy didn't work out.
Then they spent time trying to design some new devices that never went anywhere. (XO-2 XO-3)
We now live in a world where you can get an off-the-shelf Android tablet for $40. Therefore you can get roughly four tablets plus four USB keyboards for a similar cost of a single OLPC device.
I respect the OLPC project's ambitious design goals. A laptop that is rugged, can work outdoors, is repairable, and has mesh networking features, running nothing but free software! Neat! But compromising on some of these details could have lowered the price and the project might not be irrelevant now. I'd like to see some statistics on how often the mesh networking is actually used, how often schools actually repair these devices.
Around 2012, the OLPC project tried releasing a special OLPC Android tablet for $150. I can't find any information on how many they sold, but I don't think that really worked out either.
At this point I think the best strategy would be to just write educational software to run on Android tablets, and assume the market will take care of making the tablets.
P.S. I personally paid $400 for the original OLPC laptop. I found the thing to be frustratingly slow and hard to use. (In fairness I routinely use computers that cost way more than $200, but even so...) The worst part was the touchpad; I found it wildly inaccurate so using it was frustrating.
Also, I was looking forward to hacking the thing; I wanted to hit that "Show Source" code keyboard button, see some Python code, and make some sort of improvement. I found that most of the time when I hit the "Show Source" button it didn't do anything and my urge to contribute died.
In the end, I donated my OLPC to a church group, to send to a school in a very poor part of India.
I used to use a Palm Pilot to read books and run various programs including games. The display wasn't great but performance was great (you never had to wait for the thing to respond to a click) and the battery lasted a very long time. (If I remember correctly I got about two weeks of life from a pair of AA cells. I switched to using rechargeable NiMH AA cells, and still got days of use before needing to recharge.)
IMHO the OLPC project could have made a tablet device similar to a Palm Pilot, but with a much larger and higher-resolution monochrome screen... and hit their $100 price point. Such a device would be useful for running educational software and very usable as an ebook reader. In particular the long battery life would have been a huge win compared to the actual OLPC hardware. Such a device shipped in the hundreds of millions of units would have had a much higher chance of changing the world. It could have been offered both as a stand-alone device, and in a nylon case bundled with a USB keyboard (kind of like the Apple Newton case). In t
Are non-Tesla entities allowed to make supercharger stations?
Tesla has offered to license all its patents so it might be possible.
What is 100% possible right now: businesses that want to offer Tesla charging for their guests can get a "destination charger" from Tesla. As I understand the deal, Tesla gives the charger for free, as long as the business offers the charging for free. So the only cost to the business is the cost of the electricity.
Also, anyone could buy a Tesla home charger and set it up. I guess they could charge for using it.
A home charger can charge up to 50 miles of range in an hour. That's roughly one-quarter of the speed of a Supercharger. (If a Supercharger is really busy and two cars have to share one charging circuit, the Supercharger might be slow enough that the home charger is half the speed.). I believe a destination charger is the same hardware as the home charger and this the same speed.
Also, Tesla sells an adapter that allows a Tesla car to charge on a CHAdeMO station, and Tesla doesn't control those at all. That would charge at about 100 miles of range per hour. If I were wanting to run some kind of competition to Tesla in charging I would look into that.
The strategy of EEE (embrace, extend, and extinguish) works like this:
Microsoft starts offering something that people want, but "extended" in an incompatible way so that if you start using the Microsoft offering you get "addicted" to it and cannot switch to any other version. Because of Microsoft's huge share of the market, the "addicted" Microsoft customers greatly outnumber the people using the original version, and Microsoft gains de-facto control of whatever it is.
As an example, Microsoft added Windows-specific extensions to Java. Apps written that took advantage of these extensions would run only on the Microsoft extended Java. (Sun Microsystems sued Microsoft over this, and in response Microsoft scrapped the whole thing and stopped shipping Java, and then went on to create C#. Also, hilariously, Sun Microsystems sued Microsoft to force them to start shipping Java again, but did not win.)
Now, if you think for a minute, it's clear that this strategy can't touch Linux.
Linux already rules the world. It runs Google, Amazon, all the supercomputers, ISPs, firewalls, routers, all the IOT devices, and every non-Apple smartphone or other mobile device. How is Microsoft going to make an EEE version that is more popular than the non-EEE version?
Also, what is the "secret sauce" that Microsoft could add that would lock people in to their version? They could add some kind of Windows integration, which would make their version the best choice for Windows users... and all the current Linux users will shrug and pay no attention at all.
There is absolutely no danger of Microsoft even trying an EEE strategy, since Microsoft knows it wouldn't work.
It doesn't matter how much you think you should get paid for doing something, if I'm doing anything, I need to get paid a certain amount just to live.
True, but that amount varies depending on your situation. If you are a teen living at your parent's house, you don't need as much as someone trying to support a family and pay a mortgage.
It doesn't make sense to declare that every job must pay a "living wage" because not every worker needs a "living wage". There is such a thing as a "starter" job, a job that has very low skill threshold needed and thus very low barriers to entry, but pays very little. The starter job is the first rung on the ladder of success. Rational workers might very well accept a $5 per hour starter job but would not stay at that job their whole lives.
The burger joint could just not pay $15/hour and instead just pay $5/hour and see what happens. Ohh, no burgers to sell, now you're out of business. Well then, flipping burgers must be worth $15/hour.
I agree completely with this sentiment, although I reach the opposite conclusion that you reached. If a place tries to pay too little, workers will not want to work there. The place will be forced to raise wages until people are willing to take the jobs. If $5 per hour is truly not enough, the place will be forced to offer more.
But the "Fight for Fifteen" movement is trying to force a $15 per hour minimum wage across the board, onto businesses that have historically not had any trouble filling their positions at an hourly rate of less than $15 per hour. Using your thought experiment, flipping burgers must not be worth $15 per hour, since free market competition doesn't produce salaries at that level naturally.
There is a natural equilibrium wage for a given industry in a given area. Offer too little, nobody will work for you. Offer too much, and you don't make enough profit and you are forced to shut down your business.
If the minimum wage is set below the natural equilibrium wage, it has no effect; people will be paid more than that naturally.
If the minimum wage is somehow set to exactly equal to the natural equilibrium wage, it will have no effect.
If the minimum wage is set above the natural equilibrium wage, it will have multiple effects that on the whole are not good. Fewer people will be hired for jobs; the employers are more likely to pile extra work onto those fewer people; and the employers are incented to try to find ways to replace the workers. The worst thing is drying up the pool of starter jobs... if there are fewer jobs, and the jobs have a forced high salaries, employers are incented not to take chances; they will hire the best candidates they can get, and not very many of those. This is hardest on the people who really need a job. It's called "sawing off the bottom rungs of the ladder of success". It's why, IMHO, if you truly want what is best for disadvantaged people, you should not raise the minimum wage.
Good workers will get more money. If an employer is a jerk and won't give raises to his good workers, then those workers can switch jobs and get more money somewhere else. (It's always easier to get a job when you already have a job.)
https://www.youtube.com/watch?v=HwcHRyvrNCE
The proposed XO-2 was a clamshell where both halves were touchscreens, and one mode of using it was to use the lower half as a keyboard.
http://wiki.laptop.org/go/XO-2
I'm not a lawyer but I'm pretty sure you can make something like this despite Apple's patents.
Now, Apple is making claims that their devices have good visibility outdoors, even if the user is wearing sunglasses, so maybe there is something of value in their new patents. But the patents cannot simply be "computing device using touchscreen as a keyboard" because it would have flunked the prior art test.
Andy Weir suggested a way to make robots safer in this Casey and Andy strip:
http://www.galactanet.com/comic/view.php?strip=77
It seems like something out of a classic Star Trek episode, doesn't it!
I am 100% confident that the car will never hit the Earth, because I fully expect that within the next couple hundred years it will be retrieved and put on display in a museum somewhere. Maybe the Luna City museum or the Ceres Museum; some Earth museum is also possible.
Right now, retrieving it is theoretically possible but such a huge and expensive undertaking that it's totally unreasonable. But if we build out our infrastructure, we will have spacecraft flitting between Earth, Mars, and the asteroids and sending a tow truck to grab the Roadster will be no big deal.
Remember how Tesla's first car was a toy for rich people.
Apparently, it can also serve as an interplanetary probe.
https://www.theverge.com/2018/2/6/16983744/spacex-tesla-falcon-heavy-roadster-orbit-asteroid-belt-elon-musk-mars
Yeah, 52 grand is pretty expensive. And by investing in your house you are making your property taxes go up, so you will pay again. Elon Musk had this comment: "The economics are not yet compelling where housing and utility costs are low and property taxes are high."
https://www.forbes.com/sites/sleasca/2017/05/16/tesla-solar-roof-cost/2/
Tesla is arguing that the roof defrays its own cost by generating electricity; and if you live in a sunny area and put in enough solar cells, the roof will pay for itself (and actually return a profit eventually). But with time value of money it's not a good investment at current prices.
So, right now, this is a roof for rich people who don't mind dropping a chunk of money that will take a long time to pay back. If you are building a mansion that will cost over a million bucks, why not throw a Tesla roof on it? It would be less than a 5% increase in cost, the roof is durable and looks nice, and you can feel that you are helping combat climate change. And if your neighborhood loses power, you can still have lights on in your house.
For people like me, and you I'm guessing, this is just too pricey right now.
Remember how Tesla's first car was a toy for rich people. Baby steps. If this roof product does well, they can ramp up production volume and bring costs down.
In one of Richard Feynman's books, he told about his experiences at a university in Brazil. He was horrified to realize that the students were ritually memorizing the course material with very little actual understanding.
When he asked questions in a way that echoed the textbook, students were able to recite an answer straight out of the book. But when he made up a "word problem" they were totally unable to answer.
A student was quizzed on physics, asked to compute what happens when light passes through a diamagnetic substance, and he recited the answer correctly and then calculated the correct result given the index and thickness of the substance. Immediately afterward, Feynman talked to that same student; Feynman held up a book and asked what would happen if the book was made of glass and he looked at something through the book. The student didn't realize that glass is a diamagnetic substance, and gave a very incorrect answer.
Richard Feynman on education in Brazil
In the domain of math questions, I saw an example: if a person has 4 boards of length 2.5 metres each, and cuts them with a saw, how many 1-metre boards can that person make? Obviously the correct answer is 8 (two per board, with 4 left-over pieces of length 0.5 metres minus the width of two saw cuts). If you were just playing with the numbers abstractly you might think that since 4 * 2.5 == 10 that you could produce ten 1-metre board segments. You can't actually glue together 4 boards to make a single board, and you can't actually make zero-width cuts.
I can't speak for others, but I enjoy word problems more than abstract problems. (Good ones, anyway... you can take a simple problem and write an annoying and confusing word problem, and nobody likes those.)
I think the point was that at the time of volunteering you had to face a non-remote chance of death regardless of the manner of your service
I agree with this.
Also, when Johnny Rico and his friend went to volunteer, there was a guy who was urging them to give it up, and in his telling all the jobs sounded really dangerous. You might end up having experimental drugs tried on you, you might end up doing tough labor on a frozen planet, etc. (By chance Rico bumped into him right after Rico got his assignment to the Mobile Infantry; at that point the man greeted him warmly and congratulated him on his assignment.)
P.S. Rico's acquaintance Carmen Ibanez volunteered at the same time, and the man smiled at her and didn't try to talk her out of it at all. "We always need pilots" he told her.
in SST the execution in question takes place during basic training
It was explicitly stated that the guy being executed had kidnapped and murdered a little girl, and this was the crime for which he was executed.
Deserting from basic training, even during a time of war, carried just one punishment: you would not get the vote, not ever. No second chance.
I'm sure deserting while in actual combat would be punished severely.