To answer that we'd first need a working graphene transistor, and the one described in the TFA is not.
The issue currently limiting performance/watt is transistor transconductance (gain), which for bipolar transistors (at room temperature) is 1decade of output current per 60mV of input voltage change, for MOS (in subthreshold) 1decade/(80~120mV) and significantly less in saturation modes. Considering that you need ~5 decades to get ON/OFF behavior that sets the supply voltage at min. 0.5V, in practice twice as much because of variability and lower conductance in saturation. So, at least for high performance circuit, we are stuck with ~1V supply voltage and that's pretty much the end of (fast) performance scaling in CMOS.
To get better performance/watt we'd need a device that has some sort of a positive feedback based switching mechanism (breakdown, avalanche etc), which could exceed the 1decade/60mV limit. I'm not sure if we'd get that from graphene. Sure, low conductivity of graphene (ON resistance) helps too, as we could make the transistors smaller and that would reduce their gate capacitance but since energy is proportional to C*V^2, capacitance has much less effect than the voltage.
It looks like a SiC MOS transistor, with electrodes (D, G, S) made out of graphene rather than metal or polysilicon. Does it really make that much difference in performance over regular MOS transistors? If so, how much of the performance gain comes from the semiconductor material (SiC vs. Si) and how much from the interconnections? How multiple layers of interconnections are handled, if at all?
Yes, Li-ion cells supplying consumer electronics are mass produced and we are incredibly blessed to have them around (as well as research, engineering and production capacity going into them). Without that we wouldn't be able to bootstrap the EV market.
No, their form factor isn't suitable for a car (we really need larger cells for lower cost/capacity and better thermal balancing), and their chemistry is designed for short-lived high energy density products. You can get them fairly cheaply (per unit, at least) but then you spend a lot on battery assembly. If you want cells designed for EVs prepare to pay at least twice as much (per capacity, per unit that's even more). So, even with today's technology we could see ~3 times lower prices if the EV market was as developed as the consumer electronics one.
Another issue is that in a growing market production capacity lags behind the demand, producing some inertia. That's why LCDs didn't replace CRTs "overnight" - for a while LCDs were priced at premium and CRTs were dirt cheap. But that's only a transient phenomenon (it was the market *growth* that was slowing things down), the fate of CRTs was set long time ago.
I agree, and I am really surprised Nissan doesn't offer one, at least for hire. This alone could boost their sales several times.
Perhaps the problem with the trailer is that it implies a small range extender of the type I was writing about above. You simply can't take a Volt's ICE and put it on a trailer.
Honestly, if there was an electric car with 100 miles range and (detachable or not) range extender *extending* the range 3x, I would buy it tomorrow. I don't even care about efficiency of the range extender, as long as it is cheap, small and quiet.
Batteries are already good enough (see Tesla S). It is their price that hinders the market. Luckily batteries are much easier to produce at mass scale than, say, Diesel engines, so the market *will* grow exponentially (more customers -> larger scale of production -> cheaper batteries -> more customers).
Performance of batteries will improve as well but it won't make or break the deal. In a worst case you can just add a cheap 600cc range extender and make the car go 300+ miles on a full charge.
With a disruptive technology like this it is *very* unlikely we will see a 50% adoption for an extended period of time. Either EVs will catch on and it will be highly uncool to drive a gas fueled car, or they will not and the market share will settle at single digits. Sure, to go from 0% to 99% you have to pass that 50% but at this level we can expect the highest rate of growth, so I wouldn't bet on the time it will happen.
The world's got crazy, and I'm not talking just about Apple (which is no exception, btw). But, how come exporting stuff is good (in it's own, rather than as a means of getting money for importing stuff), or why import restrictions are considered to be a penalty to producers rather than consumers? Another example, people often say, even on/., that cheap immigrant workforce from overseas is bad and the solution to it is to make it expensive. What?! Since when paying more is better than paying less?
All this BS comes from a false assumption that wealth is the amount of green paper we shuffle each day. No, it's still the amount of stuff and services we can buy, so low prices are as important as high income. This seems like an easy mistake to spot, unless you are a politician, who gets a share of our income but not our losses.
I'm disappointed with Mozilla's approach to privacy (or lack of it). Currently the biggest danger for privacy is not tracking (your bank also tracks your transactions) but collecting all the available threads of information to build a fairly complete profile of the user. Yet Mozilla is pretty much ignoring the problem to the point it is difficult to differentiate Firefox from Google Chrome (a browser specifically designed for collecting information).
The only thing I ask for is a good identity manager (Multifox v.1.x is pretty good) and a convenient cookie manager (for lack of better alternatives I use CookieCuller). Things I *don't* want are "do not track" efforts, which change nothing, except for giving Mozilla an undeserved label "we care".
Your TODO list:
Make the damn identities and cookies first class components of the browser and let the users control them as easily as they control URLs or tabs. In my current setup, I have several Firefox windows open, each with a different user being logged into Google/FB/you_name_it, and with different sets of cookies allowed. This works pretty well but currently this setup takes too much fiddling to work.
Identity management should be integrated. Period. Not as a clumsy session management dialog box, which only shows up at start-up (if you ask for it). Identity name should be displayed in the url/title bar, and integrated with the context menu ("Open as...", or "New Window with Identity..."), bookmarks, URL bar etc.
Cookies are still waiting for a good manager, with some sort of user contributed black/white lists (like Adblock did for URLs). Filtering cookies should be as easy as "block cookies from this provider when browsing as..." (note that identity shows up here too).
"For Clojure, it is about keeping the language itself simple."
No, it is not. It is about simplicity and flexibility of composing small pieces of code into complex software systems, something most languages and toolkits seriously suck at. The nice thing about Clojure is that its principles are so easy to grasp - data don't change while you're looking at them, and operations on these data can be arbitrarily composed because they only take a value in and produce a value out. That's all, after using it for a while I wonder why languages weren't like this from the beginning.
The language itself, while simple, is more complex than many other Lisps. OTOH, it is much simpler than Haskell, which has similar goals (among many many others).
Xfce panel may not be very intuitive to configure (try right-click, DnD, panel settings dialog etc.) but it works pretty damn well and actually offers more functionality than Gnome 2 panel ever did. Things like vertical panels were broken for the whole lifetime of Gnome2 and have never been fixed properly (although there are some patches floating around). Or, applets which move around and not return to original position when the screen resolution changes.
Every now and then I try KDE (I used to use it 10 years ago). On average it takes about an hour to configure it to a usable state (partly because of brain-dead defaults designed for showing off the desktop rather than for usability, partly because of messy and uber-detailed configuration). When I'm done, I get a configuration, which looks pretty much like my Xfce desktop, albeit being heavier, less aesthetically pleasing, and demanding too much attention from me. No surprise I end up returning to my trusty desktop.
It doesn't mean that Xfce is perfect - no DE has ever been perfect. But it is easy, efficient, visually pleasing, and doesn't try to overwhelm me with all the coolness in the world.
Not only possible but desirable. The fact they can't filter out the content mean the system is implemented correctly from the security point of view. You wouldn't like to put your own sensitive data in a system with backdoors, as it inherently relies on trust between you and whoever happens to administrate the system.
In a way, if bringing security to an average Joe is dangerous, there is something terribly wrong with the legal system we live in. In USSR people had to apply for permission for traveling from one city to another. The justification was eerily similar - if you give people freedom they will "abuse" it (read: use against the rulers).
What if someone kills your daugther? Should we pass a bill to bring her back to life? Or maybe we just put the murderer in jail.
If I had to choose, I would much more prefer to have CP pictures floating around than having a wide-spread surveillance network looking into *all* aspects of my life.
This is a fine act of improving quality of our lifes. On one hand being killed or raped makes the victim's life pitiful (or gone), on the other - eliminating this danger is impossible and makes everybody's life poor (no one has managed to solve this problem, not even China or NK).
Printers use that high resolution because they are essentially 1-bit-per-color devices (sometimes slightly more). To get grayscale they need dithering, which reduces the effective resolution a lot (64 levels of grayscale need, in the worst case, 8x8 pixels for one logical pixel).
If there was more demand for oil it would be cheaper than gasoline - production cycle of oil is both simpler and cheaper than that of gasoline. Especially with crude oil getting heavier.
it could result in an engine that feels and behaves like a gasoline engine but has the mileage of a diesel. That would be nice.
Except that new diesels feel an behave better than new downsized petrol engines. It used to be the opposite 20 years ago, but now the roles have swapped.
I was referring to "value that the buyers decide they have, not the sellers". I agree with everything else you said.
If only one side is involved, it is not a price, just an "offer" or an "assessment". To make it a price both parties have to be happy with the deal. Every transaction is by definition a win-win situation.
That's a broader problem with the way the science works. Scientists are rarely interested in applications. The have to give some "justification", which looks reasonable enough to earn them a grant. But that's where work on applications often ends.
Remember, you get, what you measure. What matters in the academic world is the number of publications (and to a lesser degree their quality). Making breakthroughs isn't the best, or certainly the easiest way of becoming a successful scientist. It's way better to throw ideas left and right producing incremental (if any) improvements, and not to dwell on a single idea for too long.
One thing he is right about, though, is that battery cells developed for portable applications aren't particularly well suited to EVs.
Don't get me wrong - it is fantastic we have them, and have them manufactured at a mass scale. This way we can piggy-back on decades of intensive R&D that went into them. Without that there wouldn't be mass manufactured electric cars on the road now. Portable batteries have even proven pretty good in that application, but it doesn't mean we can't improve on that.
In a long term we probably want batteries where reactants are not stored in a solid form, separated only by a thin layer of electrolyte. We can afford having a pump here and there, or add a couple of (refillable?) tanks for keeping reactants and their byproducts. There were already some attempts of doing just that, but at the moment such batteries simply can't compete with the whole industry backing up the development of portable batteries. This will have to wait until EV's gain more market share.
Although I consider hydrogen a dead-end (maybe except for special applications, like airplanes), the research that goes into fuel cell may produce something useful (who says the reactants must be "H_2" and "O_2" after all?).
This may change overnight (I've seen that in some coutries where, at least at some point in time, majority of cars on the road were using LPG).
Performance of EVs is no longer a problem - there are batteries, which can take you 300 miles on a single charge. They are just not yet economically viable for lower segments of the market.
The good news is that there is absolutely no reason for the batteries or other EV components to be more expensive than, say, a gas engine. They are a lot easier to manufacture and take less resources (amount of lithium in an Li-ion battery is pretty miniscule). It's just a matter of ramping up the production.
Also, markets don't respond linearly, especially in emerging applications (Apple didn't just put a hardisk in an iPod - they put a disk big enough to store all you music in it). There's almost always a threshold "good/cheap enough", which makes all the difference. But, if you want to cross the threshold you have to reach it first (which isn't nearly as flashy). OTOH, once you're above it, it doesn't matter if your camera has 10Mpixels or 100Mpixels - you're probably better off fighting challanges that matter.
That's where speculators come in handy.
Hypothetical situation: people don't care about energy, use plenty of fuel for as long as they can (come on, production of gas isn't all that more expensive, is it?), and, suddenly, they wake up with gas prices od $100/gallon.
It's hypothetical because there are people who try to predict the future. If they are right - they get plenty of money, if not - they loose (pretty damn good incentive for being right). If many of them expect a hike in prices of oil - the prices will slowly ramp up and at the same time some amount of oil will be stock piled for later use.
Voters don't like that ("let's just burn all the oil now") and politicians are pushing for moving the reserves back into retail. But, voters don't care, and politicians only care about the next election. Who do you believe then?
Nuclear is bad. Nuclear is not safe and never will be. It is also going to be necessary for the next 50-100 years.
All strong sources of energy are inherently dangerous and expensive (in absolute terms). They differ enough from each other to make you choose your poison, that's it.
For the amount of energy nuclear plants produce, they are relatively cheap and safe.
Coal has many operational issues, but failure is limited to the plant and extremely immediate surrounding area.
Coal plants are failing continuously (as a part of their design), and by doing so they affect much larger area than nuclear plants will ever do.
Slashdot Mirror
To answer that we'd first need a working graphene transistor, and the one described in the TFA is not.
The issue currently limiting performance/watt is transistor transconductance (gain), which for bipolar transistors (at room temperature) is 1decade of output current per 60mV of input voltage change, for MOS (in subthreshold) 1decade/(80~120mV) and significantly less in saturation modes. Considering that you need ~5 decades to get ON/OFF behavior that sets the supply voltage at min. 0.5V, in practice twice as much because of variability and lower conductance in saturation. So, at least for high performance circuit, we are stuck with ~1V supply voltage and that's pretty much the end of (fast) performance scaling in CMOS.
To get better performance/watt we'd need a device that has some sort of a positive feedback based switching mechanism (breakdown, avalanche etc), which could exceed the 1decade/60mV limit. I'm not sure if we'd get that from graphene. Sure, low conductivity of graphene (ON resistance) helps too, as we could make the transistors smaller and that would reduce their gate capacitance but since energy is proportional to C*V^2, capacitance has much less effect than the voltage.
It looks like a SiC MOS transistor, with electrodes (D, G, S) made out of graphene rather than metal or polysilicon. Does it really make that much difference in performance over regular MOS transistors? If so, how much of the performance gain comes from the semiconductor material (SiC vs. Si) and how much from the interconnections? How multiple layers of interconnections are handled, if at all?
Sort of.
Yes, Li-ion cells supplying consumer electronics are mass produced and we are incredibly blessed to have them around (as well as research, engineering and production capacity going into them). Without that we wouldn't be able to bootstrap the EV market.
No, their form factor isn't suitable for a car (we really need larger cells for lower cost/capacity and better thermal balancing), and their chemistry is designed for short-lived high energy density products. You can get them fairly cheaply (per unit, at least) but then you spend a lot on battery assembly. If you want cells designed for EVs prepare to pay at least twice as much (per capacity, per unit that's even more). So, even with today's technology we could see ~3 times lower prices if the EV market was as developed as the consumer electronics one.
Another issue is that in a growing market production capacity lags behind the demand, producing some inertia. That's why LCDs didn't replace CRTs "overnight" - for a while LCDs were priced at premium and CRTs were dirt cheap. But that's only a transient phenomenon (it was the market *growth* that was slowing things down), the fate of CRTs was set long time ago.
I agree, and I am really surprised Nissan doesn't offer one, at least for hire. This alone could boost their sales several times.
Perhaps the problem with the trailer is that it implies a small range extender of the type I was writing about above. You simply can't take a Volt's ICE and put it on a trailer.
Honestly, if there was an electric car with 100 miles range and (detachable or not) range extender *extending* the range 3x, I would buy it tomorrow. I don't even care about efficiency of the range extender, as long as it is cheap, small and quiet.
Batteries are already good enough (see Tesla S). It is their price that hinders the market. Luckily batteries are much easier to produce at mass scale than, say, Diesel engines, so the market *will* grow exponentially (more customers -> larger scale of production -> cheaper batteries -> more customers).
Performance of batteries will improve as well but it won't make or break the deal. In a worst case you can just add a cheap 600cc range extender and make the car go 300+ miles on a full charge.
With a disruptive technology like this it is *very* unlikely we will see a 50% adoption for an extended period of time. Either EVs will catch on and it will be highly uncool to drive a gas fueled car, or they will not and the market share will settle at single digits. Sure, to go from 0% to 99% you have to pass that 50% but at this level we can expect the highest rate of growth, so I wouldn't bet on the time it will happen.
Simple, just use electricity we currently waste on drilling, refining and transportation of oil.
The need wasn't strong enough to justify a purchase of a dedicated reader. Especially that there is a critical mass effect involved.
The world's got crazy, and I'm not talking just about Apple (which is no exception, btw). But, how come exporting stuff is good (in it's own, rather than as a means of getting money for importing stuff), or why import restrictions are considered to be a penalty to producers rather than consumers? Another example, people often say, even on /., that cheap immigrant workforce from overseas is bad and the solution to it is to make it expensive. What?! Since when paying more is better than paying less?
All this BS comes from a false assumption that wealth is the amount of green paper we shuffle each day. No, it's still the amount of stuff and services we can buy, so low prices are as important as high income. This seems like an easy mistake to spot, unless you are a politician, who gets a share of our income but not our losses.
I'm disappointed with Mozilla's approach to privacy (or lack of it). Currently the biggest danger for privacy is not tracking (your bank also tracks your transactions) but collecting all the available threads of information to build a fairly complete profile of the user. Yet Mozilla is pretty much ignoring the problem to the point it is difficult to differentiate Firefox from Google Chrome (a browser specifically designed for collecting information).
The only thing I ask for is a good identity manager (Multifox v.1.x is pretty good) and a convenient cookie manager (for lack of better alternatives I use CookieCuller). Things I *don't* want are "do not track" efforts, which change nothing, except for giving Mozilla an undeserved label "we care".
Your TODO list:
Make the damn identities and cookies first class components of the browser and let the users control them as easily as they control URLs or tabs. In my current setup, I have several Firefox windows open, each with a different user being logged into Google/FB/you_name_it, and with different sets of cookies allowed. This works pretty well but currently this setup takes too much fiddling to work.
Identity management should be integrated. Period. Not as a clumsy session management dialog box, which only shows up at start-up (if you ask for it). Identity name should be displayed in the url/title bar, and integrated with the context menu ("Open as ...", or "New Window with Identity ..."), bookmarks, URL bar etc.
Cookies are still waiting for a good manager, with some sort of user contributed black/white lists (like Adblock did for URLs). Filtering cookies should be as easy as "block cookies from this provider when browsing as ..." (note that identity shows up here too).
"For Clojure, it is about keeping the language itself simple."
No, it is not. It is about simplicity and flexibility of composing small pieces of code into complex software systems, something most languages and toolkits seriously suck at. The nice thing about Clojure is that its principles are so easy to grasp - data don't change while you're looking at them, and operations on these data can be arbitrarily composed because they only take a value in and produce a value out. That's all, after using it for a while I wonder why languages weren't like this from the beginning.
The language itself, while simple, is more complex than many other Lisps. OTOH, it is much simpler than Haskell, which has similar goals (among many many others).
Xfce panel may not be very intuitive to configure (try right-click, DnD, panel settings dialog etc.) but it works pretty damn well and actually offers more functionality than Gnome 2 panel ever did. Things like vertical panels were broken for the whole lifetime of Gnome2 and have never been fixed properly (although there are some patches floating around). Or, applets which move around and not return to original position when the screen resolution changes.
Every now and then I try KDE (I used to use it 10 years ago). On average it takes about an hour to configure it to a usable state (partly because of brain-dead defaults designed for showing off the desktop rather than for usability, partly because of messy and uber-detailed configuration). When I'm done, I get a configuration, which looks pretty much like my Xfce desktop, albeit being heavier, less aesthetically pleasing, and demanding too much attention from me. No surprise I end up returning to my trusty desktop.
It doesn't mean that Xfce is perfect - no DE has ever been perfect. But it is easy, efficient, visually pleasing, and doesn't try to overwhelm me with all the coolness in the world.
Not only possible but desirable. The fact they can't filter out the content mean the system is implemented correctly from the security point of view. You wouldn't like to put your own sensitive data in a system with backdoors, as it inherently relies on trust between you and whoever happens to administrate the system.
In a way, if bringing security to an average Joe is dangerous, there is something terribly wrong with the legal system we live in. In USSR people had to apply for permission for traveling from one city to another. The justification was eerily similar - if you give people freedom they will "abuse" it (read: use against the rulers).
What if someone kills your daugther? Should we pass a bill to bring her back to life? Or maybe we just put the murderer in jail.
If I had to choose, I would much more prefer to have CP pictures floating around than having a wide-spread surveillance network looking into *all* aspects of my life.
This is a fine act of improving quality of our lifes. On one hand being killed or raped makes the victim's life pitiful (or gone), on the other - eliminating this danger is impossible and makes everybody's life poor (no one has managed to solve this problem, not even China or NK).
Printers use that high resolution because they are essentially 1-bit-per-color devices (sometimes slightly more). To get grayscale they need dithering, which reduces the effective resolution a lot (64 levels of grayscale need, in the worst case, 8x8 pixels for one logical pixel).
One of many oddities of mass manufacturing.
If there was more demand for oil it would be cheaper than gasoline - production cycle of oil is both simpler and cheaper than that of gasoline. Especially with crude oil getting heavier.
it could result in an engine that feels and behaves like a gasoline engine but has the mileage of a diesel. That would be nice.
Except that new diesels feel an behave better than new downsized petrol engines. It used to be the opposite 20 years ago, but now the roles have swapped.
I was referring to "value that the buyers decide they have, not the sellers". I agree with everything else you said.
If only one side is involved, it is not a price, just an "offer" or an "assessment". To make it a price both parties have to be happy with the deal. Every transaction is by definition a win-win situation.
If that was the case, bitcoins would be worth $0. A transaction always involve two parties, which (both of them) must agree it is "worth" making it.
That's a broader problem with the way the science works. Scientists are rarely interested in applications. The have to give some "justification", which looks reasonable enough to earn them a grant. But that's where work on applications often ends.
Remember, you get, what you measure. What matters in the academic world is the number of publications (and to a lesser degree their quality). Making breakthroughs isn't the best, or certainly the easiest way of becoming a successful scientist. It's way better to throw ideas left and right producing incremental (if any) improvements, and not to dwell on a single idea for too long.
One thing he is right about, though, is that battery cells developed for portable applications aren't particularly well suited to EVs.
Don't get me wrong - it is fantastic we have them, and have them manufactured at a mass scale. This way we can piggy-back on decades of intensive R&D that went into them. Without that there wouldn't be mass manufactured electric cars on the road now. Portable batteries have even proven pretty good in that application, but it doesn't mean we can't improve on that.
In a long term we probably want batteries where reactants are not stored in a solid form, separated only by a thin layer of electrolyte. We can afford having a pump here and there, or add a couple of (refillable?) tanks for keeping reactants and their byproducts. There were already some attempts of doing just that, but at the moment such batteries simply can't compete with the whole industry backing up the development of portable batteries. This will have to wait until EV's gain more market share.
Although I consider hydrogen a dead-end (maybe except for special applications, like airplanes), the research that goes into fuel cell may produce something useful (who says the reactants must be "H_2" and "O_2" after all?).
This may change overnight (I've seen that in some coutries where, at least at some point in time, majority of cars on the road were using LPG).
Performance of EVs is no longer a problem - there are batteries, which can take you 300 miles on a single charge. They are just not yet economically viable for lower segments of the market.
The good news is that there is absolutely no reason for the batteries or other EV components to be more expensive than, say, a gas engine. They are a lot easier to manufacture and take less resources (amount of lithium in an Li-ion battery is pretty miniscule). It's just a matter of ramping up the production.
Also, markets don't respond linearly, especially in emerging applications (Apple didn't just put a hardisk in an iPod - they put a disk big enough to store all you music in it). There's almost always a threshold "good/cheap enough", which makes all the difference. But, if you want to cross the threshold you have to reach it first (which isn't nearly as flashy). OTOH, once you're above it, it doesn't matter if your camera has 10Mpixels or 100Mpixels - you're probably better off fighting challanges that matter.
That's where speculators come in handy. Hypothetical situation: people don't care about energy, use plenty of fuel for as long as they can (come on, production of gas isn't all that more expensive, is it?), and, suddenly, they wake up with gas prices od $100/gallon. It's hypothetical because there are people who try to predict the future. If they are right - they get plenty of money, if not - they loose (pretty damn good incentive for being right). If many of them expect a hike in prices of oil - the prices will slowly ramp up and at the same time some amount of oil will be stock piled for later use. Voters don't like that ("let's just burn all the oil now") and politicians are pushing for moving the reserves back into retail. But, voters don't care, and politicians only care about the next election. Who do you believe then?
Nuclear is bad. Nuclear is not safe and never will be. It is also going to be necessary for the next 50-100 years.
All strong sources of energy are inherently dangerous and expensive (in absolute terms). They differ enough from each other to make you choose your poison, that's it. For the amount of energy nuclear plants produce, they are relatively cheap and safe.
Coal has many operational issues, but failure is limited to the plant and extremely immediate surrounding area.
Coal plants are failing continuously (as a part of their design), and by doing so they affect much larger area than nuclear plants will ever do.