Moore's law, the base of his argument that technology is evolving exponentially is pretty much on schedule. We are now on the Petaflop (10^15) range, with the transistor count following the predicted exponential.
RAM cost is now at 28000kB/USD, also fitting the curve
GDP per capita also is within schedule (note that the scale is logarithmic), even with the wealth transfer east (which is bound to be limited in time to ten more years give or take)
And, lastly, the core of all atacks on Kurzweil, so is life expectancy on track.
You may still believe these exponentials will hit some kind of ceiling somehow. That might be true. The numbers, however, support Kurzweil's theory. And predicting from the number of times Moore's law depletion was announced in the last twenty years, I'd wager my bets on Kurzweil.
As a result, the genome alone cannot possibly tell you how to "make" an organism...
Untrue. The genome gives you the instructions for a self-modifying program that eventually produces a human brain. All biology experiments so far seem to point to the fact that the DNA chain indeed contains all the information needed to build an organism.
For the most part, the "expensive" countries all have dying cultures, since they don't reproduce enough to survive. (Remember "replacement rate" is 2.1)
Bah. Portugal is indeed below 2.1, much as most western countries. What most people seem to forget is that Portugal had 5 million people in the XIX century, and has 10 million today. We can shrink to half, and then pickup the slack. Growth isn't good by itself. Heck, we colonized the world when we were 2.5 million strong...
These numbers are probably similar all over Europe.
You're way ahead of me. I'd miss some US states if I tried to list them all (I'm from Europe). Although I can point them on a map if someone names them. Heck, I have some trouble listing all European countries, with all the independence calls of eastern states in the last couple decades...
Looking at GDP, yes. But you need to read it in a different light. Portugal is a temperate climate country, with a socialist government, where people live reasonably well:
Medical services are free, with great quality (everyone bitches about it here, but we do hold world top-ranks in quality measures like newborn survival rate or cancer survival years).
Education is free up to Msc levels. Top students easily get state-paid Phd tuition fees.
Everyone is entitled to a minimum subsidy of 250eur/month, even if you do absolutely nothing. If you are, say, a single parent with two children, you may get 750eur/month from state subsidies. This in a country with minimum wage of a bit over 410eur
Retirement pension is state-managed and covers the entire population. Again, even if you haven't worked a day in your life
All of this greatly reduces incentives to entrepreneurship, with obvious results in economic evolution. On the flipside, it means that a salary of 1500eur/month buys you a good home, a nice car and two weeks vacation in the tropics, because you don't have to save for health, education or harsh times.
Further, Portugal suffers from low education levels when compared to the rest of Europe and namely when compared to former USSR countries. It's an effect of our dictatorship (ended in '74) that will take a generation to fix (and is indeed being fixed). The result is that industry uses little capital, uses more manpower than machinery and suffers naturally from the consequent low productivity.
Nevertheless, it's a really nice country to live in, with economic parameters difficult to explain to a Northern American.
True. That's why Portugal signed deals to be the prototype territory for Nissan and Renault to test their electric cars. The country is covering the territory with electric recharging stations and will be subsidizing electric cars in the next years. The strategic plan seems to be to actually reduce oil consumption.
The grid of recharging stations will be ready by year-end. The Nissan Leaf will start selling in 2011, together with the Opel Ampera (Chevrolet Volt Mark II). The Renault lineup is to appear between 2011 and 2013. The electric Ferrari has been promised, but no ETA yet;-)
You have to note that the geography of the country greatly helps. The country has about 1000km top to bottom, with 90% of the population near the sea, on a strip of no more than 100km. It's rather easy to cover that with recharging stations.
Very much true. Great comment. Just to clarify, we have about 50% installed capacity in gas/coal plants, with the other 50% in hydroelectric dams. Wind is just extra. Solar is minimal. If there is wind, then it's free energy, if not, fire up the gas plant. From what I recall, most of the year compensation is done by hydroelectric dams. The critical period is mid to late summer, when A/C power ramps up consumption and when dams hit their low levels. During this period, compensation is done by gas/coal plants.
Anyhow, if there is a lesson to be learned by the US here, it's not about renewable energy. It's about the power grid. The article doesn't mention it, but upgrading the power grid, which happened in the late 80s is the key to this project. The electric grid here is incredibly efficient and a feat of engineering EDP should be proud of. It is fully automatic, balances out production from a myriad of extremely unreliable sources, prioritizing renewable sources and using dams and gas/coal plants to compensate whenever needed (be it by storing energy or ramping up production). A brownout is unheard of, while ground shunting losses are minimal. It is indeed a smart grid, as all should be.
If you have a good electric grid, then whenever renewables hit the economic sweet spot they will get built. It's just that for us, the sweet spot is higher because we have no oil, no gas and limited coal in the territory.
We built lots of hydroelectric dams in the 70s and 80s, so we use that as energy stores. We built secondary dams downriver, pool water there and pump up when wind overgenerates energy. It's not incredibly efficient, but it's better than shunting the energy into the ground or stopping the wind farms.
Answering you: No, they are not self-sufficient in the sense that they are independent from the main grid. They are self-sufficient because they produce more than they consume. Doing a piece of electric network isolated wouldn't cross our mind. The country's electric grid is what North-Americans call a 'smart-grid' and is something every country in the world should have. It easily balances load accross the country, starting and stopping hydro plants and in the limit starting and stopping natural gas plants in peak season (typically late summer: depleted dams, lots of A/C running). It even automatically manages importing and exporting energy from/into Spain and France.
Your whole argument hinges on high transmission losses, which don't exist. Electric energy transmission is very efficient. For short distances (100km) it's about 98% efficiency, and even for extreme distances you won't dip below 90%. Portugal used to import electricity from France. Electricity traveled 2000km between the nuclear power plant and the grid connect point in Sines, and it's not a remarkable feat. I doubt you can't feed boston from anywhere on a 3000km radius. You can get into flyover-state territory with 3000km...
You managed to read the article with a large bias. It *does* explain how to deal with the issues of fragmentation: a) guarantee APIs are forward compatible; b) have apps declare their hardware needs; c) minimize bugs/incompatible APIs using extensive testing.
Google does assume fragmentation is inevitable. That seems to be under discussion by some people here on/. Personally, I can't fathom how is fragmentation avoidable, unless by stagnation. Stagnation is quite the opposite of the Android ecosystem, which is evolving at a very fast pace (fast hardware and software release cycles).
If indeed fragmentation is inevitable, Google has seemingly defined the problem correctly and it looks like the solution is good.
It's not like there's no other option. Basically, as a developer, you must choose between a large user base now with an unknown probability of being kicked off with no warning, or a smaller user base (in the Android market) with clear acceptance rules. It's not an easy choice. Risk has rewards. Risk also causes shit to hit the fan. Just don't complain if/when it happens.
Fragmentation is needed for a competitive environment. It's an added problem, partly for developers but mainly for google, and they are handling it quite well. Properly accounting for different hardware targets in both the hardware development and in the software development kits is a daunting task. However, and I feel everyone is repeating the fragmentation mantra without giving proper credit to Google, Android handles fragmentation quite well. Apps are always forward-compatible (write for 1.5 and you get ~100% compatibility with existing handsets), and they announce the hardware they need.
Do you need a camera? Declare it on your manifest, and the app appears on the market only to devices sporting a camera. Do you absolutely need multitouch? Declare it. Do you need an SD card? Declare it. The only drawback is that every requirement you add narrows down the range of devices your app appears in.
Would it be better if there were fewer devices all alike? It'd be like the narrowing decision would have already been made for you. Oh, right. That's the Apple way: Users are too stupid, let's decide for them.
In the end, it's different. It's not worse. It's more complicated for the developer, in exchange for a larger user base. Before anyone mentions there are more iPhones than Android devices, please first consider that: a) there are more Android devices than iPhones being sold today and; b) Android covers a much wider range of price-points, and is thus in reach of a much larger user base, so this tendency is likely here to stay (think how Nokia is still king of mobile handsets).
I have a Thinkpad X61t with a GM965 integrated intel video chip. Never had a problem with suspend/resume. It just works (under ubuntu now, and under Gentoo previously).
If something moving on a screen a row or two below your position is cause for distraction, I'd wager you have the wrong professors. Good teachers grab the students' attention and only let go after an hour or an hour and a half. I had one MBA professor that could keep the entire class engaged for a whole 5 hour period (with a 15min break at mid-time), but that's definitely an outlier.
Then, Google has complete control over what can be released, and can even uninstall apps that were previously downloaded.
This is a blatant lie. Android even allows for marketplaces other than Google's.
Google does control whatever goes into the Android Marketplace. It has no control whatsoever over the devices. I'm ok with that. If they bork the Android Marketplace, I'm sure an alternative will popup.
On the contrary all of them must have been speed related. Stationary cars kill no-one.
And herein lies the confusion between cause and condition. Speed is rarely the cause for an accident. It is always a condition. The trick is, acting on causes has much better effects than acting on conditions.
You have a simplistic view, which I don't hope to change. Please take my opinion as a call for more information. Go read on best cases in death toll reductions in Europe (Portugal's IP5 road comes to mind). Hint: The action focus was not speed.
This is an article by Chris Anderson, and the summary futzes it up and says nothing about the author. This isn't like reading some schmuck. Chris Anderson , editor of Wired, author of "The Long Tail" and the "Free" Wired article.
You're referring to teams in the Performing stage. This is the hallmark of an excellent manager. It only happens on a residual number of teams, though. Most teams get stuck in Norming, and a lot of badly managed teams get stuck in Storming:
Integer problems are in general quite complex to solve. Not the US election, which was the case here. It's easily approachable as a linear continuous problem, then constrain critical variables and then a brute-force search on the 10 or so critical states.
You got it wrong. Let me correct the text for you:
P.S. If you've got some way to analytically solve any constrained non-linear optimization problem with 50+ variables, there's probably a long line of people with medals and/or piles of cash to give you.
Linear problems are easy-peasy. A 50-simplex is solvable by-hand.
P.S. The US election result optimization is a linear problem.
When you buy an iPhone and judge if the price is right, do you take into consideration the existence of a power brick in the package? You don't, and nobody else does. The perceived value of the product is elsewhere, and that reduction in cost will never cause a reduction in price.
I do know that part of the issue is power dissipation. When you're converting 5V (from usb) to what the charger circuit in the cell phone needs, the difference ends affecting power dissipation (in the form of heat). Voltage difference between power source and charger times max current draw = amount of power you need to dissipate at that stage (or simply P=I*V). A charger made specifically with that battery in mind would minimize power wasted (on charging) and would mean less electronics on the phone.
Completely false. Transformers do have losses (namely eddy current) and do produce heat. However, it's nowhere near I*deltaV. You'd see high to medium-voltage and medium to low-voltage transformers in the power grid blowing up everywhere if your proposition was anywhere near reality.
That would reduce the cost, not add to it. (...) the consumer would end up spending less after buying three devices.
Faulty logic. Sale price is not related to production cost (other than the condition that you must be able to sell at a profit or you'll go bust). Reduced cost will produce higher margins and a larger profit on the initial product. It'll also mean the end to the "extra chargers" market.
We're talking embedded devices here. Stuff like my wireless router. Try to do a DNS poisoning attack or a man-in-the-middle when I'm typing the IP in the address bar and have the device on my network segment...
Slashdot Mirror
Moore's law, the base of his argument that technology is evolving exponentially is pretty much on schedule. We are now on the Petaflop (10^15) range, with the transistor count following the predicted exponential.
Cost of DNA sequencing, another of his examples, is today at 0.000008(USD) per base pair. Fits the curve.
RAM cost is now at 28000kB/USD, also fitting the curve
GDP per capita also is within schedule (note that the scale is logarithmic), even with the wealth transfer east (which is bound to be limited in time to ten more years give or take)
And, lastly, the core of all atacks on Kurzweil, so is life expectancy on track.
You may still believe these exponentials will hit some kind of ceiling somehow. That might be true. The numbers, however, support Kurzweil's theory. And predicting from the number of times Moore's law depletion was announced in the last twenty years, I'd wager my bets on Kurzweil.
As a result, the genome alone cannot possibly tell you how to "make" an organism...
Untrue. The genome gives you the instructions for a self-modifying program that eventually produces a human brain. All biology experiments so far seem to point to the fact that the DNA chain indeed contains all the information needed to build an organism.
For the most part, the "expensive" countries all have dying cultures, since they don't reproduce enough to survive. (Remember "replacement rate" is 2.1)
Bah. Portugal is indeed below 2.1, much as most western countries. What most people seem to forget is that Portugal had 5 million people in the XIX century, and has 10 million today. We can shrink to half, and then pickup the slack. Growth isn't good by itself. Heck, we colonized the world when we were 2.5 million strong...
These numbers are probably similar all over Europe.
You're way ahead of me. I'd miss some US states if I tried to list them all (I'm from Europe). Although I can point them on a map if someone names them. Heck, I have some trouble listing all European countries, with all the independence calls of eastern states in the last couple decades...
Looking at GDP, yes. But you need to read it in a different light. Portugal is a temperate climate country, with a socialist government, where people live reasonably well:
All of this greatly reduces incentives to entrepreneurship, with obvious results in economic evolution. On the flipside, it means that a salary of 1500eur/month buys you a good home, a nice car and two weeks vacation in the tropics, because you don't have to save for health, education or harsh times.
Further, Portugal suffers from low education levels when compared to the rest of Europe and namely when compared to former USSR countries. It's an effect of our dictatorship (ended in '74) that will take a generation to fix (and is indeed being fixed). The result is that industry uses little capital, uses more manpower than machinery and suffers naturally from the consequent low productivity.
Nevertheless, it's a really nice country to live in, with economic parameters difficult to explain to a Northern American.
True. That's why Portugal signed deals to be the prototype territory for Nissan and Renault to test their electric cars. The country is covering the territory with electric recharging stations and will be subsidizing electric cars in the next years. The strategic plan seems to be to actually reduce oil consumption.
The grid of recharging stations will be ready by year-end. The Nissan Leaf will start selling in 2011, together with the Opel Ampera (Chevrolet Volt Mark II). The Renault lineup is to appear between 2011 and 2013. The electric Ferrari has been promised, but no ETA yet ;-)
You have to note that the geography of the country greatly helps. The country has about 1000km top to bottom, with 90% of the population near the sea, on a strip of no more than 100km. It's rather easy to cover that with recharging stations.
Very much true. Great comment. Just to clarify, we have about 50% installed capacity in gas/coal plants, with the other 50% in hydroelectric dams. Wind is just extra. Solar is minimal. If there is wind, then it's free energy, if not, fire up the gas plant. From what I recall, most of the year compensation is done by hydroelectric dams. The critical period is mid to late summer, when A/C power ramps up consumption and when dams hit their low levels. During this period, compensation is done by gas/coal plants.
Anyhow, if there is a lesson to be learned by the US here, it's not about renewable energy. It's about the power grid. The article doesn't mention it, but upgrading the power grid, which happened in the late 80s is the key to this project. The electric grid here is incredibly efficient and a feat of engineering EDP should be proud of. It is fully automatic, balances out production from a myriad of extremely unreliable sources, prioritizing renewable sources and using dams and gas/coal plants to compensate whenever needed (be it by storing energy or ramping up production). A brownout is unheard of, while ground shunting losses are minimal. It is indeed a smart grid, as all should be.
If you have a good electric grid, then whenever renewables hit the economic sweet spot they will get built. It's just that for us, the sweet spot is higher because we have no oil, no gas and limited coal in the territory.
We built lots of hydroelectric dams in the 70s and 80s, so we use that as energy stores. We built secondary dams downriver, pool water there and pump up when wind overgenerates energy. It's not incredibly efficient, but it's better than shunting the energy into the ground or stopping the wind farms.
Answering you: No, they are not self-sufficient in the sense that they are independent from the main grid. They are self-sufficient because they produce more than they consume. Doing a piece of electric network isolated wouldn't cross our mind. The country's electric grid is what North-Americans call a 'smart-grid' and is something every country in the world should have. It easily balances load accross the country, starting and stopping hydro plants and in the limit starting and stopping natural gas plants in peak season (typically late summer: depleted dams, lots of A/C running). It even automatically manages importing and exporting energy from/into Spain and France.
Your whole argument hinges on high transmission losses, which don't exist. Electric energy transmission is very efficient. For short distances (100km) it's about 98% efficiency, and even for extreme distances you won't dip below 90%. Portugal used to import electricity from France. Electricity traveled 2000km between the nuclear power plant and the grid connect point in Sines, and it's not a remarkable feat. I doubt you can't feed boston from anywhere on a 3000km radius. You can get into flyover-state territory with 3000km...
You managed to read the article with a large bias. It *does* explain how to deal with the issues of fragmentation: a) guarantee APIs are forward compatible; b) have apps declare their hardware needs; c) minimize bugs/incompatible APIs using extensive testing.
Google does assume fragmentation is inevitable. That seems to be under discussion by some people here on /. Personally, I can't fathom how is fragmentation avoidable, unless by stagnation. Stagnation is quite the opposite of the Android ecosystem, which is evolving at a very fast pace (fast hardware and software release cycles).
If indeed fragmentation is inevitable, Google has seemingly defined the problem correctly and it looks like the solution is good.
It's not like there's no other option. Basically, as a developer, you must choose between a large user base now with an unknown probability of being kicked off with no warning, or a smaller user base (in the Android market) with clear acceptance rules. It's not an easy choice. Risk has rewards. Risk also causes shit to hit the fan. Just don't complain if/when it happens.
The problem with Android is the fragmentation.
Fragmentation is needed for a competitive environment. It's an added problem, partly for developers but mainly for google, and they are handling it quite well. Properly accounting for different hardware targets in both the hardware development and in the software development kits is a daunting task. However, and I feel everyone is repeating the fragmentation mantra without giving proper credit to Google, Android handles fragmentation quite well. Apps are always forward-compatible (write for 1.5 and you get ~100% compatibility with existing handsets), and they announce the hardware they need.
Do you need a camera? Declare it on your manifest, and the app appears on the market only to devices sporting a camera. Do you absolutely need multitouch? Declare it. Do you need an SD card? Declare it. The only drawback is that every requirement you add narrows down the range of devices your app appears in.
Would it be better if there were fewer devices all alike? It'd be like the narrowing decision would have already been made for you. Oh, right. That's the Apple way: Users are too stupid, let's decide for them.
In the end, it's different. It's not worse. It's more complicated for the developer, in exchange for a larger user base. Before anyone mentions there are more iPhones than Android devices, please first consider that: a) there are more Android devices than iPhones being sold today and; b) Android covers a much wider range of price-points, and is thus in reach of a much larger user base, so this tendency is likely here to stay (think how Nokia is still king of mobile handsets).
I have a Thinkpad X61t with a GM965 integrated intel video chip. Never had a problem with suspend/resume. It just works (under ubuntu now, and under Gentoo previously).
If something moving on a screen a row or two below your position is cause for distraction, I'd wager you have the wrong professors. Good teachers grab the students' attention and only let go after an hour or an hour and a half. I had one MBA professor that could keep the entire class engaged for a whole 5 hour period (with a 15min break at mid-time), but that's definitely an outlier.
Then, Google has complete control over what can be released, and can even uninstall apps that were previously downloaded.
This is a blatant lie. Android even allows for marketplaces other than Google's.
Google does control whatever goes into the Android Marketplace. It has no control whatsoever over the devices. I'm ok with that. If they bork the Android Marketplace, I'm sure an alternative will popup.
And herein lies the confusion between cause and condition. Speed is rarely the cause for an accident. It is always a condition. The trick is, acting on causes has much better effects than acting on conditions.
You have a simplistic view, which I don't hope to change. Please take my opinion as a call for more information. Go read on best cases in death toll reductions in Europe (Portugal's IP5 road comes to mind). Hint: The action focus was not speed.
Now, go RTFA...
http://en.wikipedia.org/wiki/Forming-storming-norming-performing
Now that I'm at it, team phases require different leadership styles. Goleman is a good reference:
http://www.collaborativenetworks.net/jeff/downloads/htm/handout.htm
Integer problems are in general quite complex to solve. Not the US election, which was the case here. It's easily approachable as a linear continuous problem, then constrain critical variables and then a brute-force search on the 10 or so critical states.
You got it wrong. Let me correct the text for you:
Linear problems are easy-peasy. A 50-simplex is solvable by-hand.
P.S. The US election result optimization is a linear problem.
Might as well grab a list of all induhviduals and go throught time kicking their gender-appropriate gonads in proper alphabetical order.
When you buy an iPhone and judge if the price is right, do you take into consideration the existence of a power brick in the package? You don't, and nobody else does. The perceived value of the product is elsewhere, and that reduction in cost will never cause a reduction in price.
I do know that part of the issue is power dissipation. When you're converting 5V (from usb) to what the charger circuit in the cell phone needs, the difference ends affecting power dissipation (in the form of heat). Voltage difference between power source and charger times max current draw = amount of power you need to dissipate at that stage (or simply P=I*V). A charger made specifically with that battery in mind would minimize power wasted (on charging) and would mean less electronics on the phone.
Completely false. Transformers do have losses (namely eddy current) and do produce heat. However, it's nowhere near I*deltaV. You'd see high to medium-voltage and medium to low-voltage transformers in the power grid blowing up everywhere if your proposition was anywhere near reality.
Faulty logic. Sale price is not related to production cost (other than the condition that you must be able to sell at a profit or you'll go bust). Reduced cost will produce higher margins and a larger profit on the initial product. It'll also mean the end to the "extra chargers" market.
We're talking embedded devices here. Stuff like my wireless router. Try to do a DNS poisoning attack or a man-in-the-middle when I'm typing the IP in the address bar and have the device on my network segment...