Next time you're taking off or landing from a major airport, take a look at the suburbs. Hundreds of thousands (heck, probably millions) of square meters of roof, mostly unshaded. Here in Denver, maximum solar flux (noon in June) is about a kilowatt per square meter. Typical household peak load over an hour is about 12 kilowatts, average load over the course of the day is closer to a single kilowatt. I've got maybe 20 square meters of south facing roof (and about the same west facing). 50% efficiency would have a peak output of about 15 kilowatts. Peak hourly load for a typical household runs about 12 kilowatts. If I had adequate storage, I could be nearly self-sufficient. Even without, the local utility is required to accept my excess and run my meter backwards -- I could come close to a zero electric bill. At least for the southwestern US, 50% efficient cheap solar cells would go a long ways towards making the suburbs quite energy efficient.
Now if you want to make the suburbs really happy, find a way to take that excess electricity and efficiently combine water and carbon dioxide to get liquid hydrocarbons that can run an ICE:^)
The US needs to hand over some control of the root servers and Europe needs to trust the US a little more...
It would be nice if someone gave us a better description of what the rest of the world thinks "hand over some control" actually means.
As I understand it, the protocol for communicating between the root servers allows for a maximum of 13 names (A-M).
Is the rest of the world asking that the US turn responsibility for servers with those names over to them, eg, France will be responsible for "K"?
The real responsibility of the root servers is to point to the servers responsible for the top-level domains (eg,.com).
I would be very, very surprised if France was not already in control of the servers responsible for the.fr TLD.
Is the rest of the world asking that they be given responsibility for some of the non-country TLDs (eg,.com)?
The US courts periodically get involved in resolving ownership of names, infringement of copyright, etc.
If I were to somehow gain ownership of cocacola.com and use it to advertise my own cola products, the US courts would quickly find me to be infringing on the trademark and make me stop.
Is the rest of the world asking that such legal resolution be moved elsewhere?
Methanol is a good choice for fuelling cars too, since it generates more power than gasoline...
Bzzzzt!
Methanol contains only about half as much energy per liter as gasoline.
Here's a summary table of several fuels.
Maintaining the same driving range requires a tank that's twice as big.
Obtaining the same power output from a fuel-injected ICE will require
injecting twice as much each cycle.
It is potentially useful in a fuel-cell to power an electric car,
which can overall be more efficient at extracting energy than an ICE,
but it will have to be twice as efficient in order to deliver the same MPG.
Assuming comparable ICE power plants,
methanol has to be priced at half the price of gasoline
to deliver the same miles-per-dollar.
You might need to do some creative googling to find sources of dry ice, but last time I was looking it was fairly cheap
At least here in the US, every time I've needed dry ice,
I was able to get it at my local supermarket.
Stop at the customer service desk and tell them what you want.
Usually came in 10 or 20 pound blocks in some sort of cardboard wrapper.
Not sure exactly what they use it for --
meat lockers if the power goes out?
keeping frozen items cold enough in the delivery van? --
but they weren't adverse to selling it to me.
Take a pair of gloves with you, as
inadvertently laying the back of your hand against it is unpleasant.
It's an interesting if expensive alternative to
ice and salt in an ice cream churn.
I've tried mixing dry ice and water
in hopes of having the "fog rolling down the sides" effect
at the churn,
but have never been entirely satisfied with either the effect or
the resulting ice cream.
I am not surprised that the total number of wiretaps requested is small, and only mildly surprised that the number that were denied is zero (I would have expected a handfull, but not many more than that).
It would be interesting to see the historical statistics, but (a) the police learned that the judges were serious about the need to show cause for a wiretap and
(b) the bad guys learned to not say important things over the phone.
If I were a bad guy, and real-time voice communications with other bad guys was important, I would use public wifi access networks with a non-standard audio application (this is not rocket science today -- hell, it wasn't really rocket science when I was writing prototype apps incorporating real-time audio transport back in 1993) and strong encryption (readily available).
Best of luck to the FBI trying to tap such conversations at Starbuck's today, MacDonald's tomorrow, and the public library on alternate Wednesdays.
Didn't we just have an article about the decline of the US in sciences and engineering?
Tell me again how it will all be better if we just somehow get more people to study those subjects in school.
It's jobs, man.
Show me long-term stable growth in science and engineering jobs and I'll show you plenty of students willing to take on the challenges.
During the recession of the early 90s, US companies laid off employees by the thousands ever other week. During the past 5 years, US companies having been laying of employees by the tens of thousands.
Even if these people were being rehired in suitable jobs, there are other problems in lack of stability.
When you have to spend,
pretty much as a minimum,
except for the truly gifted,
four hard years for a BS,
six for an MS,
eight for a PhD, and
then more time really learning a specialty,
most people are looking for some stability in their job.
You read articles that suggest that today's college students should
expect to change,
not just jobs,
but CAREERS,
four or five times in their life.
Imagine being a new EE and being told that
every five or six years,
you need to go through a major change --
from IP network electronics to
power generation systems to
implantable medical devices.
I'll argue all day that darn few people
can even do it,
let alone look forward to it.
Between it and PL/1 I concluded I hated computers and didn't get back into them for almost 10 years.
Huh.
I learned APL a couple of years later, and it was what convinced me I really liked computers.
Perhaps it was because the numerical analysis and statistical modeling problems I was working on were well-suited to the APL notation, but even without that:
an interpreted language,
symbolic debugging,
garbage collection for intermediate memory use,
the ability to write self-modifying code.
Cool stuff, especially for the time.
I still use an APL interpreter on Linux in preference to a calculator for many casual operations...
Sure, other uses will be found for genetic markers... but... the biggest and most obvious use for this info is to deny health "insurance" coverage to people likely to need it. This quite simply isn't a factor in systems where everyone is entitled to health care.
Indeed, uses that have already started.
Women with a particular BRCA-1 marker have an 80% chance of developing breast cancer before age 65.
Assuming a cheap test,
most women should probably be tested.
Even with a more expensive test,
women with a family history should probably be tested.
Positive test results may indicate that differences in health monitoring and/or treatment are appropriate.
It seems likely that as tests for genetic markers become cheap,
they will be incorporated along with other mandatory blood tests performed on newborns --
eg, PKU screening.
I'm beginning to think that it's a national disgrace that
the US will be spending energy on the question
"How do we keep this technology from being used
to deny people health care?" while
the rest of the industrialized countries get to ask
"How can we use this technology to provide cost-effective health care?"
Even if the 'The big IF' situation you described (a little upstart having something Microsoft wants) comes true, then what will Billy do? Buy them.
In the long term, I would expect something more like the early days of electronics, and the price that little players can get for their patents will be quite small.
A handful of large corporations -- RCA, AT&T, Zenith, etc -- held extensive patent portfolios.
So large and broad, in fact, that it was difficult to build anything involving electronic circuits without infringing on one or more of them.
These big players had cross-licensing agreements with each other, involving no exchanges of money, so for example, RCA didn't have to worry about AT&T's patents.
But if you were a little player, it was a dangerous playing field.
Imagine you are a little player with a single valuable patent.
If none of the big players will license it from you, you get no revenue from it.
If none of the other little players can use it because they also have to infringe on big player patents, you still get no revenue from them it.
As you go broke, one of the big players makes you a low-ball offer that looks pretty darned good.
Once they own it, it's a freebie for all the other big players because of the cross-licensing.
Those costs are sunk costs. Only incremental costs over incremental benefits makes sense...
Indeed.
Of course, the incremental costs for Office for Linux are larger than simply porting code.
One area where there would be large costs would be in support.
To pick a simple example,
both Windows and MacOS provide a single widget set for user interfaces.
The widget set is guaranteed to be properly installed on any box running those.
The developers of those widget sets go to great lengths to provide backwards compatibility so that application software doesn't get broken across versions.
I even suspect that given the dominance of Office, those developers include testing against Office as part of their normal processes.
For Linux, MS would be faced with questions such as which widget sets, which versions, etc.
Automated installation is harder.
Customer support is harder.
The size of the market is uncertain, and it is not clear (at least to me) that Office for Linux would be profitable.
Astounding -- three year increase in expectancy in eight calendar years.
That certainly will put the bite on pension funds, both public and private.
But the life insurance companies ought to do well.
Compared to someone starting to draw Social Security in 1940, today's retiree at age 65 has an expected "years to live" that's only about six years longer, not 15.
==========
This may be true on average in the USA, but it directly contradicts a recent study in the UK which says that the average 65-year old male can now expect to live into his 80's.
A useful summary from several sources for the US is given
here.
The life expectancy for a white male turning 65 in 1940 was (by interpolation) just about 12.25.
The life expectancy for a white male turning 65 in 2005 would be (by interpolation and extrapolation) about 17.25.
So this year's 65-year-old can expect to make it to 82.
But 1940's 65-year-old could expect to make it to 77, which surprises a lot of people.
The bigger difference between then and now was the fraction of the cohort that made it to 65.
I would be surprised if the UK case was really that different -- can you provide a pointer to the study you mention?
Which brings up the point, do you really WANT to live 300 years? We already tend to go downhill after our 20's, and each decade after is compounded by more health problems. Now some people will claim that uber-nano technology, and some franken-science will keep us in great shape, but simply put; every part in our body wears out with time.
Even assuming that routine wear-and-tear can be defeated in some fashion,
what about the design capacity of that critical organ, the brain?
Heinlein touched on it in some of his works about long-lived people: is the storage capacity of human memory up to holding 300 years worth of data?
As one of Heinlein's characters comments at some point, "I'd spend a whole morning looking for the book that I wanted to finish reading, only to realize that that was 20 years ago."
Some of the symptoms of various dementias among the elderly suggest that storage is, in some ways, LIFO.
In those cases, the afflicted people regress back to their 20s or even teens.
The long-ago memories remain intact, but they lose everything else.
The proof of increased lifespan (and not just lack of infant mortality) is the potential crisis in pensions in the western world. People were expected to live only 5-10 years after retirement. People now routinely live 20-25 years and some even 30 or more.
Correct in principle, but not in the details.
Over the last 100 years, life expectancy at age 65 in the US has been increasing at a linear rate of about one month per year.
Other industrialized countries may have rates of increase somewhat higher or lower than that.
Compared to someone starting to draw Social Security in 1940, today's retiree at age 65 has an expected "years to live" that's only about six years longer, not 15.
Studies done in the early years of Social Security included this effect -- the 1940 forecasts of how long people would live in the 1990s were almost exactly correct.
A bigger effect on pension economics is the fraction of the population that reaches retirement age.
Industrialized countries are safer places than they were 100, or even 50, years ago.
Fewer people die in their 40s and 50s.
While more people live long enough to claim SS benefits, those same people also make contributions for many more years.
The SS "crisis" in the US is overstated.
An interesting place to start is the non-partisan Congressional Budget Office's "A 125-Year Picture of the Federal Government's Share of the Economy, 1950 to 2075", which can be found online here.
Using today's benefit formulas,
SS expenses as a fraction of the national economy are predicted to stabilize at just over 6% of GDP
(Figure 2 summarizes this nicely).
I claim that we can afford this.
According to this study, by 2075, by far the largest single component of the federal budget will be interest on the debt.
My understanding is that US law treats "telephone systems" differently to computer systems.
Indeed,
in the US,
telephone service and data service are regulated very differently.
At the present time, the FCC appears to me to be making its distinctions based on some basic attributes of the service that is being sold:
if the service uses numbers from the North American numbering plan for telephone service
(eg, 303.555.1212)
or some other country's telephone numbering plan,
and interconnects with the public telephone network,
they regard it as a telephone service and consider it subject to the requirements of that service:
E911 emergency capability, wiretap capability, etc.
There are public interest reasons for taking such a position.
The 911 number has been drilled into kids' heads;
there are dozens of cases annually where the little kid dials 911 and tells the operator "My mommy is sick", and saves her life.
Wiretap has a long and successful history;
every year there are convictions of criminals where information gained by legal wiretap is pivotal.
As someone who was writing prototype audio-over-IP applications more than 12 years ago
in order to help the telcos evaluate the competitive threat the technology posed
(my answer: along about 2002 this stuff becomes a significant threat to your basic business),
I say that the authorities are fighting a long-term losing battle,
at least on the wiretap front.
Writing such an application and including strong encryption is not a hard problem.
Attempting to regulate every application that includes two-way audio is just nuts.
Are you going to make Counterstrike servers include a wiretap capability?
If I were a bad guy,
my colleagues and I would talk using
an independently-developed audio-over-IP application that did not use either SIP or H.323, and
that did include strong encryption.
There are, of course, additional techniques that could be used to make things even harder to track.
I'm very pro-nuclear power. But this is a serious issue. Honestly, it'd be less of an issue if we had a reasonable method of handling our own security other than "lets bomb everyone we are a bit nervous about". But we don't have that, and thus the practicality issue looms large.
I'm pro fission power too.
If I were a developing country,
I'd be even more pro fission,
and pro breeder reactors,
since it appears to be perhaps the only technology that
lets me grow my energy use up to the level of the developed economies and sustain it over the long term.
One of the ways we could address the security concerns is if we had solved the engineering problems of the thorium fuel cycle,
since it is much harder to get weapons grade material from that as compared to a uranium cycle.
Instead,
we shut down our experimental thorium breeder in 1982.
Assume that in most places, the existing electricity grid will be used to shuffle power around the neighborhood -- and to allow access to the big-boy power generators who pick up the slack when the microgrids are net consumers.
Many states allow individuals to sell power back into the grid.
Some even require that the "price" paid for such power is the same retail that the small customer pays, so accounting is simply a matter of spinning the meter backwards.
Most of those states, however, require that the small generator monitor the commercial power,
and if it fails,
to quit pushing power into the grid.
The nominal reason is one of safety -- some of the repairs done to the local wiring are danerous if the wires are "live".
Granted, a sufficiently sophisticated distributed control system should be able to handle this.
But I'm not sure I'd want to put my repair staff at risk on that basis.
In order to use nuclear power in a widespread fashion, we'd relaly have to have fast breeder reactors, to extend the lifetime of our supply of fissionable materials.
The problem is that fast breeder reactors are perfect for making weapons-grade Plutonium too.
Two points.
If our military cannot adequately secure the transport of material between the commercial reactors and the reprocessing facilities, within our own borders, there's something wrong.
Thorium fuel cycles can also be used in breeders, produce much less plutonium, and the waste products that require long-term storage tend to much shorter half-lives.
But we're not operating even a single thorium-cycle research reactor.
Seriously, if you want to think in terms of 100-150 years, this is a solved problem, and without the need for stone tablets.
Pigment-based inks on acid-free paper.
Silver-based black and white photo chemistry on acid-free paper.
Stitched bindings, not glue.
Store in a trunk where there's neglible light.
Put the trunk in the attic of a house where it's reasonably safe from large amounts of water (rain or flood).
Civil War documents using these techniques have survived nicely to the present day.
The Bell Labs archives have Alexander Graham Bell's original laboratory notebooks, still easily legible.
To date, there are no reliable archival media for this length of time for audio or moving pictures.
Write it down.
Sketch it (as silver-based photographic materials are getting harder and harder to find).
And you can be the source material for the historians of 2155:^)
To abandon New Orleans would mean abandoning over 400 years of tradition, history, and a unique and quirky culture unlike anywhere else in the country.
Not to mention that a great deal of the middle of the US requires a sizeable port city somewhere near the mouth of the Mississippi.
Add up the tonnage and the value of cargo that passes through New Orleans and South Louisiana, and you get very large numbers indeed.
Enormous amounts of bulky goods are transported up and down the Mississippi, the Missouri, and the Ohio Rivers.
It is not practical to move the transhipment point, from ocean-going vessel to river-going barge, too far upstream.
...by using technology to change school kids' experience, develop a cheap and effective e-book reader. Roughly speaking,
A solid block of plastic capable of surviving being dropped regularly,
Integrated low-power high-contrast very-high-resolution (think 300 dpi or better) display that can be read under all lighting conditions where paper books can be read (I'm thinking some of the e-paper technologies might be good),
Low-power -- need to be able to run for an entire day without being plugged in (two days would be better),
Reasonable physical factor -- say, 8x12x1 inch, five pounds max, three pounds would be better.
Cheap enough the school system can just hand 'em out.
At the beginning of the semester, the teachers download each kid's books for the semester into the device.
Doesn't have to be a computer, does have to be a viable replacement for a book.
No more 3rd graders lugging 40-pound backpacks.
I recall a talk with a teacher once about kids who did well in elementary school and then seemed to hit a wall in junior high.
As she explained it, the instruction "style" changes fairly dramatically between the two.
In elementary school, children learn to read, but are not typically expected to acquire knowledge independently by reading -- all the material will be covered orally in class as well.
In junior high, class becomes more of a forum for discussing the material, and the children must acquire considerable material from the assigned reading.
Of course, this shift continues the higher you go in the education system (I recall graduate courses where you were expected to read and remember insane quantities of material, most of which would never be discussed in class, but all of which was fair game on exams).
Anyway, the point is that elementary school kids generally need the interaction with teachers every day.
Distance learning is possible, but makes things much more difficult for many of the young kids.
Unfortunately, the long-term solution to that particular energy problem is the same as it is for the kids' parents' commute:
higher-density housing,
more but smaller schools,
more kids within walking distance.
But that's a 20-60 year undertaking.
What has stiffled innovation? Ironically, buisness. Since the 80s an increasing importance has been placed on short term profits. This shift in focus from long-term to short-term gains conincides with the shift in executive compensation packages tied to stock performance. The need to cut costs and increase profits always existed, and rightfully so, but that drive has now caused many industries to cut of their own nose to spite their face.
Indeed.
Thank you for pointing this out.
When Andy Grove was leading Intel, and was saying things about the US school system not producing enough engineers, one wanted to ask him: And what have you and Intel done to encourage engineering students?
No, not how many partial scholarships have you sponsored.
No, not how much equipment you may given to departments.
How many engineers, net, have you added to your work force?
And how many of those were in the US?
Have you checked, lately, to see how many US citizens can obtain permission to work in India?
In China?
In Israel?
It is up to business,
including giants like Intel and little ones,
to create engineering and science jobs.
By that measure,
looking about,
it would seem that rather than too few engineering and science students,
the US has too many!
Slashdot Mirror
Now if you want to make the suburbs really happy, find a way to take that excess electricity and efficiently combine water and carbon dioxide to get liquid hydrocarbons that can run an ICE :^)
It would be nice if someone gave us a better description of what the rest of the world thinks "hand over some control" actually means. As I understand it, the protocol for communicating between the root servers allows for a maximum of 13 names (A-M). Is the rest of the world asking that the US turn responsibility for servers with those names over to them, eg, France will be responsible for "K"? The real responsibility of the root servers is to point to the servers responsible for the top-level domains (eg, .com).
I would be very, very surprised if France was not already in control of the servers responsible for the .fr TLD.
Is the rest of the world asking that they be given responsibility for some of the non-country TLDs (eg, .com)?
The US courts periodically get involved in resolving ownership of names, infringement of copyright, etc.
If I were to somehow gain ownership of cocacola.com and use it to advertise my own cola products, the US courts would quickly find me to be infringing on the trademark and make me stop.
Is the rest of the world asking that such legal resolution be moved elsewhere?
Bzzzzt! Methanol contains only about half as much energy per liter as gasoline. Here's a summary table of several fuels. Maintaining the same driving range requires a tank that's twice as big. Obtaining the same power output from a fuel-injected ICE will require injecting twice as much each cycle. It is potentially useful in a fuel-cell to power an electric car, which can overall be more efficient at extracting energy than an ICE, but it will have to be twice as efficient in order to deliver the same MPG. Assuming comparable ICE power plants, methanol has to be priced at half the price of gasoline to deliver the same miles-per-dollar.
At least here in the US, every time I've needed dry ice, I was able to get it at my local supermarket. Stop at the customer service desk and tell them what you want. Usually came in 10 or 20 pound blocks in some sort of cardboard wrapper. Not sure exactly what they use it for -- meat lockers if the power goes out? keeping frozen items cold enough in the delivery van? -- but they weren't adverse to selling it to me. Take a pair of gloves with you, as inadvertently laying the back of your hand against it is unpleasant.
It's an interesting if expensive alternative to ice and salt in an ice cream churn. I've tried mixing dry ice and water in hopes of having the "fog rolling down the sides" effect at the churn, but have never been entirely satisfied with either the effect or the resulting ice cream.
I am not surprised that the total number of wiretaps requested is small, and only mildly surprised that the number that were denied is zero (I would have expected a handfull, but not many more than that). It would be interesting to see the historical statistics, but (a) the police learned that the judges were serious about the need to show cause for a wiretap and (b) the bad guys learned to not say important things over the phone. If I were a bad guy, and real-time voice communications with other bad guys was important, I would use public wifi access networks with a non-standard audio application (this is not rocket science today -- hell, it wasn't really rocket science when I was writing prototype apps incorporating real-time audio transport back in 1993) and strong encryption (readily available). Best of luck to the FBI trying to tap such conversations at Starbuck's today, MacDonald's tomorrow, and the public library on alternate Wednesdays.
Didn't we just have an article about the decline of the US in sciences and engineering? Tell me again how it will all be better if we just somehow get more people to study those subjects in school. It's jobs, man. Show me long-term stable growth in science and engineering jobs and I'll show you plenty of students willing to take on the challenges.
Even if these people were being rehired in suitable jobs, there are other problems in lack of stability. When you have to spend, pretty much as a minimum, except for the truly gifted, four hard years for a BS, six for an MS, eight for a PhD, and then more time really learning a specialty, most people are looking for some stability in their job. You read articles that suggest that today's college students should expect to change, not just jobs, but CAREERS, four or five times in their life. Imagine being a new EE and being told that every five or six years, you need to go through a major change -- from IP network electronics to power generation systems to implantable medical devices. I'll argue all day that darn few people can even do it, let alone look forward to it.
Huh. I learned APL a couple of years later, and it was what convinced me I really liked computers. Perhaps it was because the numerical analysis and statistical modeling problems I was working on were well-suited to the APL notation, but even without that: an interpreted language, symbolic debugging, garbage collection for intermediate memory use, the ability to write self-modifying code. Cool stuff, especially for the time.
I still use an APL interpreter on Linux in preference to a calculator for many casual operations...
Indeed, uses that have already started. Women with a particular BRCA-1 marker have an 80% chance of developing breast cancer before age 65. Assuming a cheap test, most women should probably be tested. Even with a more expensive test, women with a family history should probably be tested. Positive test results may indicate that differences in health monitoring and/or treatment are appropriate. It seems likely that as tests for genetic markers become cheap, they will be incorporated along with other mandatory blood tests performed on newborns -- eg, PKU screening.
I'm beginning to think that it's a national disgrace that the US will be spending energy on the question "How do we keep this technology from being used to deny people health care?" while the rest of the industrialized countries get to ask "How can we use this technology to provide cost-effective health care?"
In the long term, I would expect something more like the early days of electronics, and the price that little players can get for their patents will be quite small.
A handful of large corporations -- RCA, AT&T, Zenith, etc -- held extensive patent portfolios. So large and broad, in fact, that it was difficult to build anything involving electronic circuits without infringing on one or more of them. These big players had cross-licensing agreements with each other, involving no exchanges of money, so for example, RCA didn't have to worry about AT&T's patents. But if you were a little player, it was a dangerous playing field.
Imagine you are a little player with a single valuable patent. If none of the big players will license it from you, you get no revenue from it. If none of the other little players can use it because they also have to infringe on big player patents, you still get no revenue from them it. As you go broke, one of the big players makes you a low-ball offer that looks pretty darned good. Once they own it, it's a freebie for all the other big players because of the cross-licensing.
Indeed. Of course, the incremental costs for Office for Linux are larger than simply porting code. One area where there would be large costs would be in support. To pick a simple example, both Windows and MacOS provide a single widget set for user interfaces. The widget set is guaranteed to be properly installed on any box running those. The developers of those widget sets go to great lengths to provide backwards compatibility so that application software doesn't get broken across versions. I even suspect that given the dominance of Office, those developers include testing against Office as part of their normal processes. For Linux, MS would be faced with questions such as which widget sets, which versions, etc. Automated installation is harder. Customer support is harder. The size of the market is uncertain, and it is not clear (at least to me) that Office for Linux would be profitable.
Astounding -- three year increase in expectancy in eight calendar years. That certainly will put the bite on pension funds, both public and private. But the life insurance companies ought to do well.
==========
This may be true on average in the USA, but it directly contradicts a recent study in the UK which says that the average 65-year old male can now expect to live into his 80's.
A useful summary from several sources for the US is given here. The life expectancy for a white male turning 65 in 1940 was (by interpolation) just about 12.25. The life expectancy for a white male turning 65 in 2005 would be (by interpolation and extrapolation) about 17.25. So this year's 65-year-old can expect to make it to 82. But 1940's 65-year-old could expect to make it to 77, which surprises a lot of people. The bigger difference between then and now was the fraction of the cohort that made it to 65. I would be surprised if the UK case was really that different -- can you provide a pointer to the study you mention?
Even assuming that routine wear-and-tear can be defeated in some fashion, what about the design capacity of that critical organ, the brain? Heinlein touched on it in some of his works about long-lived people: is the storage capacity of human memory up to holding 300 years worth of data? As one of Heinlein's characters comments at some point, "I'd spend a whole morning looking for the book that I wanted to finish reading, only to realize that that was 20 years ago." Some of the symptoms of various dementias among the elderly suggest that storage is, in some ways, LIFO. In those cases, the afflicted people regress back to their 20s or even teens. The long-ago memories remain intact, but they lose everything else.
Correct in principle, but not in the details. Over the last 100 years, life expectancy at age 65 in the US has been increasing at a linear rate of about one month per year. Other industrialized countries may have rates of increase somewhat higher or lower than that. Compared to someone starting to draw Social Security in 1940, today's retiree at age 65 has an expected "years to live" that's only about six years longer, not 15. Studies done in the early years of Social Security included this effect -- the 1940 forecasts of how long people would live in the 1990s were almost exactly correct. A bigger effect on pension economics is the fraction of the population that reaches retirement age. Industrialized countries are safer places than they were 100, or even 50, years ago. Fewer people die in their 40s and 50s. While more people live long enough to claim SS benefits, those same people also make contributions for many more years.
The SS "crisis" in the US is overstated. An interesting place to start is the non-partisan Congressional Budget Office's "A 125-Year Picture of the Federal Government's Share of the Economy, 1950 to 2075", which can be found online here. Using today's benefit formulas, SS expenses as a fraction of the national economy are predicted to stabilize at just over 6% of GDP (Figure 2 summarizes this nicely). I claim that we can afford this. According to this study, by 2075, by far the largest single component of the federal budget will be interest on the debt.
Indeed, in the US, telephone service and data service are regulated very differently. At the present time, the FCC appears to me to be making its distinctions based on some basic attributes of the service that is being sold: if the service uses numbers from the North American numbering plan for telephone service (eg, 303.555.1212) or some other country's telephone numbering plan, and interconnects with the public telephone network, they regard it as a telephone service and consider it subject to the requirements of that service: E911 emergency capability, wiretap capability, etc. There are public interest reasons for taking such a position. The 911 number has been drilled into kids' heads; there are dozens of cases annually where the little kid dials 911 and tells the operator "My mommy is sick", and saves her life. Wiretap has a long and successful history; every year there are convictions of criminals where information gained by legal wiretap is pivotal.
As someone who was writing prototype audio-over-IP applications more than 12 years ago in order to help the telcos evaluate the competitive threat the technology posed (my answer: along about 2002 this stuff becomes a significant threat to your basic business), I say that the authorities are fighting a long-term losing battle, at least on the wiretap front. Writing such an application and including strong encryption is not a hard problem. Attempting to regulate every application that includes two-way audio is just nuts. Are you going to make Counterstrike servers include a wiretap capability? If I were a bad guy, my colleagues and I would talk using an independently-developed audio-over-IP application that did not use either SIP or H.323, and that did include strong encryption. There are, of course, additional techniques that could be used to make things even harder to track.
I'm pro fission power too. If I were a developing country, I'd be even more pro fission, and pro breeder reactors, since it appears to be perhaps the only technology that lets me grow my energy use up to the level of the developed economies and sustain it over the long term. One of the ways we could address the security concerns is if we had solved the engineering problems of the thorium fuel cycle, since it is much harder to get weapons grade material from that as compared to a uranium cycle. Instead, we shut down our experimental thorium breeder in 1982.
Assume that in most places, the existing electricity grid will be used to shuffle power around the neighborhood -- and to allow access to the big-boy power generators who pick up the slack when the microgrids are net consumers. Many states allow individuals to sell power back into the grid. Some even require that the "price" paid for such power is the same retail that the small customer pays, so accounting is simply a matter of spinning the meter backwards. Most of those states, however, require that the small generator monitor the commercial power, and if it fails, to quit pushing power into the grid. The nominal reason is one of safety -- some of the repairs done to the local wiring are danerous if the wires are "live".
Granted, a sufficiently sophisticated distributed control system should be able to handle this. But I'm not sure I'd want to put my repair staff at risk on that basis.
Two points.
If our military cannot adequately secure the transport of material between the commercial reactors and the reprocessing facilities, within our own borders, there's something wrong.
Thorium fuel cycles can also be used in breeders, produce much less plutonium, and the waste products that require long-term storage tend to much shorter half-lives. But we're not operating even a single thorium-cycle research reactor.
Seriously, if you want to think in terms of 100-150 years, this is a solved problem, and without the need for stone tablets. Pigment-based inks on acid-free paper. Silver-based black and white photo chemistry on acid-free paper. Stitched bindings, not glue. Store in a trunk where there's neglible light. Put the trunk in the attic of a house where it's reasonably safe from large amounts of water (rain or flood). Civil War documents using these techniques have survived nicely to the present day. The Bell Labs archives have Alexander Graham Bell's original laboratory notebooks, still easily legible. To date, there are no reliable archival media for this length of time for audio or moving pictures. Write it down. Sketch it (as silver-based photographic materials are getting harder and harder to find). And you can be the source material for the historians of 2155 :^)
Not to mention that a great deal of the middle of the US requires a sizeable port city somewhere near the mouth of the Mississippi. Add up the tonnage and the value of cargo that passes through New Orleans and South Louisiana, and you get very large numbers indeed. Enormous amounts of bulky goods are transported up and down the Mississippi, the Missouri, and the Ohio Rivers. It is not practical to move the transhipment point, from ocean-going vessel to river-going barge, too far upstream.
At the beginning of the semester, the teachers download each kid's books for the semester into the device. Doesn't have to be a computer, does have to be a viable replacement for a book. No more 3rd graders lugging 40-pound backpacks.
Unfortunately, the long-term solution to that particular energy problem is the same as it is for the kids' parents' commute: higher-density housing, more but smaller schools, more kids within walking distance. But that's a 20-60 year undertaking.
You mean like this?
Indeed. Thank you for pointing this out. When Andy Grove was leading Intel, and was saying things about the US school system not producing enough engineers, one wanted to ask him: And what have you and Intel done to encourage engineering students? No, not how many partial scholarships have you sponsored. No, not how much equipment you may given to departments. How many engineers, net, have you added to your work force? And how many of those were in the US? Have you checked, lately, to see how many US citizens can obtain permission to work in India? In China? In Israel? It is up to business, including giants like Intel and little ones, to create engineering and science jobs. By that measure, looking about, it would seem that rather than too few engineering and science students, the US has too many!