...boss could use it to walk into your cubicle even when he's on a trip to Japan...
Obvious meaning - boss flies to Japan, uses virtual reality suite in Japan and telepresence robot in office to walk into your cubicle.
Alternative meaning - boss uses telepresence to make his trip to Japan. So there is no problem taking off the telepresence head set and having a walk round the office to see what the workers are up to, while he is "in Japan"
As I explain in my essay about how telepresence will displace business air travel:
Telepresence is not teleconferencing. The telepresence camera is
under the control of the near end, not the far end. It can roam
and see things it is not meant to see. It is a tool for bosses not
workers. The location is laid open to the inspection of the
visiting dignitary, while the presence itself reveals nothing of the
near end.
I say that Blackboard's system is notorious for being insecure. When I'm asked what is wrong with it, I whip out my copy of the cease and desist letter. "More than my jobs worth to tell you", I say. "but it will void your fire insurance if you buy a security system from Blackboard".
One less bidder for me to compete with.
Hey, wait a minute. Insurance contracts are based on utmost good faith. If you don't tell your insurers about problems with your access control system, they can refuse to pay. If you know that there are problems, but cannot find out what they are, you are obliged to inform your insurer, who may wish to alter your premium. I remember after 9/11, the insurance was limited to $3.5billion per incident. The insurers went to court to claim that the two planes should be considered a single incident, and thus halve the pay out.
If there is a fire and a big loss, flaws in the access control system that were concealed from the insurers could get real messy
If someone reads one of my posts, decides I'm not getting enough and types in a personalised email offering me Viagra, I don't call that spam. Sure it is a potential problem, but it is not the actual problem we have today.
Todays problem is that the emails are written by computers and then injected into a channel intended for humans. So there is an endless flood. Humans can no longer read email in person. They have to use programs to filter out the crap injected into the system by other programs. The spammers crime is making email unusable by humans. They are the enemies of free speech.
The ISPs are obliged to block the sites, so no-one can view them, even if they had the list. Mind you, seeing www.ashcroftsucks.org on the list might raise suspicions
Copyright is inefficient due to under-use of intellectual property. E.g. Photoshop is priced thousands of times higher than the marginal cost of production. Society would be better off in the short run if it did not have to pay a monopoly price.
Copyleft is inefficient due to under-production of intellectual property. In particular it struggles to solve this collective action problem: If there is a boring piece of software that a million users would each pay a $100 for, and it would cost $10million to get it written it, it is in societies interest for it to be written.
Alan Greenspan dodged. He could have stated the dilemma plainly, but chose instead to waffle. Oh, well, perhaps another time.
Double the production volume, take a 20% cut in cost. With a bit of logarithm bashing:
first car $1,000,000 ten cars $ 476,500 one hundred cars $ 227,061 one thousand cars $ 108,197 ten thousand cars $ 51,557, one hundred thousand cars $ 24,567 one million cars $ 11,706
I've forgotten where I got that rule of thumb from, it might be a semi-conductor industry one. Look at the hypothetical car prices. That is about how it goes, from prototypes through hand built luxury to large volume production. It is not a bad rule of thumb.
It is utterly different from the economics of software
Technological advance is continually altering the shape and nature of our economic
processes and, in particular, is promoting the trend toward increasing
conceptualization of U.S. GDP.
What he means is:
Technological advance is continually altering the shape and nature of our economic
processes and, in particular, is promoting the trend towards an increasing proportion of our
wealth existing as knowledge.
How does one measure wealth? Good question. Next. How does one measure GDP? Well, one comes up with an ammount in dollars. If we monetize knowledge, then indeed current trends lead towards an "increasing
conceptualization of U.S. GDP." But will we be better off? Talking of GDP instead of wealth begs an important question.
The consumption of physical property is rivalrous: if I eat your sandwich, you go hungry. The consumption of ideas is non-rivalrous: even if I use calculus all the time, your use of it is uneffected, you don't have to wait your turn. If the Internet allow dissemination of ideas at negligible cost, then knowledge becomes an intensive property, like temperature or pressure, and is not an extensive property, like mass, energy, or entropy. Given that the consumption of physical property is rivalrous, it makes perfect sense to go round counting things, and saying that ten automobiles are worth ten times as much as one automobile. Conversely, it is complete gibberish to claim that we are twice as well off becase Newton and Leibnitz both invented calculus, thus giving us twice as much calculus.
Greenspan's emphasis on GDP ducks the challenge of measuring wealth that is tied up in intensive rather than extensive qualities.
The only argument I've seen against an anti-spam law is concerns over freedom of speech. These concerns are easily met. The law need only say:
Port 25 (the Simple Mail Transfer Protocol port) is reserved for human to human communication. It is an offence to cause or permit a robot to post to multiple messages to it.
It is easy to craft an exception for mailing lists that do the three step, subscribe, validate, confirm thing.
First ammendent jurisprudence lets the government regulate speech provided its regulations are narrowly tailored to serve a compelling government interest. What kinds of things count as compelling the US government? Other rights in the Bill of Rights, and laws of physics. For example, if you have two radio transmitters broadcasting on the same frequency each will interfere with the other. This is the justification for Government regulation of the radio spectrum. One way of looking at it is that freedom of speech is played off against freedom of speech. It is hard to oppose government regulation of the radio spectrum on the grounds of freedom of speech if the consequence is that neither speaker will be heard. Equally, it would be very hard to argue in court for the admission of robots to the designated human-to-human channel on freedom of speech grounds when the cause of the regulation was that humans were finding it hard to talk to each other because their channel was being swamped by robot traffic.
Notice also that this proposal avoids violating the first ammendment because it focuses on solving the problem we have, and leaving alone a similar problem that we don't have. Specifically, some-one who wishes to offer you a penis enlarger remains at liberty to get your email address from any source and type in and send an email to you, with their offer. They may repeat this, obtaining somebody elses email address, typing in a fresh message, and sending them a similar offer. They may repeat this as long as their stamina holds out and their fingers will keep tapping away at the keyboard. All that is forbidden is automating the process and then using the humans only port.
Governments spend billions on software. That money comes from taxation. So the revenue raising side of that is already in place and taking money out of your pocket.
The problem lies in the licences that Governments sign up to. The money that the Government raises in taxes is paying for software development, but it goes via companies who own the software. Government could spend some of the money it already raises to spend on software development and use it to develop GPL software.
For example, Government offices need an office suit of some kind, and indeed they have one. Surely the Government office suit should come bundled with Linux for all citizens to use, and perhaps contribute too. We have already paid the taxes!
What do these folks think, seeing a nipple
or the occasional double entry will mutate
their kids into criminals?
Damn sure it will. It was youthful exposure to double entry
that set the likes of Kenneth Lay and Robert Maxwell on the road to committing serious crimes.
For the sake of our children we must block access to dubious sites like this, full of references to double entries.
He is assuming that you can just multiply the word by word probabilities together. This is a standard assumption. If you don't do something like this you get a combinatorial explosion, just like you said. More to the point, if you don't do something like this, your data becomes sparse. In the limit of making no assumptions you are reduced to recognising only the spam you have already seen, you have no capacity for generalisation and all the new spam gets through. No statistical method is any use if it doesn't generalise. Any method that works in practise has some kind of assumption hidden inside to make it go.
One reason I like the Bayesian approach is that it is pretty transparent. When an implementation is making the independence assumption, it is clearly apparent, and if you need to relax the assumption, for example by looking at word pairs, it is clear enough how to go about it. Graham does discuss this towards the end of his article.
Often the main effect of the independence assumption in practise is to exaggerate the confidence with which the classifier classifies things. Since Graham is not using his probabilities as input to subsequent processing he gets away with this,
The big technological road block is getting from the script to the emotions. With an actor, the director can give directions such as "No, no, your character is cereberal. When he gets angry it has to be cold and steely, less is more." Without an actor, the CGI puppet master has to spend hours tinkering to try to get the right effect.
So I see a battle, CGI puppet masters versus actors. A good CGI puppet master can do several characters. As the sole creator of ensemble pieces his work is freed from the constraints of the egos of the actors. The advantages of CGI props and settings can be exploited to the full without the difficulties of integrating with live action, or the expense of real props and location shooting. Against that, CGI puppet masters will slog for hours to get OK renditions of emotional nuances that come to actors, literally, naturally.
Modern armies have been moving away from having large numbers of soldiers. The Iraqi army in 1991 was about 3 million, suffered 200 000 casualties. Raw numbers didn't help.
Pilots in the airforce is another example. Unmanned aircraft are starting to become important to the military.
You seem to be telling me that the "hot intern" at
Whitehouse.com isn't staying at 1600 Pennsylvannia Avenue. Have things really changed that much since Clinton?
Let the flying area be 10 meters by 10 meters. 50mph is a little over 20 meters per second. Thus the volume of flow required is 2000 cubic meters per second.
Each cubic meter of air has a mass of about 1 kg, so it has a kinetic energy, half m v squared, of 1/2 x 1 x 20 x 20 = 200 Joules. We need 2000 cubic meters per second, so 400kW.
This is a bit more that 500hp, so the idea is not impractical because the fans would take too much power:-)
Put your tuning fork on the bridge of your violin. If the A-string is in tune it will resonate. Bring your little finger close, and as it touches, you not only get a loud buzz, but you can feel the string hitting your finger.
I thought sky divers wore baggy clothes, to reduce their terminal velocity, so that they could spend more time in free fall. How far can you push this? Could you wear a flying squirrel suit, with cloth between you arms and legs, to get your terminal velocity down to say 50mph?
If that worked, could it be the basis for a fairground ride: Enourmous fans blast air upwards at 50mph. Customers wear flying squirrel suits, and get to fly about in the air stream, a bit like being a bird.
Do the Beeb, the Guardian, and the Times give you the story behind the story?
Like me, you have spotted that large government debts create a rentier class, a new "landed" gentry, into whose sevice we have sold our children. And, if you are like me, you didn't get the idea from the Beeb, the Guardian, or the Times; you had to work it out for yourself.
I resented paying the licence fee for crap BBC programs so much that I got rid of my TV, so that I didn't have to pay it any more.
The BBC news reports wind me up. They put the same spin on stories as every-one else. They put positive spin on rising house prices.(The cost of housing is a cost, rising house prices are bad news) They commit dreadful economic howlers, saying that shortages occur inspite of governments capping prices, rather than because governments cap prices, and hoping that trade-protection will save jobs (<sarcasm>much like it did in the 1930's</sarcasm>)
The documentaries have been dumbed down. Now they are so spun out that they are unwatchably tedious. I can remember the James Burke programs from twenty years ago that were a stunning torrent of information and insight.
I think there are two key problems with the BBC.
1)Its the same people. You work for the Guardian for a bit, move to the BBC, move to ITV. Its the same employees who don't know any better.
2)They feel they have to justify the licence fee by keeping the viewing figures up. So if the fashion is for celebrity gossip in the news slot, the BBC competes with ITV for who can get most viewers for their celebrity gossip. They lack the faith in the concept of public service to stick to serious news and argue for the licence fee on the grounds that it is worthwhile to provide serious news even when it is out of fashionable.
I'm T-total, so it annoys me when public policy is discussed as though alcohol were not a drug, and drinking is not recreational drug use. The BBC spins alcohol as not-a-drug, just like the other broadcasters, even though the BBC doesn't carry adverts for alcohol, (or anything).
I never felt that the BBC took advantage of its independence of business to tell me secrets. It is really strange in the UK at the moment, because our two big parties, Conservative, and New Labour, are corporatist/big business soft right parties. 'Old' Labour was left wing, but even back in the eighties, when we had left versus right politics, the licence fee didn't do very much for editorial independence. The BBC depended on the politicians for the licence fee. The Conservative politicians listened to businessmen, the same businessmen who ran the companies that placed the adverts in other media. The Labour politicians listened to the workers who worked in the big unionised companies that placed the adverts in other media. If ITV took an anti-car stance, the advertisers of cars on ITV would bring it back on message. If the BBC took an anti-car stance, either the car workers union leader would talk to the Labour prime minister, or the managing director of Britsh Motor Corporation would talk to the Conservative prime minister. The BBC never developed a distinctive voice.
...I'm the one who ends up paying for stuff like guarding railroad tracks or onsite storage of nuclear waste to avoid moving it by train. So when I'm told that my taxes have to be spent because it is easy to derail a train, I say: let me see the documents. I want to be the judge of whether it is easy enough to be a real threat. It is not right that the government can whip up a scare and then ban the documents that would let me judge for myself whether it is for real or not.
Muscles are single acting; they can pull but they cannot
push. Joints are double acting; they are powered both
ways. The secret is that muscles come in antagonistic pairs,
one to flex the joint, and on the other side, one to extend it.
This is like the p-type and n-type transistors in
CMOS. Turn on the n-type transistor to pull the output
low. Turn on the p-type transistor to pull the output
high. Both off at the same time for tri-state. Both on at
the same time to short out the power supply and blow up the
chip. In the human body, both muscles are off in the relaxed
state. If both muscles are slightly on, this pre-stresses
the tendons, taking up any slack, and effectively stiffening
the joint. This is what you do for exacting fine work, e.g.
embroidery or surgery. This is why such work is tiring, even
when the external forces you exert are small. For ordinary
work, you must coordinate your muscles so that they are non-overlapping.
When you type on a mechanical typewriter, you push
hard. I've just been measuring my old Olivetti Lettera
22. The keyboard is open underneath so it is a simple matter
to dangle an icecream tub underneath and fill it with water
until the rachet clicks to advance the carriage. 1.12
kg. 17mm of key travel. (* 9.81 1.12 0.017) = 186mJ. If you
are typing 30 four letter words a minute that is
(* (/ 30 60.0) (+ 4 1) 0.186) = 0.464 W. It is not hard
physical work.
1.12 kg, say 10N, feels like a lot if you are not used to
it, but the significance lies elsewhere. It is way more than
the force exerted by the relaxed tone in the muscles that
control your finger. So to type a character you turn a
flexor full on, and turn it back off again. The typewriter
is geared at about 6 to 1, much like a piano, so the hammer
is flying pretty fast. Its momentum slams it into the paper,
making the impression and the rebound and the little coil
spring in the typewriter bring your finger and the key back
up. You literally never lift a finger. The springs
in the machine lift your fingers for you. You can type with
the extensor muscles relaxed all the time. Touch typing on a
manual typewriter requires alot of coordination, but it does
not require every kind of co-ordination. In particular you
do not have to co-ordinate your flexors and extensors to
avoid having them both on at the same time, because you
never turn on your extensors at all.
A modern mouse is very different. If you just plonk your
hand down on top of it you click all three buttons. You have
to use you extensors to not click. When I restarted
using a computer after a lengthy illness, I rapidly got
pains in my arms, from holding my fingers off the mouse
buttons all the time. I had to learn to be just tense enough
to stop the natural curl of my fingers from clicking the
buttons. What happens when I click a mouse button? What is
supposed to happen is that the extensor is turned off then
the flexor is turned on, then the flexor is turned off, then
the extensor is turned back on, so that they don't
overlap. I've not done any electro-myography, but I don't
believe it is happening like that. Briefly relaxing a muscle
that is kept tense most of the time is difficult and time
consumming. I bet that the flexor is turned on hard to over
come the extensor. How much damage does that do? It probably
depends alot on the office environment. If you are generally
relaxed and have only just enough tension in your extensors
to avoid accidental mouse clicks, I cannot see it doing much
harm. If work is fraught, and you tense up to avoid
mistakes, beware. The forces exerted when your flexors and
extensors are on at the same time add up internally, but
cancel externally. You might think that you cannot be
stressing your tendons because the switches have a light
action and you are not exerting much force, but if
that force is the difference between the force exerted by
the flexor and the extensor, your tissues might be under a
great deal of internal mechanical stress.
I suspect that much the same goes for a modern
keyboard. You have to actively lift your fingers off the
keys after the stroke. You don't have the option of
flexor-only typing. So when work gets hectic and pressured,
and your coordination is not 100%, you get flexors working
harder to overcome extensors that are not being fully turned
off, and lots of internal mechanical stress.
My theory is that these internal stresses are larger than
with a clnky old mechanical typewriter and are the cause of
RSI.
How can one find out if this theory is true? One way is
to get a researcher interested enough in this theory to use
electro-myography to find out if both muscles are indeed
being turned on at the same time. Another way is to get a
keyboard and a mouse with `heavy' long travel keys. This
would make sense in a prospective study, in which you equip
half a cohort of new users with the clunky mouse and key
board, and follow up after five years to see who has RSI and
who hasn't. It doesn't make much sense as a treatment. If
you have learned to type on a light keyboard with your
extensors turned on, the extra force needed to operate a
heavier keyboard might be translated by habit into more
activation of the extensors as well as more activation of
the flexors. I cannot see a heavier keyboard in itself
working as therapy, unless the sufferer can learn the flexor
only typing technique it permits, and avoid falling back
into flexor/extensor overlap habits when work gets
hectic.
Slashdot Mirror
Alternative meaning - boss uses telepresence to make his trip to Japan. So there is no problem taking off the telepresence head set and having a walk round the office to see what the workers are up to, while he is "in Japan"
I posted a link earlier.
The height is only root three over two of the side length. It would take a 16% lip to make it impossible for the cover to fall down the whole,
I say that Blackboard's system is notorious for being insecure. When I'm asked what is wrong with it, I whip out my copy of the cease and desist letter. "More than my jobs worth to tell you", I say. "but it will void your fire insurance if you buy a security system from Blackboard".
One less bidder for me to compete with.
Hey, wait a minute. Insurance contracts are based on utmost good faith. If you don't tell your insurers about problems with your access control system, they can refuse to pay. If you know that there are problems, but cannot find out what they are, you are obliged to inform your insurer, who may wish to alter your premium. I remember after 9/11, the insurance was limited to $3.5billion per incident. The insurers went to court to claim that the two planes should be considered a single incident, and thus halve the pay out.
If there is a fire and a big loss, flaws in the access control system that were concealed from the insurers could get real messy
If someone reads one of my posts, decides I'm not getting enough and types in a personalised email offering me Viagra, I don't call that spam. Sure it is a potential problem, but it is not the actual problem we have today.
Todays problem is that the emails are written by computers and then injected into a channel intended for humans. So there is an endless flood. Humans can no longer read email in person. They have to use programs to filter out the crap injected into the system by other programs. The spammers crime is making email unusable by humans. They are the enemies of free speech.
The ISPs are obliged to block the sites, so no-one can view them, even if they had the list. Mind you, seeing www.ashcroftsucks.org on the list might raise suspicions
The dilemma for GDP maximisers is easy to state:
Copyright is inefficient due to under-use of intellectual property. E.g. Photoshop is priced thousands of times higher than the marginal cost of production. Society would be better off in the short run if it did not have to pay a monopoly price.
Copyleft is inefficient due to under-production of intellectual property. In particular it struggles to solve this collective action problem: If there is a boring piece of software that a million users would each pay a $100 for, and it would cost $10million to get it written it, it is in societies interest for it to be written.
Alan Greenspan dodged. He could have stated the dilemma plainly, but chose instead to waffle. Oh, well, perhaps another time.
Double the production volume, take a 20% cut in cost. With a bit of logarithm bashing:
first car $1,000,000
ten cars $ 476,500
one hundred cars $ 227,061
one thousand cars $ 108,197
ten thousand cars $ 51,557,
one hundred thousand cars $ 24,567
one million cars $ 11,706
I've forgotten where I got that rule of thumb from, it might be a semi-conductor industry one. Look at the hypothetical car prices. That is about how it goes, from prototypes through hand built luxury to large volume production. It is not a bad rule of thumb.
It is utterly different from the economics of software
Greenspan says:
What he means is: How does one measure wealth? Good question. Next. How does one measure GDP? Well, one comes up with an ammount in dollars. If we monetize knowledge, then indeed current trends lead towards an "increasing conceptualization of U.S. GDP." But will we be better off? Talking of GDP instead of wealth begs an important question.The consumption of physical property is rivalrous: if I eat your sandwich, you go hungry. The consumption of ideas is non-rivalrous: even if I use calculus all the time, your use of it is uneffected, you don't have to wait your turn. If the Internet allow dissemination of ideas at negligible cost, then knowledge becomes an intensive property, like temperature or pressure, and is not an extensive property, like mass, energy, or entropy. Given that the consumption of physical property is rivalrous, it makes perfect sense to go round counting things, and saying that ten automobiles are worth ten times as much as one automobile. Conversely, it is complete gibberish to claim that we are twice as well off becase Newton and Leibnitz both invented calculus, thus giving us twice as much calculus.
Greenspan's emphasis on GDP ducks the challenge of measuring wealth that is tied up in intensive rather than extensive qualities.
The only argument I've seen against an anti-spam law is concerns over freedom of speech. These concerns are easily met. The law need only say:
Port 25 (the Simple Mail Transfer Protocol port) is reserved for human to human communication. It is an offence to cause or permit a robot to post to multiple messages to it.
It is easy to craft an exception for mailing lists that do the three step, subscribe, validate, confirm thing.
First ammendent jurisprudence lets the government regulate speech provided its regulations are narrowly tailored to serve a compelling government interest. What kinds of things count as compelling the US government? Other rights in the Bill of Rights, and laws of physics. For example, if you have two radio transmitters broadcasting on the same frequency each will interfere with the other. This is the justification for Government regulation of the radio spectrum. One way of looking at it is that freedom of speech is played off against freedom of speech. It is hard to oppose government regulation of the radio spectrum on the grounds of freedom of speech if the consequence is that neither speaker will be heard. Equally, it would be very hard to argue in court for the admission of robots to the designated human-to-human channel on freedom of speech grounds when the cause of the regulation was that humans were finding it hard to talk to each other because their channel was being swamped by robot traffic.
Notice also that this proposal avoids violating the first ammendment because it focuses on solving the problem we have, and leaving alone a similar problem that we don't have. Specifically, some-one who wishes to offer you a penis enlarger remains at liberty to get your email address from any source and type in and send an email to you, with their offer. They may repeat this, obtaining somebody elses email address, typing in a fresh message, and sending them a similar offer. They may repeat this as long as their stamina holds out and their fingers will keep tapping away at the keyboard. All that is forbidden is automating the process and then using the humans only port.
Governments spend billions on software. That money comes from taxation. So the revenue raising side of that is already in place and taking money out of your pocket.
The problem lies in the licences that Governments sign up to. The money that the Government raises in taxes is paying for software development, but it goes via companies who own the software. Government could spend some of the money it already raises to spend on software development and use it to develop GPL software.
For example, Government offices need an office suit of some kind, and indeed they have one. Surely the Government office suit should come bundled with Linux for all citizens to use, and perhaps contribute too. We have already paid the taxes!
Here is a prediction of a real change, with serious money making opportunities for businesses that get the timing right.
Damn sure it will. It was youthful exposure to double entry that set the likes of Kenneth Lay and Robert Maxwell on the road to committing serious crimes. For the sake of our children we must block access to dubious sites like this, full of references to double entries.
He is assuming that you can just multiply the word by word probabilities together. This is a standard assumption. If you don't do something like this you get a combinatorial explosion, just like you said. More to the point, if you don't do something like this, your data becomes sparse. In the limit of making no assumptions you are reduced to recognising only the spam you have already seen, you have no capacity for generalisation and all the new spam gets through. No statistical method is any use if it doesn't generalise. Any method that works in practise has some kind of assumption hidden inside to make it go.
One reason I like the Bayesian approach is that it is pretty transparent. When an implementation is making the independence assumption, it is clearly apparent, and if you need to relax the assumption, for example by looking at word pairs, it is clear enough how to go about it. Graham does discuss this towards the end of his article.
Often the main effect of the independence assumption in practise is to exaggerate the confidence with which the classifier classifies things. Since Graham is not using his probabilities as input to subsequent processing he gets away with this,
The big technological road block is getting from the script to the emotions. With an actor, the director can give directions such as "No, no, your character is cereberal. When he gets angry it has to be cold and steely, less is more." Without an actor, the CGI puppet master has to spend hours tinkering to try to get the right effect.
So I see a battle, CGI puppet masters versus actors. A good CGI puppet master can do several characters. As the sole creator of ensemble pieces his work is freed from the constraints of the egos of the actors. The advantages of CGI props and settings can be exploited to the full without the difficulties of integrating with live action, or the expense of real props and location shooting. Against that, CGI puppet masters will slog for hours to get OK renditions of emotional nuances that come to actors, literally, naturally.
Modern armies have been moving away from having large numbers of soldiers. The Iraqi army in 1991 was about 3 million, suffered 200 000 casualties. Raw numbers didn't help.
Pilots in the airforce is another example. Unmanned aircraft are starting to become important to the military.
You seem to be telling me that the "hot intern" at Whitehouse.com isn't staying at 1600 Pennsylvannia Avenue. Have things really changed that much since Clinton?
Let the flying area be 10 meters by 10 meters.
:-)
50mph is a little over 20 meters per second.
Thus the volume of flow required is 2000 cubic meters per second.
Each cubic meter of air has a mass of about 1 kg, so it has a kinetic energy, half m v squared, of 1/2 x 1 x 20 x 20 = 200 Joules. We need 2000 cubic meters per second, so 400kW.
This is a bit more that 500hp, so the idea is not impractical because the fans would take too much power
Put your tuning fork on the bridge of your violin. If the A-string is in tune it will resonate. Bring your little finger close, and as it touches, you not only get a loud buzz, but you can feel the string hitting your finger.
I thought sky divers wore baggy clothes, to reduce their terminal velocity, so that they could spend more time in free fall. How far can you push this? Could you wear a flying squirrel suit, with cloth between you arms and legs, to get your terminal velocity down to say 50mph?
If that worked, could it be the basis for a fairground ride: Enourmous fans blast air upwards at 50mph. Customers wear flying squirrel suits, and get to fly about in the air stream, a bit like being a bird.
Do the Beeb, the Guardian, and the Times give you the story behind the story?
Like me, you have spotted that large government debts create a rentier class, a new "landed" gentry, into whose sevice we have sold our children. And, if you are like me, you didn't get the idea from the Beeb, the Guardian, or the Times; you had to work it out for yourself.
I resented paying the licence fee for crap BBC programs so much that I got rid of my TV, so that I didn't have to pay it any more.
The BBC news reports wind me up. They put the same spin on stories as every-one else. They put positive spin on rising house prices.(The cost of housing is a cost, rising house prices are bad news) They commit dreadful economic howlers, saying that shortages occur inspite of governments capping prices, rather than because governments cap prices, and hoping that trade-protection will save jobs (<sarcasm>much like it did in the 1930's</sarcasm>)
The documentaries have been dumbed down. Now they are so spun out that they are unwatchably tedious. I can remember the James Burke programs from twenty years ago that were a stunning torrent of information and insight.
I think there are two key problems with the BBC.
1)Its the same people. You work for the Guardian for a bit, move to the BBC, move to ITV. Its the same employees who don't know any better.
2)They feel they have to justify the licence fee by keeping the viewing figures up. So if the fashion is for celebrity gossip in the news slot, the BBC competes with ITV for who can get most viewers for their celebrity gossip. They lack the faith in the concept of public service to stick to serious news and argue for the licence fee on the grounds that it is worthwhile to provide serious news even when it is out of fashionable.
I'm T-total, so it annoys me when public policy is discussed as though alcohol were not a drug, and drinking is not recreational drug use. The BBC spins alcohol as not-a-drug, just like the other broadcasters, even though the BBC doesn't carry adverts for alcohol, (or anything).
I never felt that the BBC took advantage of its independence of business to tell me secrets. It is really strange in the UK at the moment, because our two big parties, Conservative, and New Labour, are corporatist/big business soft right parties. 'Old' Labour was left wing, but even back in the eighties, when we had left versus right politics, the licence fee didn't do very much for editorial independence. The BBC depended on the politicians for the licence fee. The Conservative politicians listened to businessmen, the same businessmen who ran the companies that placed the adverts in other media. The Labour politicians listened to the workers who worked in the big unionised companies that placed the adverts in other media. If ITV took an anti-car stance, the advertisers of cars on ITV would bring it back on message. If the BBC took an anti-car stance, either the car workers union leader would talk to the Labour prime minister, or the managing director of Britsh Motor Corporation would talk to the Conservative prime minister. The BBC never developed a distinctive voice.
...I'm the one who ends up paying for stuff like guarding railroad tracks or onsite storage of nuclear waste to avoid moving it by train. So when I'm told that my taxes have to be spent because it is easy to derail a train, I say: let me see the documents. I want to be the judge of whether it is easy enough to be a real threat. It is not right that the government can whip up a scare and then ban the documents that would let me judge for myself whether it is for real or not.
Shells and skeletons fall to the bottom of the sea and turn into limestone.(calcium carbonate)
Peat bogs fail to rot, but thicken and thicken, eventually turning to coal.
Muscles are single acting; they can pull but they cannot push. Joints are double acting; they are powered both ways. The secret is that muscles come in antagonistic pairs, one to flex the joint, and on the other side, one to extend it.
This is like the p-type and n-type transistors in CMOS. Turn on the n-type transistor to pull the output low. Turn on the p-type transistor to pull the output high. Both off at the same time for tri-state. Both on at the same time to short out the power supply and blow up the chip. In the human body, both muscles are off in the relaxed state. If both muscles are slightly on, this pre-stresses the tendons, taking up any slack, and effectively stiffening the joint. This is what you do for exacting fine work, e.g. embroidery or surgery. This is why such work is tiring, even when the external forces you exert are small. For ordinary work, you must coordinate your muscles so that they are non-overlapping.
When you type on a mechanical typewriter, you push hard. I've just been measuring my old Olivetti Lettera 22. The keyboard is open underneath so it is a simple matter to dangle an icecream tub underneath and fill it with water until the rachet clicks to advance the carriage. 1.12 kg. 17mm of key travel. (* 9.81 1.12 0.017) = 186mJ. If you are typing 30 four letter words a minute that is (* (/ 30 60.0) (+ 4 1) 0.186) = 0.464 W. It is not hard physical work.
1.12 kg, say 10N, feels like a lot if you are not used to it, but the significance lies elsewhere. It is way more than the force exerted by the relaxed tone in the muscles that control your finger. So to type a character you turn a flexor full on, and turn it back off again. The typewriter is geared at about 6 to 1, much like a piano, so the hammer is flying pretty fast. Its momentum slams it into the paper, making the impression and the rebound and the little coil spring in the typewriter bring your finger and the key back up. You literally never lift a finger. The springs in the machine lift your fingers for you. You can type with the extensor muscles relaxed all the time. Touch typing on a manual typewriter requires alot of coordination, but it does not require every kind of co-ordination. In particular you do not have to co-ordinate your flexors and extensors to avoid having them both on at the same time, because you never turn on your extensors at all.
A modern mouse is very different. If you just plonk your hand down on top of it you click all three buttons. You have to use you extensors to not click. When I restarted using a computer after a lengthy illness, I rapidly got pains in my arms, from holding my fingers off the mouse buttons all the time. I had to learn to be just tense enough to stop the natural curl of my fingers from clicking the buttons. What happens when I click a mouse button? What is supposed to happen is that the extensor is turned off then the flexor is turned on, then the flexor is turned off, then the extensor is turned back on, so that they don't overlap. I've not done any electro-myography, but I don't believe it is happening like that. Briefly relaxing a muscle that is kept tense most of the time is difficult and time consumming. I bet that the flexor is turned on hard to over come the extensor. How much damage does that do? It probably depends alot on the office environment. If you are generally relaxed and have only just enough tension in your extensors to avoid accidental mouse clicks, I cannot see it doing much harm. If work is fraught, and you tense up to avoid mistakes, beware. The forces exerted when your flexors and extensors are on at the same time add up internally, but cancel externally. You might think that you cannot be stressing your tendons because the switches have a light action and you are not exerting much force, but if that force is the difference between the force exerted by the flexor and the extensor, your tissues might be under a great deal of internal mechanical stress.
I suspect that much the same goes for a modern keyboard. You have to actively lift your fingers off the keys after the stroke. You don't have the option of flexor-only typing. So when work gets hectic and pressured, and your coordination is not 100%, you get flexors working harder to overcome extensors that are not being fully turned off, and lots of internal mechanical stress.
My theory is that these internal stresses are larger than with a clnky old mechanical typewriter and are the cause of RSI.
How can one find out if this theory is true? One way is to get a researcher interested enough in this theory to use electro-myography to find out if both muscles are indeed being turned on at the same time. Another way is to get a keyboard and a mouse with `heavy' long travel keys. This would make sense in a prospective study, in which you equip half a cohort of new users with the clunky mouse and key board, and follow up after five years to see who has RSI and who hasn't. It doesn't make much sense as a treatment. If you have learned to type on a light keyboard with your extensors turned on, the extra force needed to operate a heavier keyboard might be translated by habit into more activation of the extensors as well as more activation of the flexors. I cannot see a heavier keyboard in itself working as therapy, unless the sufferer can learn the flexor only typing technique it permits, and avoid falling back into flexor/extensor overlap habits when work gets hectic.