The station is in geostationary orbit. The counterweight is beyond geostationary, exerting considerable tension on the cable.
If it is beyond geosync, then its orbital period is larger than one day. The "tension" it will exert on the station is thus not radial away from the earth but towards slowing the station down. Slowed down, the station will fall closer to the earth.
The whole thing will simply wind up around the planet.
And at all of/. there's nobody capable of doing the imple math to recognize that.
The counterweight tethered past geostationary will swing backwards in orbit,
More-or-less correct.
then swing forwards again as a pendulum.
No, it won't. There's no restoring force here anywhere. If you are trying to orbit any kind of "counterweight" past geosync, then the orbital period of that counterweight will be longer than one day and thus it'll start wrapping the entire tether around the planet.
To quote someone's sig around here: None of this is rocket science, really:
Write down the total force on a mass element dm in the cable. We
require that the solution is stable, i.e. that none of the mass
elements moves in a coordinate system that rotates with the
earth. Don't forget the pseudo-forces hat are introduced through the
non-inertial system.
a) for undergrads: assume the thing to be a tower (i.e. straight line,
angle with ground 90 degree). Aquire a solution for the total force by
integrating over dm. This will yield a quadratic expression for the
required height of the tower. For the solution to be real, the
expresion under the root must be positive. Verify that this is only
the case if the rotational kinetic energy of the system in question
exceeds the gravitational potential energy of the earth. Verify also,
that this condition is not given for the earth.
b) for graduate students: Write down the continuous Lagrangian for the
system. Assume, for idealisation, infinite tensile strength and
unlimited elasticity of the material. Choose a proper energy Ep as
generalized coordinate. Solve the Euler-lagrange equations for the
problem and show that there is no bound solution for Ep if the above
condition is not fulfilled.
Sorry. A space elevator is not an engineering problem because it is a physical impossibility.
(Extra credit: Compare the two energies and compute how fast the earth would have to spin to allow for such a system, i.e. for there to be a co-rotating satellite with exactly 1 day period that would actually keep the cable under tension.)
From the TFA: Alien visitors coming to Earth 100,000 years hence will find no obvious signs that an advanced civilisation ever lived here.
Umm... They would probably come to that conclusion right now.:-)
Of course they could have a quick look at the Moon instead where there will be a couple LEMs, rovers, and footprints that will still be visible millions of years from now...
Uhm , our 1 big city in the US isn't even in the top 10 in the world.
Depends on how you count it. If you're counting the city proper, the Tokyo isn't in the top ten either. If you count the metro areas (which people usually do -- the whole splotch on the map, usually called "Urban Agglomeration") then New York/Newark is #3 worldwide (after Tokyo and Mexico City).
In all seriousness, though, those who are alive will want to survive any way they can, and those who have already inserted themselves at the top ("old money") will stay there and only occasionally boot the wealthy-but-not-quite-so-wealthy ones to the level of the proles as resources become more and more scarce.
There's an old saying that I once saw attributed to George Orwell -- but I have yet to see any credible source for it. It goes something like this:
All societies have three layers: the high, the middle, and the low. The objective of the high is to maintain the status quo; the objective of the middle is to trade places with the high; the objective of the low is to make all people equal.
So who the hell is actually clicking on all these online advertisements?
Many others have pointed it out (and I'll second them): If I'm in the market for a XYZ-gizmo and I visit a website that I like an they have an ad for someone that sells XYZ-gizmos then I think it is only fair to click that ad and make the purchase through that vendor. The gizom is the same wherever I buy it, but in this way there's a dime or two that'll flow to the website as income. Speaking as someone who's paying for 5GB of traffic per day through advertising, this strikes me as only fair.
But there's more to advertising than the immediate revenue. If Lockheed's market share goes up, Boeing is going to want to raise its visibility. Even if they don't strictly "sell" anything through an ad - if they can get a couple bright students to think of them as their next employer it might already be worth it. It can be desirable to protect a market share even in a not-so-profitable industry simply to prevent the competition from gobbling it up. There's the whole question of branding: "I saw that ad for XYZ-gizmos from the ABS company. I didn't even know they were in that business..." well, now you know. And next year you may remember when it comes to replacing your XYZ-gizmo.
Don't think that every dollar spent on advertising has to show up in next months bank statement as revenue...
Great. Now, how do you measure profits made from advertising, because as I understand it, that is the issue under discussion here.
The same way businesses always have. This is not a new problem.
Actually the comparison between traditional media and the 'net is rather awkward. There is a much more predictable connection between particular markets and particular advertising channels in the traditional media. You spend X dollars on an ad in magazine Y and you can expect Z dollars revenue -- plus/minus seasonal variability and all, but in general you can tell which demographic consumes what media. But an advertising dollar spent at YouTube the week before Google announced they were going to buy them would have yielded a completely different result from a dollar spent a week later. You get completely different revenue figures the day before a site is slashdotted from the day after. Markets, demographics, communications channels are much more volatile and ephemeral on the internet than anywhere else. The websites with the most loyal followership do not come close to comparing with even mediocre magazines or even radio shows.
Run an ad on some radio show for a month, then not for a month. Repeat. You're gonna see what's up -- there's going to be a pattern that correlates with your spending. But run an ad on a website for a month and the result is a lot less predictable. Maybe you'll see a correlation with revenue. Maybe not.
Amongst the things that make the net so much more unpredictable are the much higher influence people have over what they see (are you really going to tune to another station to escape a 30 second ad? I know I don't - but I routinely flip to another tab to see what's going on there while the other tab is cutting through ad-crap) and the fact that they have the ability to respond - on the net you cannot just see an ad, you can talk about the ad, you can direct others to it, you can make fun of it in some forum somewhere.
I hear a radio ad and I buy the product and the cycle ends there. But I see an ad on the 'net and I post to my favorite forum "hey guys check this out, XYZ computers is having a sale on the ABC gizmo...". This means there's a much wider range of effects that any one ad can have (including, I might add, "Hey guys, have you seen that racist XYZ ad? That's it! I'm gonna boycott their crap until...").
I can tell your from experience (this is one of the more popular demonstrations in the power engineering lab) that cheap watt-meters can be terribly wrong with loads that are not simple resistors.
In other words: you are NOT actually telling us anything "from experience" at all. You are telling us about a demo you saw in a class. That's a huge difference.
Can you actually NAME any "cheap wattmeter" anywhere out there in real consumer land that does NOT properly factor load angles into its results? Any one?
When I read "cheap wattmeter" I think something like this which costs under $20 (certainly qualifies as cheap in my book) and outputs W and VA independently (and also the power factor, i.e. cos(phi)).
(It also integrates over about a second or so between outputs, which makes your comment about needing fast ADCs to measure switching power supplies completely moot.)
Is there truly any equipment out there that is actually marketed to actual customers as measuring your power consumption that does not do any of these?
For if there isn't, then you've merely told us that your engineering professor managed to rig up something invalid -- which isn't exactly... err... "insghtful".
Those LED's cost next to nothing, I would guess on the order of pennies per decade.
This is only marginally on topic, but: Why guess? Is it really so hard to multiply a couple numbers? Are you really saving yourself effort buy operating on the basis of ignorant guesses when you could inform yourself within a few seconds by taking a run-of-the mill number for an LED, say 1.6V 20mA makes 32mW, makes 32mWh per hour. Times 24 is 768mWh per day, comes to somewhere under 300Wh per year. Around here we pay ~10c/kWh so you're talking dollars per decade, not pennies per decade. Per LED.
This took me more time to type than it would've taken you to compute. I'm kinda annoyed by this out in the Real World[tm] but on/. I somehow kinda expect people to prefer a simple multiplication over making potentially stupid statements...
Here's a cute little first-year CS problem: show that with the current calendar the 13th of a month has a higher probability of falling on a friday than any other day of the week.
Maybe I'm missing some subtlety in the OP somewhere, but if GPUs weren't better at what they're doing than CPUs, there wouldn't be a point in having a GPU in the first place.
...and if you have a problem that can be expressed in terms of the problem space the GPU is designed to handle, then that problem is going to run faster on the gpu than on the CPU.
I'm all for abstracting the API's to make it easier for developers to write apps that'll work well in more places.
Let's say I agree with this.
I just don't like the idea of combining the UI into one big project in a way that might make it harder to get diversity and choice for users.
Well - that's the problem, no? If it isn't "one big project" then it's going to be "a dozen small projects". And whenever someone doesn't like something, they'll for their own version and their own way of doing things because exactly there isn't one big project that would be hampered by it. The philosophy is "if you don't like my stuff, use something else (or write your own)" and then people do that and use something else and the whole operation gets balkanized into a hundred fractions.
I'm surprised that some chip-manufacturer (I'm thinking VIA, but it could be ATI or nVidia as well) hasn't come up with a hardware-accelerator for desktops yet. In the end, MacOS and Windows and KDE and Gnome and whatever the hell else is out there all "look and feel" exactly the same: they have buttons and windows and menus and if ever someone invents something truly different (say tabs), all the others immediately adopt it as well. The rest is really cosmetics. I'm waiting for the lowball-mobo that carries a specialized graphics chip designed to allow you to run WinVista but without much 3D-fanfare or any such thing. Once we have that, the hardware interface to that chip (i.e. the Vista GUI API) will in effect define how to talk to a GUI from then on -- everybody will want that chip and all the pretty linux desktops will devolve into skins on top of that API. Because in the end, you cannot beat dedicated hardware for sheer speed and efficiency of operation.
Linux had a chance at developing something of an "generalized GUI API" layer (in the sense of openGL) that goes between your linux and your GUI - soemthing that lives on the level of X. Something that encapsulates the user interaction sufficiently well away from the OS as to be abstractable into hardware. We did that with GL. Before that we did it with sound. Before that we did it with network interfaces. Figure out what it is that you do all the time and make a chip that does it so you don't load the CPU with that task. I fear the chance has been blown as the next opportunity to make money by doing this abstraction will be the production of cheap MoBos with onboard-video that is Aero-capable.
Slashdot Mirror
However, unlike warp drive, it's at least not violating the laws of physics.
Yes, it is very much violating the laws of physics.
Of course I cannot force you to open a mechanics book. That would be up to your motivation.
The station is in geostationary orbit. The counterweight is beyond geostationary, exerting considerable tension on the cable.
If it is beyond geosync, then its orbital period is larger than one day. The "tension" it will exert on the station is thus not radial away from the earth but towards slowing the station down. Slowed down, the station will fall closer to the earth.
The whole thing will simply wind up around the planet.
And at all of /. there's nobody capable of doing the imple math to recognize that.
Pitiful, really.
Why was that modded "flamebait"?? It's the first realistic comment in the entirel thread!
The counterweight tethered past geostationary will swing backwards in orbit,
More-or-less correct.
then swing forwards again as a pendulum.
No, it won't. There's no restoring force here anywhere. If you are trying to orbit any kind of "counterweight" past geosync, then the orbital period of that counterweight will be longer than one day and thus it'll start wrapping the entire tether around the planet.
To quote someone's sig around here: None of this is rocket science, really:
Write down the total force on a mass element dm in the cable. We require that the solution is stable, i.e. that none of the mass elements moves in a coordinate system that rotates with the earth. Don't forget the pseudo-forces hat are introduced through the non-inertial system.
a) for undergrads: assume the thing to be a tower (i.e. straight line, angle with ground 90 degree). Aquire a solution for the total force by integrating over dm. This will yield a quadratic expression for the required height of the tower. For the solution to be real, the expresion under the root must be positive. Verify that this is only the case if the rotational kinetic energy of the system in question exceeds the gravitational potential energy of the earth. Verify also, that this condition is not given for the earth.
b) for graduate students: Write down the continuous Lagrangian for the system. Assume, for idealisation, infinite tensile strength and unlimited elasticity of the material. Choose a proper energy Ep as generalized coordinate. Solve the Euler-lagrange equations for the problem and show that there is no bound solution for Ep if the above condition is not fulfilled.
Sorry. A space elevator is not an engineering problem because it is a physical impossibility.
(Extra credit: Compare the two energies and compute how fast the earth would have to spin to allow for such a system, i.e. for there to be a co-rotating satellite with exactly 1 day period that would actually keep the cable under tension.)
Because the problem still remains--research causes cancer in rats.
Actually, apparently not all research...
...and I promise I'm going to learn how to use 'blockquote' soon ...
Uhhh... If Alaska isn't "the north", then I don't know what is. How can Alaska have "a higher birth rate than the north"? That doesn't make sense.
Odd Even, good name but i can see the problem for english people
I'll give you even odds they woldn't even notice.
Neat map.
I never figured that Montana and Alaska count as "the south" these days.
Live and learn...
Uhm , our 1 big city in the US isn't even in the top 10 in the world.
Depends on how you count it. If you're counting the city proper, the Tokyo isn't in the top ten either. If you count the metro areas (which people usually do -- the whole splotch on the map, usually called "Urban Agglomeration") then New York/Newark is #3 worldwide (after Tokyo and Mexico City).
(Source: geohive.com)
In all seriousness, though, those who are alive will want to survive any way they can, and those who have already inserted themselves at the top ("old money") will stay there and only occasionally boot the wealthy-but-not-quite-so-wealthy ones to the level of the proles as resources become more and more scarce.
There's an old saying that I once saw attributed to George Orwell -- but I have yet to see any credible source for it. It goes something like this:
All societies have three layers: the high, the middle, and the low. The objective of the high is to maintain the status quo; the objective of the middle is to trade places with the high; the objective of the low is to make all people equal.
... slashdotted...
So who the hell is actually clicking on all these online advertisements?
Many others have pointed it out (and I'll second them): If I'm in the market for a XYZ-gizmo and I visit a website that I like an they have an ad for someone that sells XYZ-gizmos then I think it is only fair to click that ad and make the purchase through that vendor. The gizom is the same wherever I buy it, but in this way there's a dime or two that'll flow to the website as income. Speaking as someone who's paying for 5GB of traffic per day through advertising, this strikes me as only fair.
But there's more to advertising than the immediate revenue. If Lockheed's market share goes up, Boeing is going to want to raise its visibility. Even if they don't strictly "sell" anything through an ad - if they can get a couple bright students to think of them as their next employer it might already be worth it. It can be desirable to protect a market share even in a not-so-profitable industry simply to prevent the competition from gobbling it up. There's the whole question of branding: "I saw that ad for XYZ-gizmos from the ABS company. I didn't even know they were in that business..." well, now you know. And next year you may remember when it comes to replacing your XYZ-gizmo.
Don't think that every dollar spent on advertising has to show up in next months bank statement as revenue...
Actually the comparison between traditional media and the 'net is rather awkward. There is a much more predictable connection between particular markets and particular advertising channels in the traditional media. You spend X dollars on an ad in magazine Y and you can expect Z dollars revenue -- plus/minus seasonal variability and all, but in general you can tell which demographic consumes what media. But an advertising dollar spent at YouTube the week before Google announced they were going to buy them would have yielded a completely different result from a dollar spent a week later. You get completely different revenue figures the day before a site is slashdotted from the day after. Markets, demographics, communications channels are much more volatile and ephemeral on the internet than anywhere else. The websites with the most loyal followership do not come close to comparing with even mediocre magazines or even radio shows.
Run an ad on some radio show for a month, then not for a month. Repeat. You're gonna see what's up -- there's going to be a pattern that correlates with your spending. But run an ad on a website for a month and the result is a lot less predictable. Maybe you'll see a correlation with revenue. Maybe not.
Amongst the things that make the net so much more unpredictable are the much higher influence people have over what they see (are you really going to tune to another station to escape a 30 second ad? I know I don't - but I routinely flip to another tab to see what's going on there while the other tab is cutting through ad-crap) and the fact that they have the ability to respond - on the net you cannot just see an ad, you can talk about the ad, you can direct others to it, you can make fun of it in some forum somewhere.
I hear a radio ad and I buy the product and the cycle ends there. But I see an ad on the 'net and I post to my favorite forum "hey guys check this out, XYZ computers is having a sale on the ABC gizmo...". This means there's a much wider range of effects that any one ad can have (including, I might add, "Hey guys, have you seen that racist XYZ ad? That's it! I'm gonna boycott their crap until...").
I can tell your from experience (this is one of the more popular demonstrations in the power engineering lab) that cheap watt-meters can be terribly wrong with loads that are not simple resistors.
In other words: you are NOT actually telling us anything "from experience" at all. You are telling us about a demo you saw in a class. That's a huge difference.
Can you actually NAME any "cheap wattmeter" anywhere out there in real consumer land that does NOT properly factor load angles into its results? Any one?
When I read "cheap wattmeter" I think something like this which costs under $20 (certainly qualifies as cheap in my book) and outputs W and VA independently (and also the power factor, i.e. cos(phi)).
(It also integrates over about a second or so between outputs, which makes your comment about needing fast ADCs to measure switching power supplies completely moot.)
Is there truly any equipment out there that is actually marketed to actual customers as measuring your power consumption that does not do any of these?
For if there isn't, then you've merely told us that your engineering professor managed to rig up something invalid -- which isn't exactly ... err ... "insghtful".
Those LED's cost next to nothing, I would guess on the order of pennies per decade.
This is only marginally on topic, but: Why guess? Is it really so hard to multiply a couple numbers? Are you really saving yourself effort buy operating on the basis of ignorant guesses when you could inform yourself within a few seconds by taking a run-of-the mill number for an LED, say 1.6V 20mA makes 32mW, makes 32mWh per hour. Times 24 is 768mWh per day, comes to somewhere under 300Wh per year. Around here we pay ~10c/kWh so you're talking dollars per decade, not pennies per decade. Per LED.
This took me more time to type than it would've taken you to compute. I'm kinda annoyed by this out in the Real World[tm] but on /. I somehow kinda expect people to prefer a simple multiplication over making potentially stupid statements...
I have recently switched to a steam powered laptop.
Oh, was that you? : http://www.engadget.com/2006/10/13/steampunk-lapto p-comes-complete-with-morse-key/
The start of today is JD2454021.5 and at noon is JD2454022.0
Which, if my rusty astronomy serves, would make it MJD54022, right?
5+4+2+2=13 You can't make this shit up...
Vaguely more interesting and almost veering ON topic for slashdot, in binary notation todays date would be
1010 1101 11111010110
...which just so happens to have thirteen "1"s in it.
Here's a cute little first-year CS problem: show that with the current calendar the 13th of a month has a higher probability of falling on a friday than any other day of the week.
And for the rest of us, the overwhelming majority of rational folks, [...]
Wow.
What country do you live in?
Can I come?
Maybe I'm missing some subtlety in the OP somewhere, but if GPUs weren't better at what they're doing than CPUs, there wouldn't be a point in having a GPU in the first place.
...and if you have a problem that can be expressed in terms of the problem space the GPU is designed to handle, then that problem is going to run faster on the gpu than on the CPU.
ASCII silly question, get a silly ANSI.
I'm all for abstracting the API's to make it easier for developers to write apps that'll work well in more places.
Let's say I agree with this.
I just don't like the idea of combining the UI into one big project in a way that might make it harder to get diversity and choice for users.
Well - that's the problem, no? If it isn't "one big project" then it's going to be "a dozen small projects". And whenever someone doesn't like something, they'll for their own version and their own way of doing things because exactly there isn't one big project that would be hampered by it. The philosophy is "if you don't like my stuff, use something else (or write your own)" and then people do that and use something else and the whole operation gets balkanized into a hundred fractions.
I'm surprised that some chip-manufacturer (I'm thinking VIA, but it could be ATI or nVidia as well) hasn't come up with a hardware-accelerator for desktops yet. In the end, MacOS and Windows and KDE and Gnome and whatever the hell else is out there all "look and feel" exactly the same: they have buttons and windows and menus and if ever someone invents something truly different (say tabs), all the others immediately adopt it as well. The rest is really cosmetics. I'm waiting for the lowball-mobo that carries a specialized graphics chip designed to allow you to run WinVista but without much 3D-fanfare or any such thing. Once we have that, the hardware interface to that chip (i.e. the Vista GUI API) will in effect define how to talk to a GUI from then on -- everybody will want that chip and all the pretty linux desktops will devolve into skins on top of that API. Because in the end, you cannot beat dedicated hardware for sheer speed and efficiency of operation.
Linux had a chance at developing something of an "generalized GUI API" layer (in the sense of openGL) that goes between your linux and your GUI - soemthing that lives on the level of X. Something that encapsulates the user interaction sufficiently well away from the OS as to be abstractable into hardware. We did that with GL. Before that we did it with sound. Before that we did it with network interfaces. Figure out what it is that you do all the time and make a chip that does it so you don't load the CPU with that task. I fear the chance has been blown as the next opportunity to make money by doing this abstraction will be the production of cheap MoBos with onboard-video that is Aero-capable.
We'll see...