1. Getting all the ratios correct to build a plane capable of lifting the payload and the batteries required for overnight endurance is part of the challenge, but I believe that "good enough" solutions are already with us or are right around the corner.
2. Of course satellites need to last for 10 years - there's no way to service them at all. Their design, build and deployment costs are astronomical (heh) because of that. The equivalent costs of aeronautical platforms will be so much lower by comparison that the maintenance costs will hardly matter. If they crash, they crash - same with aircraft today, but aircraft - particularly commercial aircraft - have a record of safety I'm not sure you're fully appreciating. I would suggest that if these aircraft were going to be autonomously or remotely piloted that they carry automatic parachute systems as a fail-safe. You also, in general, will probably want to be orbiting these things above farmland or water rather than skyscrapers.
3. You need to add up *all* of the operational costs. That fleet of satellites is phenomenally expensive by comparison.
So why hasn't it been done so far?
I posit that it's a historical accident. The first need we had was for trans-oceanic telecommunications links and continental multi-casting (that is, satellite distribution of TV programming). For those uses, geostationary satellites are superior because they're higher up and thus have a larger horizon. That, and the technology for a pilotless solar powered airplane didn't exist. If, instead, our need was based around, say, metro- or regional-area wireless data transmission...
Among the advantage to stratospheric platforms over geostationary orbit satellites:
1. Antenna aiming isolation. The same downlink frequency can be shared by many more users since the downlink antennas could be directional enough (like satellite dishes are) to reject other sources. This is true of geostationary orbit satellites, but the locations can only really be varied along an arc over the equator. Aeronautical platforms can be moved in two dimensions (obviously they really are in 3 dimensions, but the antenna aiming will still simply be azimuth and elevation).
2. Better geographical isolation. Because they're lower, their horizon area will be much, much smaller. This means their service areas can be more easily limited. This can be done with antenna geometry for geostationary orbit satellites, but in general they can still see almost an entire hemisphere of the earth, so they can still raise the noise floor in otherwise out-of-service areas.
3. Less delay. Geostationary orbits are far enough away that they introduce a delay that is unacceptable for most duplex applications (such as telephone or Internet traffic). Aeronautical platforms would be close enough to the ground that that delay would likely not be a factor.
4. More servicable. The cost of launching geostationary payloads means that the payloads must be designed for a long service life, which raises the cost of the payload significantly, as well as the risks. Aeronautical platforms simply have to come down for a landing, be serviced, and then take off and fly back to station.
5. Less chance of solar outages. Twice a year, there are a few days in a row where geostationary satellites transit the sun. Often, their signal gets drowned out when that happens. Aeronautical platforms stationed outside of the tropics, the problem can be avoided pretty easily (simply place the station to the north of the service area for the Northern hemisphere or to the south of the service area for the Southern hemisphere).
It's a reference to the movie "For A Few Dollars More." The Bank of El Paso decided that stealth was a better option and disguised their safe with a wood cabinet and left the giant iron bank vault empty. Of course, the villian learned the secret and stole the wooden cabinet instead.
if we get a bank vault, the thieves will just plan and figure out a way to get into the vault.
Cue the Ennio Morricone soundtrack...
"I would like to relate a nice little parable. Once upon a time, there was a carpenter. You don't think a carpenter can make money, eh? Well, you're wrong. This one did well, because he was a builder of safes...."
We've seen this before. The industry is constantly cycling between specialized co-processors (what I loosely call asymetric multiprocessing) to increase performance and increasingly powerful central processors and dumb peripherals to decrease cost and bus latencies. What's old is new again.
You should see how thin some dishes on real satellites are.
Just to spell it out for everyone... Of course those dishes can be thin - there's no wind in space to deform them. Down here dishes have to stand up to wind load.
By "many," I believe you probably mean "one." The one in particular was the Stanford Prison Experiment, run by Philip Zimbardo back in 1971. It proved your point, but had such a profound negative impact on the participants that it had to be terminated early, and ethical considerations would prevent it from being repeated.
Your VoIP box IS able to generate a clocking good enough to make the modem able to talk to IT. If it then does a local softmodem function it could recover the data stream, and the application could pack THAT up in some protocol other than VoIP for transport across the internet to a similar function on the other end. Then the timing issues can be sidestepped, with the TIFF stream from the FAX or the PPP (or whatever) on the modem slipped coherently by appropriate idle insertion/deletion or the equivalent. The machine at the far end won't see the identical bit stream, but it will see a stream that is semantically equivalent. Then your modem or FAX machine would work just fine.
This, ironically, reminds me of old Telebit modems in the past. Sort of. Bear with me.
Telebit had some of the first high speed modems I ever recall seeing (19.2 kbps back when I had a 2400 bps modem on my computer). They managed this by letting the modems talk to each other with half-duplex. This had a fairly obvious negative impact on latency, but if you were running a simple one-way transmission protocol, the bulk transfer rates were impressive (for the time).
The problem was that the file transfer protocols in use at the time - XMODEM, Kermit and UUCP-g, didn't play that. So the Telebits had a special register in them which would make them perform protocol endpoint emulation. If you were going to use UUCP-g over your dialup connection, then when the transfer actually started, the Telebit on your end would ack all of your packets for you, send just the data across the link, and the remote modem would eat all of the acks coming from the remote machine.
When I signed up for DSL, they didn't offer it naked, so I had to get a measured-rate POTS line to go along with it. I used it for fax reception and got a Vonage box for voice service.
When I got an alarm, it was a happy coincidence - I hooked the RJ31X jack for the alarm up to that POTS line along with the fax modem. So our alarm system isn't dependent on the Internet or Vonage being up.
If it weren't for that, I probably would have opted for the GSM module available for our alarm. But the monthly service cost for that probably equals or exceeds the cost of the POTS line running along side the DSL, and that can at least serve an extra purpose.
Everyone talks about the big problem being that you have to give the key to the fellow who's going to watch the movie, but even that understates the difficulties facing DRM schemes.
Recently, I put up a GeoCache puzzle cache. The idea was that folks would have to figure out the puzzle to find out the GPS coordinates of the cache. I was very clever and devious. I was humbled when the thing was found within 6 hours of publication.
How was it done?
To make a long story short, it was a "known plaintext attack." Since I am required to publicize a pair of coordinates somewhere within a couple miles of the cache (to make the geocache site's search engine work correctly - so that folks from New York won't solve the puzzle and get screwed when the cache is 2000 miles away), this lets attackers look for solutions that result in numbers "near" the posted coordinates.
This is what makes movie DRM untenable. Since the format of the disks is publicly known (to insure that UNencrypted disks operate correctly), attackers know that they can discard solutions after decrypting very little of the ciphertext (probably just one byte).
With sufficiently large keys, even that becomes a huge problem, but the fact that the format of the plaintext is known is still a huge advantage for the attackers.
Um, no. Try $5000+ a for HIGH END PC. Do your homework before you make such ridiculous statements. Not only are you right, but even in this space Macs are competitive.
$5045 buys you a Mac Pro with 4 3 GHz Xeon cores (2x dual core), 4 GB of RAM, a 750 GB SATA hard disk, an ATI Radeon X1900 512M (2x dual-link DVI) video card, a 16x DVD+/-RW (DL) optical drive.
Just fooling around on Dell's site for a few minutes I wound up with a price for an XPS 710 H2C with a quad core 3.2 GHz ("Factory overclocked." I am not kidding.), 4 GB RAM, 500GB RAID 1 (2x 500 GB drives), a similar optical drive (though they also throw in an extra DVD-ROM optical drive) and an NVidia GeForce 8800 GTX with 768M running Windows XP Pro for $5224. It's pretty close. Replace the Mac Pro's single 750 GB drive with 2x 500GB and the price is $5244.
The least expensive mac has a 1.66 GHz Core Duo processor, 512M of RAM, a 60GB hard disk and a DVD / CDRW drive and Gig-E. It is $599.
Your comparison machine is used, not as fast, has a smaller hard drive and (presumably) no optical drive. Try again. Remember, the object of the game is to find the same configuration for less money.
That old canard is getting very tired. When compared to equivalent hardware and bundles, Macs are very competitive.
What IS true is that the least expensive mac is more expensive than the least expensive PC. But, of course, the least expensive PC is invariably a cheap piece of crap one step up from a calculator.
Ha ha, but "hit-to-kill" is in contrast, for example, to the current generation of TOW missiles, which try to fly just over a tank and explode directly above it. That is, they intentionally try to "miss" the target so that they can deploy blast energy at a more vulnerable location of the target.
For most FOSS licences, however, there is no consideration, no requirement to agree to terms (eg. as required by UCITA), and hence no contract.
Sure there is. The consideration is permission to use and (possibly) redistribute the software. I agree that the requirement to agree to the terms is an implied one for most licenses. The GPL makes it explicit ("nothing else besides this license permits you to...."), to its credit.
In fact, the BSD license has only been tested once (so far as I am aware), and the case was settled, so there is probably no case law to go on, but at least to some extent, the usual practices in an industry are permitted to be taken into consideration by the judge.
My analysis of the paper is that the author confuses what is permitted with what is required.
Let's take a look:
Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
So long as you include the disclaimer as required, and you don't use the author's name in vain, you can do as you like.
There is no part of the license that says that you cannot distribute modified forms under more restrictive licensing, provided that you also perform the acts required by the BSD license and require the same of any sub-licensees that distribute. There is no part of the license that says that you cannot distirbute UNmodified forms under more restrictive licensing, but presumably anyone receiving a copy from you under more restrictive licensing could figure out that you obtained it from a source that merely required adherence to the BSD license, throw away the copy you provided and get their own.
In general, acts that are not specified as prohibited in a contract (in this case the license is a contractural term. You are agreeing to abide by the license in return for being provided value in the form of the code covered by the license) are permitted (modulo exceptions that aren't worthy of mention here). Since sublicensing is not mentioned, it is permitted - provided the original conditions are always met by anyone redistributing and/or using the code.
Slashdot Mirror
1. Getting all the ratios correct to build a plane capable of lifting the payload and the batteries required for overnight endurance is part of the challenge, but I believe that "good enough" solutions are already with us or are right around the corner.
2. Of course satellites need to last for 10 years - there's no way to service them at all. Their design, build and deployment costs are astronomical (heh) because of that. The equivalent costs of aeronautical platforms will be so much lower by comparison that the maintenance costs will hardly matter. If they crash, they crash - same with aircraft today, but aircraft - particularly commercial aircraft - have a record of safety I'm not sure you're fully appreciating. I would suggest that if these aircraft were going to be autonomously or remotely piloted that they carry automatic parachute systems as a fail-safe. You also, in general, will probably want to be orbiting these things above farmland or water rather than skyscrapers.
3. You need to add up *all* of the operational costs. That fleet of satellites is phenomenally expensive by comparison.
So why hasn't it been done so far?
I posit that it's a historical accident. The first need we had was for trans-oceanic telecommunications links and continental multi-casting (that is, satellite distribution of TV programming). For those uses, geostationary satellites are superior because they're higher up and thus have a larger horizon. That, and the technology for a pilotless solar powered airplane didn't exist. If, instead, our need was based around, say, metro- or regional-area wireless data transmission...
Among the advantage to stratospheric platforms over geostationary orbit satellites:
1. Antenna aiming isolation. The same downlink frequency can be shared by many more users since the downlink antennas could be directional enough (like satellite dishes are) to reject other sources. This is true of geostationary orbit satellites, but the locations can only really be varied along an arc over the equator. Aeronautical platforms can be moved in two dimensions (obviously they really are in 3 dimensions, but the antenna aiming will still simply be azimuth and elevation).
2. Better geographical isolation. Because they're lower, their horizon area will be much, much smaller. This means their service areas can be more easily limited. This can be done with antenna geometry for geostationary orbit satellites, but in general they can still see almost an entire hemisphere of the earth, so they can still raise the noise floor in otherwise out-of-service areas.
3. Less delay. Geostationary orbits are far enough away that they introduce a delay that is unacceptable for most duplex applications (such as telephone or Internet traffic). Aeronautical platforms would be close enough to the ground that that delay would likely not be a factor.
4. More servicable. The cost of launching geostationary payloads means that the payloads must be designed for a long service life, which raises the cost of the payload significantly, as well as the risks. Aeronautical platforms simply have to come down for a landing, be serviced, and then take off and fly back to station.
5. Less chance of solar outages. Twice a year, there are a few days in a row where geostationary satellites transit the sun. Often, their signal gets drowned out when that happens. Aeronautical platforms stationed outside of the tropics, the problem can be avoided pretty easily (simply place the station to the north of the service area for the Northern hemisphere or to the south of the service area for the Southern hemisphere).
film at 11.
It's a reference to the movie "For A Few Dollars More." The Bank of El Paso decided that stealth was a better option and disguised their safe with a wood cabinet and left the giant iron bank vault empty. Of course, the villian learned the secret and stole the wooden cabinet instead.
Cue the Ennio Morricone soundtrack...
"I would like to relate a nice little parable. Once upon a time, there was a carpenter. You don't think a carpenter can make money, eh? Well, you're wrong. This one did well, because he was a builder of safes...."
We've seen this before. The industry is constantly cycling between specialized co-processors (what I loosely call asymetric multiprocessing) to increase performance and increasingly powerful central processors and dumb peripherals to decrease cost and bus latencies. What's old is new again.
Mostly tweaking people who have no sense of humor.
Not it.
Since we're talking about Windows machines, I can tell you for certain which comes first.
Just to spell it out for everyone... Of course those dishes can be thin - there's no wind in space to deform them. Down here dishes have to stand up to wind load.
The article makes the same mistake, so I don't blame the editors.
The dish in question is part of a terrestrial point-to-point link. It's not a satellite dish, since satellites aren't involved.
That was the Milgram experiment from 1961, which didn't have a thing to do with prison.
By "many," I believe you probably mean "one." The one in particular was the Stanford Prison Experiment, run by Philip Zimbardo back in 1971. It proved your point, but had such a profound negative impact on the participants that it had to be terminated early, and ethical considerations would prevent it from being repeated.
This, ironically, reminds me of old Telebit modems in the past. Sort of. Bear with me.
Telebit had some of the first high speed modems I ever recall seeing (19.2 kbps back when I had a 2400 bps modem on my computer). They managed this by letting the modems talk to each other with half-duplex. This had a fairly obvious negative impact on latency, but if you were running a simple one-way transmission protocol, the bulk transfer rates were impressive (for the time).
The problem was that the file transfer protocols in use at the time - XMODEM, Kermit and UUCP-g, didn't play that. So the Telebits had a special register in them which would make them perform protocol endpoint emulation. If you were going to use UUCP-g over your dialup connection, then when the transfer actually started, the Telebit on your end would ack all of your packets for you, send just the data across the link, and the remote modem would eat all of the acks coming from the remote machine.
When I signed up for DSL, they didn't offer it naked, so I had to get a measured-rate POTS line to go along with it. I used it for fax reception and got a Vonage box for voice service.
When I got an alarm, it was a happy coincidence - I hooked the RJ31X jack for the alarm up to that POTS line along with the fax modem. So our alarm system isn't dependent on the Internet or Vonage being up.
If it weren't for that, I probably would have opted for the GSM module available for our alarm. But the monthly service cost for that probably equals or exceeds the cost of the POTS line running along side the DSL, and that can at least serve an extra purpose.
Everyone talks about the big problem being that you have to give the key to the fellow who's going to watch the movie, but even that understates the difficulties facing DRM schemes.
Recently, I put up a GeoCache puzzle cache. The idea was that folks would have to figure out the puzzle to find out the GPS coordinates of the cache. I was very clever and devious. I was humbled when the thing was found within 6 hours of publication.
How was it done?
To make a long story short, it was a "known plaintext attack." Since I am required to publicize a pair of coordinates somewhere within a couple miles of the cache (to make the geocache site's search engine work correctly - so that folks from New York won't solve the puzzle and get screwed when the cache is 2000 miles away), this lets attackers look for solutions that result in numbers "near" the posted coordinates.
This is what makes movie DRM untenable. Since the format of the disks is publicly known (to insure that UNencrypted disks operate correctly), attackers know that they can discard solutions after decrypting very little of the ciphertext (probably just one byte).
With sufficiently large keys, even that becomes a huge problem, but the fact that the format of the plaintext is known is still a huge advantage for the attackers.
Um, no. Try $5000+ a for HIGH END PC. Do your homework before you make such ridiculous statements. Not only are you right, but even in this space Macs are competitive.
$5045 buys you a Mac Pro with 4 3 GHz Xeon cores (2x dual core), 4 GB of RAM, a 750 GB SATA hard disk, an ATI Radeon X1900 512M (2x dual-link DVI) video card, a 16x DVD+/-RW (DL) optical drive.
Just fooling around on Dell's site for a few minutes I wound up with a price for an XPS 710 H2C with a quad core 3.2 GHz ("Factory overclocked." I am not kidding.), 4 GB RAM, 500GB RAID 1 (2x 500 GB drives), a similar optical drive (though they also throw in an extra DVD-ROM optical drive) and an NVidia GeForce 8800 GTX with 768M running Windows XP Pro for $5224. It's pretty close. Replace the Mac Pro's single 750 GB drive with 2x 500GB and the price is $5244.
Nice straw-man.
The least expensive mac has a 1.66 GHz Core Duo processor, 512M of RAM, a 60GB hard disk and a DVD / CDRW drive and Gig-E. It is $599.
Your comparison machine is used, not as fast, has a smaller hard drive and (presumably) no optical drive. Try again. Remember, the object of the game is to find the same configuration for less money.
That old canard is getting very tired. When compared to equivalent hardware and bundles, Macs are very competitive.
What IS true is that the least expensive mac is more expensive than the least expensive PC. But, of course, the least expensive PC is invariably a cheap piece of crap one step up from a calculator.
Mind you, that's after all of the denial, anger, bargaining and depression.
We already are.
Ha ha, but "hit-to-kill" is in contrast, for example, to the current generation of TOW missiles, which try to fly just over a tank and explode directly above it. That is, they intentionally try to "miss" the target so that they can deploy blast energy at a more vulnerable location of the target.
Sure there is. The consideration is permission to use and (possibly) redistribute the software. I agree that the requirement to agree to the terms is an implied one for most licenses. The GPL makes it explicit ("nothing else besides this license permits you to...."), to its credit.
In fact, the BSD license has only been tested once (so far as I am aware), and the case was settled, so there is probably no case law to go on, but at least to some extent, the usual practices in an industry are permitted to be taken into consideration by the judge.
Go read it again.
I didn't say a license is a contract. A license is part of a contract. In return for value, you agree to abide by the license.
So, if a licence doesn't give you permission to do something (that would be prohibited by copyright law), then you can't do it
Copyright law does not outlaw adding terms when redistributing copies, providing the redistribution is permitted by the copyright holder.
Let's take a look:
So long as you include the disclaimer as required, and you don't use the author's name in vain, you can do as you like.There is no part of the license that says that you cannot distribute modified forms under more restrictive licensing, provided that you also perform the acts required by the BSD license and require the same of any sub-licensees that distribute. There is no part of the license that says that you cannot distirbute UNmodified forms under more restrictive licensing, but presumably anyone receiving a copy from you under more restrictive licensing could figure out that you obtained it from a source that merely required adherence to the BSD license, throw away the copy you provided and get their own.
In general, acts that are not specified as prohibited in a contract (in this case the license is a contractural term. You are agreeing to abide by the license in return for being provided value in the form of the code covered by the license) are permitted (modulo exceptions that aren't worthy of mention here). Since sublicensing is not mentioned, it is permitted - provided the original conditions are always met by anyone redistributing and/or using the code.