The thing to do is to hold the ISP accountable if they don't hold the user accountable.
For example - I had this host that kept sending me half-megabyte virus executables via mail. I identified the ISP as Netvision in Israel. I tried to contact them repeately. They did nothing to stop this - they did not contact the user, they did not disconnect the user, they did not block the user's ability to send mail, NOTHING.
In cases like this, then HELL YES I say hold the ISP accoutable - they have failed to hold the user accountable.
If I start making prank calls from my phone, the phone company will kill my line if they get called about it. ISPs should be no exception.
That last bit about the radio station is definitely in the catagory of "Not Smart" - a conviction by the FCC for such activities might be just a bit of a rough spot in a job interview for a radio tech.
But yes, it's a bit scary to think about how many millions of dollars of test gear spends most of it's time being a radio reciever - I'm sure that this happens at R&S, Tek, and Agilent too.
How'd you pull off the TV trick? I'm assuming you used the HP to get the video signal, one 500 to get the audio, and the Tek to display, but how'd you get the raster?
I was wondering about this as well - the number of observations we are making of planets in other solar systems is already astonishingly large given the angular distance these objects subtend as viewed from Earth (true, many of the observations are indirect and thus don't directly depend upon the angular distance).
I wonder if the breakthroughs in bypassing the diffraction limit will allow for direct imaging of larger bodies (Jovian worlds at Jovian orbital distances).
Hello. Let me welcome you to Slashdot - I can see you are new here.
You see, you fatally misunderstand the meaning of the moderation values. It's an easy mistake for somebody new here to make, as the "guidelines" are misleading.
The real meanings of the moderations are:
Insightful
"Him say things me not understand - him smart!"
Informative
"Him put link in post!"
Interesting
"Me agree with him!"
Underrated
"Me agree with him, but me afraid of metamods!"
Troll
"YOU RONG! YOU SUCK!"
Flamebait
(see Troll)
Offtopic
"Me not like your response to me post about thing not in story."
Overrated
"YOU RONG! but me afraid of metamods!"
Remember, the rules here are very simple (as are most of the users):
Asking others to take responsibility for their actions is BAD and WRONG and TROLL!
Stealing from big business is GOOD and INSIGHTFUL.
Stealing from Free Software is BAD and WRONG and FLAMEBAIT.
There was the time I was working on an amateur radio repeater that was co-located with several commercial systems, including a commercial FM radio station. We had my COM-120B (a US$20,000 piece of test equipment). At the same time the FM radio station's engineers were out working on their gear, with a IFR-500S (a decade+ old US$10,000 piece of gear that cannot do half the things the 120 can). It was kind of funny watching those guys look over at us with envy.
Then there were the times a friend of mine borrowed the telephone company's bucket truck to help me put up my tower. Of course, he used to own the phone company, so....
"I don't think there is a chance in hell of our getting a million downloads in four days, so I will make this grandiose gesture knowing full well I will never have to put up on it - much like the Taco Bell/MIR offer.
However, I will use this PR stunt to get lots of free advertising from lazy reporters who are too stupid to figure out how I am using them."
No, the circuit labeled "voltage doubler" is incorrectly labeled - it is a full wave recitfier, just not the "classical" configuration.
Look up "Rectifier" in Wikipedia - they got it right.
Actually, most switching power supplies use a modification and combination of the two approaches - for a nominal 120VAC they use the 2 diode configuration and feed the other leg into the common point between both caps, and for a nominal 240VAC they feed the other leg into a second pair of diodes - allowing 300VDC to be generated from both 120VAC and 240VAC with a single switching circuit.
Of course, in a modern power supply design neither of these are acceptable as they BOTH generate a terrible power factor due to the caps only charging at the voltage crests - hence the need for a power factor correction circuit before the rectifiers in order to get certification for the EU.
First of all - have you ever designed a power supply, or worked with people who have? I do.
I suggest you go look at Vicor's web site to get some basic information about modern switched power supplies, as you don't seem to know as much as you think you do.
120VAC is 170Vpeak-center or 340 Vp-p. Full wave rectifying the voltage gives you about 300V (minus losses). If you doubt me, do the math on sine wave to RMS.
It is *not* just a matter of varying the duty cycle of a PWM supply - the very long pulse widths saturate the coil and cause losses, and the very short duty cycles cause increased switching losses due to the rise time of the switches.
The main power bus will be about 170VDC, since the power brick will presumably be AC powered.
Not necessarily - remember the requirement to accept 12VDC input? You can do one of two things - you can run from the 12V bus, or you can have a second DC/DC converter to take 12VDC to the higher voltage.
Besides, you wouldn't generate 170VDC, but 300VDC - you full wave rectify the signal and dump the +170VDC into one cap, and the -170VDC into another.
80% efficiency is unacceptable here. With decent quality components, you should be able to get up to about 95% efficiency out of a switching power supply...
Only for a fixed output voltage design running off a high voltage main bar. If you want to support a wide range of voltages (like the requirements state) you lose efficiency, and if you want to run off a low voltage buss you lose efficiency, and if you want to cascade supplies (either 120VAC to 300VDC to 12VDC to output, or 12VDC to 300VDC to output, as required to allow for both AC mains and battery operation) your efficiency goes down as the product of the stages.
A shared ground is absolutely unacceptable in this application.
And if you noticed, I addressed this - you can do this, but then each of your final output stages needs to be a fully galvanicly isolated flyback design - which is more expensive and less efficient than the other possible topologies that do not provide ground isolation.
Let us examine the implications of each of your requirements in turn to see what impact they will have on the final design:
5 output ports
In and of itself, not a real problem - but it does imply that the final output circuits will have to be replicated 5 times, which will increase the bill of materials.
This also runs afowl of the "0-1-infinity" rule - if you provide N ports, the user will need N+1 ports - and will bitch mightily that "you were a moron for only supplying N ports since everybody knows you need N+1 ports you 'tard!"
Each port variable from 2V to 24V
That is really a very wide voltage swing. If the main internal power bus is less than 26V, it means each output module must be a buck/boost converter (which is more complex and expensive than the other solutions, as well as less efficient).
each port capable of 50 watts without significant voltage drop
The term "voltage drop" is meaningless in this context - what you really mean is "with good regulation." Voltage drop would be the loss due to resistance from the power module to the device being powered. Now, for 24V a 50W supply is 2 amps - not a really big problem. However, for 2V a 50W supply is 25A - that takes BIG wire to carry, and special connectors. If you are willing to specify that power in terms of a given voltage (e.g. 50W max at 24V, max 4A current) this problem goes away.
Then there is the fact that your 5 ports at 50W each is 250W of power delivered to the devices - allowing for a total effeciency of 80% that is 313W input from the wall, and 63W of heat dissipated in the power system.
a handful of accessory connectors / converters including 1 plug male -> 4 plugs female, big diameter plug to small plug, extension cables, etc.
No real problem except that this is more stuff to lose - you would want to sell these things seperately and let the end user buy what he needs when he needs it.
silent power supply if possible / air cooled, or very, very quiet fan.
See the power requirements - you are going to have a problem without forced air cooling, and if you want that to be quiet you have to have a fairly large device so that you can move lots of air slowly.
Use a 12VDC battery as input
Again, this means that each section has to be a buck/boost converter. You also are going to be pulling 26A from that battery - you will need, again, heavy wires and special connectors for this. You also run into the problem of supplying a battery box for this - you will NOT get UL/CE approval if you just allow Jethro to set a wet cell car battery down on the floor next to this thing. You also run into the problem of adding charging circuits (more cost) to the power supply, adding short-circuit protection on the battery terminals, and the fact that you may need to draw as much as 600W to run all the gizmos and charge the battery.
You also have the problem that, in general, the efficiency of a DC/DC converter goes down as the input voltage goes down - you are MUCH more efficient making power from a 300V rail than from a 12V rail.
You also run into the possiblity of ground loops - you have one ground path from the computer to the gizmo, and one ground path from the gizmo to the power supply. The noise pickup will be a function of the loop area, which will be very large, and of the noise injection of all the other devices. Seperate transformers break the ground loop by not having a DC path from the gizmo to the wall. While it is possible to break the ground loop the same way with a bus power device like this, it increases the cost of the final output stages.
Now, IF we could convince the gizmo makers to standardize on a small set of standard voltages, and IF we could get them to design their devices with the possiblity of ground loops and power supply noise in mind, we could have a device that made those standard voltages and a distribution mechanism for them.
And if all the/bots would boycott bad Star Wars films George Lucas would stop making them.
All the mapping programs I've worked with treat ferry routes "like a road" - they have the ferry route in the database just as they would a road, and they just have a flag that says "This is a ferry route".
I have yet to see a routing program that knew of the schedule of the ferry and would then say "At this point, wait 6.3 hours for the ferry to arrive".
The problem with extending this to the airlines would be the time dependence of the different routes. While it does not matter WHEN I arrive at the junction of I-35 and I-40 (I-40 will still be there and still be available), if my plane arrives 1 hour too late, the next plane will be gone, and the route will be invalidated.
And then there is the fact that the "cost" of taking a given road is pretty easily computed - take the length of the road, multiply by an estimated speed for that type of road - done. Now, try to compute the "cost" of a flight - is that super-duper-mega-saver-standby, or "I need to fly out TONIGHT" pay through the nose?
Having a routing program that takes into account the different costs of air travel - especially when you remove the constraints on city of departure and arrival - you'd get into a VERY nasty bit of design. Consider: is it faster/cheaper to drive to Wichita, get a plane to Dallas, get a plane to LA, drive to Acton, or is it cheaper to fly into Burbank, or into Flagstaff, or to drive to OKC and fly out, or to KCI, or to Salina, or.... - and each of those choices imposes time constraints on the connections.
I won't speculate on the P/NP completeness of the problem, and I won't say that it COULDN'T be solved electronically, but I think you'd still be better planning the air part of the trips with a dedicated tool and much input from the user, then planning the road trip side of things with a map program.
They used to be there to protect the workers, and they attracted leaders who were concerned with protecting the workers. Then they grew large, and powerful, and their leaders became those who were concerned with being in charge of large and powerful organizations. The unions began to act not to protect the workers, but to protect the size and power of the unions.
(Note: you can s/union/political party/g , or s/union/businesses/g, or s/union/government/g. and replace "workers" as appropriate, and not change the truth of this - this is not (just) a bash against the unions.)
One way to help thin this sort of thing down a bit is to use a non-standard high port (above 1024) for your SSH daemon.
This keeps the 5|<r1p7 |<1dd3z from being able to trivially find you SSH server.
Ideally, you want to do this in combination with code that watches for a port-scan and adds a firewall rule to block the scanning address.
Yes, this won't completely stop abuses of your SSH server - there's always a chance that somebody will stumble across it, so you should keep it up to date on security patches and disable password login (in other words REQUIRE a keypair to access). But this greatly reduces the amount of crap
I applaud the efforts of these hardware vendors to support Free drivers, but:
These guys are NOT the market leaders - they are very much in a niche market. They are behind and falling even more behind in terms of market penetration - as more embedded systems move to Intel, nVidia, and ATI these guys lose ground.
They are hoping that by having Free drivers, they will pick up some sales in embedded widgets that they might otherwise not get. Yes, these guys make their money on hardware, AND they don't feel there is any "magic" IPR that can be exposed by releasing their programming specifications.
So, what is the likely impact upon ATI and nVidia?
Almost none.
Now, *IF* we start seeing ATI and nVidia slipping in their sales to hardware vendors (laptop makers, makers of embedded widgets, etc.) *then* we *might* see ATI and nVidia getting worried. And if a winning lottery ticket happens to blow into my hand I might be a millionare.
But since the segment of the embedded market to which ATI and nVidia sell (as well as the laptop market and motherboard manufacturing market) are still 90% or more "powered by Windows®", there is still little motivation for ATI or nVidia to re-examine their stance on releasing information about their chips - they will continue to fear that releasing programming specs will somehow reveil the "magic" in their chips.
The purpose of all the new TLDs is "To allow domain registrars to make more money."
That is why every little movie simply must have its own.com domain, rather than having a virtual directory under the domain of the publisher - e.g. paramount.com/drecky_summer_movie/
That is why Joe's garage on the corner down the street must have JoesGarage.com, or at least JoesGarageAtFifthAndMain.com, rather than joesgarage.ict.ks.us.
Domains get cheaper the further down the heirarchy they get - domain registrars cannot charge as much for *.lawyers.com as they can for *.com.
Unlike physical items like land or gold, new TLDs can be created ad infinitum, so the registrars "figger" (they don't "figure" or "reason" or "think" - that is beyond them) they can get ICANN to keep creating new TLDs and they can continue to make the same amount of money forever.
Of course, that has worked out so spectactularly well in the case of.biz - after all, I know that when I see a.biz domain I feel great trust for the domain holder, as we all know that.biz mean business, and that anybody with a.biz must therefor be trustworthy!
. . . .
Excuse me, I had to replace the sarcasm fuse in my keyboard.
You either a) use something other than a transistor to create the signal, such as a tube (tubes do high frequency, high voltage, and high power really well). or b) you use the transistor itself - you wire it up as an oscillator and see how fast it can run.
This sounds an awful lot like they are giving the zero-gain bandwidth of the transistor - the frequency at which the transistor does NOT amplfy a signal anymore.
So, at 599GHz the transistor will amplify a little. At 600 GHz the transistor takes as much power to drive the input as it is able to switch at the output. At 601 GHz the transistor takes more power to control than it can switch.
Given a 600 GHz zero-gain bandwidth transistor you ARE NOT going to make a 600 GHz clockspeed processor.
He built the place. It started out as a planetarium at a state fair, and Max (and Patty Carey) worked their asses off to make it one of the leading space museums in the world.
He is ANYTHING but a moron. He was one of the cageiest individuals around. He spent years combing junkyards in Florida, Huston, and Huntsville, finding gear that NASA had thown away when the program it was associated with was no longer funded.
He found the best people to restore the artifacts, and built a museum collection that was the envy of other space museums.
Before you spout off on the subject (and moderators, before you moderate this tripe as insightful) you might want to actually do some research on the history of the Cosmosphere.
All of that makes this EVEN WORSE - Max could have just as easily continued to do as he had done, locating artifacts in junkheaps, having Spaceworks (the artifact restoration arm of the Cosmosphere) restore them, and legally sell them. He didn't have to do this!
And if he did indeed misappropriate artifacts (and while it sure looks that way, do remember - he has not yet been convicted in a court of law), then that was not merely a carrer-limiting move, that was a carrer-ending move - no museum will ever touch him again.
I've been saying for some time I'd like to see a FPS incorporate light discipline and sound discipline - in other words, standing in a doorway with bright light behind you makes you a good target for the bad guys, waving your flashlight around needlessly lets the bad guys know where you are, making lots of noise lets the bad guys know where you are....
Imagine having to dive from a brightly lit room into a dark room quickly (in order to not get shot when you are in the door), then having to wait for a while to let your eyes adjust, then flipping a light switch and watching the bad guys shoot out the lights....
A robot to grade the site - level the dirt off. A robot to dig the foundation. A robot to do the walls (in progress). A robot to do the interior finish work. A robot to do the exterior finish work.
And a really ANNOYING robot to run around with a megaphone yelling at all the other robots "Four Hours! We only have FOUR HOURS until the family gets back!" and "Mr. Robotic Bus Driver - MOVE THAT BUS!"
--
Seriously, give the amount of pre-fabrication that can be done on a conventional house now (wall segments, roof joists, and floors built at the factory and shipped out), I wonder if a concrete extruder is the best way to go - perhaps building a human-controlled crane with more dexterity to raise the pre-fab wall sections, and a computer-controlled wall-nailer, and so on would be a better approach to achive rapid and cost-effective house construction than a house-sized 3D litho rig.
The car analogy is flawed, because you can order your car with quite a bit of customization.
For example, when I had to buy a new car (damn soccer moms in minivans not paying attention to where they are going!) I was very clear to the dealer: "No leather seats. Cloth Seats."
The dealer's response was "But leather seats are on sale right now - no added cost! A $900 value!"
"No. I do not want leather seats. They take more care than cloth. They are too hot in the summer and too cold in the winter unless they are cooled and heated - and then that is just one more thing to go wrong. Cloth seats."
I got my cloth seats.
This bunk that they cannot sell a laptop without an OS because:
"We need to be able to test the machine." (then test it and scrub the disk when you are done.)
"We cannot ship without an operating system." (fine - install FreeDOS on it.)
"We cannot refund your money - we have an agreement with Microsoft." (The court says differently.)
The wear and tear upon the bulb and ballast also varies based upon the type of ballast.
The old style "dumb" ballast and ignition system (which used a bimetallic switch to short the bulb out, heat the electrodes in the bulb, and build up a magnetic field on the ballast, then opened the switch to generate a high-voltage spike to strike the arc) were much more intolerant of start cycles than modern, high-frequency switching ballasts.
The newer ballasts (which also have the advantage of increasing the frequency of the current through the bulb from 60Hz to several tens of kilohertz) don't take nearly as much wear from a start cycle.
Slashdot Mirror
The thing to do is to hold the ISP accountable if they don't hold the user accountable.
For example - I had this host that kept sending me half-megabyte virus executables via mail. I identified the ISP as Netvision in Israel. I tried to contact them repeately. They did nothing to stop this - they did not contact the user, they did not disconnect the user, they did not block the user's ability to send mail, NOTHING.
In cases like this, then HELL YES I say hold the ISP accoutable - they have failed to hold the user accountable.
If I start making prank calls from my phone, the phone company will kill my line if they get called about it. ISPs should be no exception.
That last bit about the radio station is definitely in the catagory of "Not Smart" - a conviction by the FCC for such activities might be just a bit of a rough spot in a job interview for a radio tech.
But yes, it's a bit scary to think about how many millions of dollars of test gear spends most of it's time being a radio reciever - I'm sure that this happens at R&S, Tek, and Agilent too.
How'd you pull off the TV trick? I'm assuming you used the HP to get the video signal, one 500 to get the audio, and the Tek to display, but how'd you get the raster?
I was wondering about this as well - the number of observations we are making of planets in other solar systems is already astonishingly large given the angular distance these objects subtend as viewed from Earth (true, many of the observations are indirect and thus don't directly depend upon the angular distance).
I wonder if the breakthroughs in bypassing the diffraction limit will allow for direct imaging of larger bodies (Jovian worlds at Jovian orbital distances).
You see, you fatally misunderstand the meaning of the moderation values. It's an easy mistake for somebody new here to make, as the "guidelines" are misleading.
The real meanings of the moderations are:
Insightful "Him say things me not understand - him smart!" Informative "Him put link in post!" Interesting "Me agree with him!" Underrated "Me agree with him, but me afraid of metamods!" Troll "YOU RONG! YOU SUCK!" Flamebait (see Troll) Offtopic "Me not like your response to me post about thing not in story." Overrated "YOU RONG! but me afraid of metamods!"
Remember, the rules here are very simple (as are most of the users):
Well, let's see.
There was the time I was working on an amateur radio repeater that was co-located with several commercial systems, including a commercial FM radio station. We had my COM-120B (a US$20,000 piece of test equipment). At the same time the FM radio station's engineers were out working on their gear, with a IFR-500S (a decade+ old US$10,000 piece of gear that cannot do half the things the 120 can). It was kind of funny watching those guys look over at us with envy.
Then there were the times a friend of mine borrowed the telephone company's bucket truck to help me put up my tower. Of course, he used to own the phone company, so....
In other words:
"I don't think there is a chance in hell of our getting a million downloads in four days, so I will make this grandiose gesture knowing full well I will never have to put up on it - much like the Taco Bell/MIR offer.
However, I will use this PR stunt to get lots of free advertising from lazy reporters who are too stupid to figure out how I am using them."
No, the circuit labeled "voltage doubler" is incorrectly labeled - it is a full wave recitfier, just not the "classical" configuration.
Look up "Rectifier" in Wikipedia - they got it right.
Actually, most switching power supplies use a modification and combination of the two approaches - for a nominal 120VAC they use the 2 diode configuration and feed the other leg into the common point between both caps, and for a nominal 240VAC they feed the other leg into a second pair of diodes - allowing 300VDC to be generated from both 120VAC and 240VAC with a single switching circuit.
Of course, in a modern power supply design neither of these are acceptable as they BOTH generate a terrible power factor due to the caps only charging at the voltage crests - hence the need for a power factor correction circuit before the rectifiers in order to get certification for the EU.
First of all - have you ever designed a power supply, or worked with people who have? I do.
I suggest you go look at Vicor's web site to get some basic information about modern switched power supplies, as you don't seem to know as much as you think you do.
120VAC is 170Vpeak-center or 340 Vp-p. Full wave rectifying the voltage gives you about 300V (minus losses). If you doubt me, do the math on sine wave to RMS.
It is *not* just a matter of varying the duty cycle of a PWM supply - the very long pulse widths saturate the coil and cause losses, and the very short duty cycles cause increased switching losses due to the rise time of the switches.
Not necessarily - remember the requirement to accept 12VDC input? You can do one of two things - you can run from the 12V bus, or you can have a second DC/DC converter to take 12VDC to the higher voltage.
Besides, you wouldn't generate 170VDC, but 300VDC - you full wave rectify the signal and dump the +170VDC into one cap, and the -170VDC into another.
Only for a fixed output voltage design running off a high voltage main bar. If you want to support a wide range of voltages (like the requirements state) you lose efficiency, and if you want to run off a low voltage buss you lose efficiency, and if you want to cascade supplies (either 120VAC to 300VDC to 12VDC to output, or 12VDC to 300VDC to output, as required to allow for both AC mains and battery operation) your efficiency goes down as the product of the stages.
And if you noticed, I addressed this - you can do this, but then each of your final output stages needs to be a fully galvanicly isolated flyback design - which is more expensive and less efficient than the other possible topologies that do not provide ground isolation.
5 output ports In and of itself, not a real problem - but it does imply that the final output circuits will have to be replicated 5 times, which will increase the bill of materials.
Each port variable from 2V to 24V That is really a very wide voltage swing. If the main internal power bus is less than 26V, it means each output module must be a buck/boost converter (which is more complex and expensive than the other solutions, as well as less efficient). each port capable of 50 watts without significant voltage drop The term "voltage drop" is meaningless in this context - what you really mean is "with good regulation." Voltage drop would be the loss due to resistance from the power module to the device being powered. Now, for 24V a 50W supply is 2 amps - not a really big problem. However, for 2V a 50W supply is 25A - that takes BIG wire to carry, and special connectors. If you are willing to specify that power in terms of a given voltage (e.g. 50W max at 24V, max 4A current) this problem goes away.This also runs afowl of the "0-1-infinity" rule - if you provide N ports, the user will need N+1 ports - and will bitch mightily that "you were a moron for only supplying N ports since everybody knows you need N+1 ports you 'tard!"
a handful of accessory connectors / converters including 1 plug male -> 4 plugs female, big diameter plug to small plug, extension cables, etc. No real problem except that this is more stuff to lose - you would want to sell these things seperately and let the end user buy what he needs when he needs it. silent power supply if possible / air cooled, or very, very quiet fan. See the power requirements - you are going to have a problem without forced air cooling, and if you want that to be quiet you have to have a fairly large device so that you can move lots of air slowly. Use a 12VDC battery as input Again, this means that each section has to be a buck/boost converter. You also are going to be pulling 26A from that battery - you will need, again, heavy wires and special connectors for this. You also run into the problem of supplying a battery box for this - you will NOT get UL/CE approval if you just allow Jethro to set a wet cell car battery down on the floor next to this thing. You also run into the problem of adding charging circuits (more cost) to the power supply, adding short-circuit protection on the battery terminals, and the fact that you may need to draw as much as 600W to run all the gizmos and charge the battery.Then there is the fact that your 5 ports at 50W each is 250W of power delivered to the devices - allowing for a total effeciency of 80% that is 313W input from the wall, and 63W of heat dissipated in the power system.
You also have the problem that, in general, the efficiency of a DC/DC converter goes down as the input voltage goes down - you are MUCH more efficient making power from a 300V rail than from a 12V rail.
You also run into the possiblity of ground loops - you have one ground path from the computer to the gizmo, and one ground path from the gizmo to the power supply. The noise pickup will be a function of the loop area, which will be very large, and of the noise injection of all the other devices. Seperate transformers break the ground loop by not having a DC path from the gizmo to the wall. While it is possible to break the ground loop the same way with a bus power device like this, it increases the cost of the final output stages.
Now, IF we could convince the gizmo makers to standardize on a small set of standard voltages, and IF we could get them to design their devices with the possiblity of ground loops and power supply noise in mind, we could have a device that made those standard voltages and a distribution mechanism for them.
And if all the
I'm not holding my breath.
Ehhh, not quite.
All the mapping programs I've worked with treat ferry routes "like a road" - they have the ferry route in the database just as they would a road, and they just have a flag that says "This is a ferry route".
I have yet to see a routing program that knew of the schedule of the ferry and would then say "At this point, wait 6.3 hours for the ferry to arrive".
The problem with extending this to the airlines would be the time dependence of the different routes. While it does not matter WHEN I arrive at the junction of I-35 and I-40 (I-40 will still be there and still be available), if my plane arrives 1 hour too late, the next plane will be gone, and the route will be invalidated.
And then there is the fact that the "cost" of taking a given road is pretty easily computed - take the length of the road, multiply by an estimated speed for that type of road - done. Now, try to compute the "cost" of a flight - is that super-duper-mega-saver-standby, or "I need to fly out TONIGHT" pay through the nose?
Having a routing program that takes into account the different costs of air travel - especially when you remove the constraints on city of departure and arrival - you'd get into a VERY nasty bit of design. Consider: is it faster/cheaper to drive to Wichita, get a plane to Dallas, get a plane to LA, drive to Acton, or is it cheaper to fly into Burbank, or into Flagstaff, or to drive to OKC and fly out, or to KCI, or to Salina, or.... - and each of those choices imposes time constraints on the connections.
I won't speculate on the P/NP completeness of the problem, and I won't say that it COULDN'T be solved electronically, but I think you'd still be better planning the air part of the trips with a dedicated tool and much input from the user, then planning the road trip side of things with a map program.
No.
Unions are there to protect the unions.
They used to be there to protect the workers, and they attracted leaders who were concerned with protecting the workers. Then they grew large, and powerful, and their leaders became those who were concerned with being in charge of large and powerful organizations. The unions began to act not to protect the workers, but to protect the size and power of the unions.
(Note: you can s/union/political party/g , or s/union/businesses/g, or s/union/government/g. and replace "workers" as appropriate, and not change the truth of this - this is not (just) a bash against the unions.)
One way to help thin this sort of thing down a bit is to use a non-standard high port (above 1024) for your SSH daemon.
This keeps the 5|<r1p7 |<1dd3z from being able to trivially find you SSH server.
Ideally, you want to do this in combination with code that watches for a port-scan and adds a firewall rule to block the scanning address.
Yes, this won't completely stop abuses of your SSH server - there's always a chance that somebody will stumble across it, so you should keep it up to date on security patches and disable password login (in other words REQUIRE a keypair to access). But this greatly reduces the amount of crap
And don't forget:
The case is not made of magnetically neutralized plastic.
The circuit board is not pressed virgin Yak wool.
There's not a tube in sight.
No, they will do what they do already - they will type the movie name into [google|yahoo|MSN] and be taken to the site by the search engine.
I applaud the efforts of these hardware vendors to support Free drivers, but:
These guys are NOT the market leaders - they are very much in a niche market. They are behind and falling even more behind in terms of market penetration - as more embedded systems move to Intel, nVidia, and ATI these guys lose ground.
They are hoping that by having Free drivers, they will pick up some sales in embedded widgets that they might otherwise not get. Yes, these guys make their money on hardware, AND they don't feel there is any "magic" IPR that can be exposed by releasing their programming specifications.
So, what is the likely impact upon ATI and nVidia?
Almost none.
Now, *IF* we start seeing ATI and nVidia slipping in their sales to hardware vendors (laptop makers, makers of embedded widgets, etc.) *then* we *might* see ATI and nVidia getting worried. And if a winning lottery ticket happens to blow into my hand I might be a millionare.
But since the segment of the embedded market to which ATI and nVidia sell (as well as the laptop market and motherboard manufacturing market) are still 90% or more "powered by Windows®", there is still little motivation for ATI or nVidia to re-examine their stance on releasing information about their chips - they will continue to fear that releasing programming specs will somehow reveil the "magic" in their chips.
The purpose of all the new TLDs is "To allow domain registrars to make more money."
.com domain, rather than having a virtual directory under the domain of the publisher - e.g. paramount.com/drecky_summer_movie/
.biz - after all, I know that when I see a .biz domain I feel great trust for the domain holder, as we all know that .biz mean business, and that anybody with a .biz must therefor be trustworthy!
That is why every little movie simply must have its own
That is why Joe's garage on the corner down the street must have JoesGarage.com, or at least JoesGarageAtFifthAndMain.com, rather than joesgarage.ict.ks.us.
Domains get cheaper the further down the heirarchy they get - domain registrars cannot charge as much for *.lawyers.com as they can for *.com.
Unlike physical items like land or gold, new TLDs can be created ad infinitum, so the registrars "figger" (they don't "figure" or "reason" or "think" - that is beyond them) they can get ICANN to keep creating new TLDs and they can continue to make the same amount of money forever.
Of course, that has worked out so spectactularly well in the case of
.
.
.
.
Excuse me, I had to replace the sarcasm fuse in my keyboard.
You either a) use something other than a transistor to create the signal, such as a tube (tubes do high frequency, high voltage, and high power really well). or b) you use the transistor itself - you wire it up as an oscillator and see how fast it can run.
OK, I cannot RTFM right now as it is /.ed, but:
This sounds an awful lot like they are giving the zero-gain bandwidth of the transistor - the frequency at which the transistor does NOT amplfy a signal anymore.
So, at 599GHz the transistor will amplify a little. At 600 GHz the transistor takes as much power to drive the input as it is able to switch at the output. At 601 GHz the transistor takes more power to control than it can switch.
Given a 600 GHz zero-gain bandwidth transistor you ARE NOT going to make a 600 GHz clockspeed processor.
How Max got the job at the Cosmosphere is simple.
He built the place. It started out as a planetarium at a state fair, and Max (and Patty Carey) worked their asses off to make it one of the leading space museums in the world.
He is ANYTHING but a moron. He was one of the cageiest individuals around. He spent years combing junkyards in Florida, Huston, and Huntsville, finding gear that NASA had thown away when the program it was associated with was no longer funded.
He found the best people to restore the artifacts, and built a museum collection that was the envy of other space museums.
Before you spout off on the subject (and moderators, before you moderate this tripe as insightful) you might want to actually do some research on the history of the Cosmosphere.
All of that makes this EVEN WORSE - Max could have just as easily continued to do as he had done, locating artifacts in junkheaps, having Spaceworks (the artifact restoration arm of the Cosmosphere) restore them, and legally sell them. He didn't have to do this!
And if he did indeed misappropriate artifacts (and while it sure looks that way, do remember - he has not yet been convicted in a court of law), then that was not merely a carrer-limiting move, that was a carrer-ending move - no museum will ever touch him again.
I've been saying for some time I'd like to see a FPS incorporate light discipline and sound discipline - in other words, standing in a doorway with bright light behind you makes you a good target for the bad guys, waving your flashlight around needlessly lets the bad guys know where you are, making lots of noise lets the bad guys know where you are....
Imagine having to dive from a brightly lit room into a dark room quickly (in order to not get shot when you are in the door), then having to wait for a while to let your eyes adjust, then flipping a light switch and watching the bad guys shoot out the lights....
So, we need:
A robot to grade the site - level the dirt off.
A robot to dig the foundation.
A robot to do the walls (in progress).
A robot to do the interior finish work.
A robot to do the exterior finish work.
And a really ANNOYING robot to run around with a megaphone yelling at all the other robots "Four Hours! We only have FOUR HOURS until the family gets back!" and "Mr. Robotic Bus Driver - MOVE THAT BUS!"
--
Seriously, give the amount of pre-fabrication that can be done on a conventional house now (wall segments, roof joists, and floors built at the factory and shipped out), I wonder if a concrete extruder is the best way to go - perhaps building a human-controlled crane with more dexterity to raise the pre-fab wall sections, and a computer-controlled wall-nailer, and so on would be a better approach to achive rapid and cost-effective house construction than a house-sized 3D litho rig.
For example, when I had to buy a new car (damn soccer moms in minivans not paying attention to where they are going!) I was very clear to the dealer: "No leather seats. Cloth Seats."
The dealer's response was "But leather seats are on sale right now - no added cost! A $900 value!"
"No. I do not want leather seats. They take more care than cloth. They are too hot in the summer and too cold in the winter unless they are cooled and heated - and then that is just one more thing to go wrong. Cloth seats."
I got my cloth seats.
This bunk that they cannot sell a laptop without an OS because:
Is just that - bunk.
The wear and tear upon the bulb and ballast also varies based upon the type of ballast.
The old style "dumb" ballast and ignition system (which used a bimetallic switch to short the bulb out, heat the electrodes in the bulb, and build up a magnetic field on the ballast, then opened the switch to generate a high-voltage spike to strike the arc) were much more intolerant of start cycles than modern, high-frequency switching ballasts.
The newer ballasts (which also have the advantage of increasing the frequency of the current through the bulb from 60Hz to several tens of kilohertz) don't take nearly as much wear from a start cycle.
The "cost" of turning on a flourescent light being higher than leaving it running is an urban myth.
Yes, a flourescent takes more power for a few cycles when it strikes.
The total energy taken to strike the arc in the light is less than a few seconds of runtime.