Am I the only one who finds it obnoxious that a gig or two of drive space is locked up in the restore partition, only to be used during such a reinstall? It could just as easily be kept on CD, returning the drive space to active use.
Speaking of which, when is someone gonna write a virus that infects the system restore partition, so it can survive such "clean" reformats?
The trouble here is that I don't believe consciousness is easily measured in binary terms. A lot of the things that make it work are analog, by which I'm referring to things like the size of synaptic gaps, the positions of neurons, and the minute influence of things like external magnetic fields on the motion of chemicals in the brain.
To what resolution can you measure those things? What is the smallest change in a parameter that would affect the outcome of a thought?
If the universe were a pegboard and matter could only occupy definite locations, then you'd be able to describe the brain's layout in terms that would make it easy to quantify and store. But because position is analog, I submit that measuring the capacity for consciousness is folly.
It's quite a big assumption to say that a certain number of bits, or energy states, are required to cause a thought or a perception. Yes, the energy can be measured in discrete quantities, but the actions it generates have to cross analog distances, and it's those nuances that contain more detail than a simple quantified representation can hold.
But... I got to talking about this with my brother a while back, and the way he was explaining the math, it seems that as you exceed light speed, yes everything does flip negative, but what happens as you break two or three times c? The way it worked in my head, you'd experience normal time if you were moving at 3c.
Personally, I'm more concerned with the poorly understood reaction between the DoE and actual science.
The way I see it, cold fusion is such a tremendously holy grail, and the Pons-Fleischman experiment was simple enough to replicate, it would've made more sense to throw some more experimental funding at it years ago. A handful of failed attempts to replicate the results are discouraging, yes, but the potential benefits should've justified a bit more tinkering back when it was announced.
Maybe I'm missing it, maybe the threshhold of debunking was passed and everyone gave up on it as a fluke. Maybe it still is a fluke, albeit a somewhat more convincing one.
Obviously not the whole scientific community gave up on the idea, or today's announcement never would've happened. What did these folks know that kept them working on it?
In other news, automated cameras have been installed, issuing tickets to those who run red lights in the middle of the night when there's no other traffic, including police, around.
Vehicles have been equipped with "black box" devices, recording operations without the driver's knowledge or consent.
Eavesdropping equipment has also been installed in new vehicles, giving the ability to listen in on the occupants at any time.
Law enforcement uses special equipment to "see through" walls and observe the occupants inside a building, without a warrant because it's observable from the street.
I started this post attempting to be sarcastic, but every terrifying example of surveillance I could come up with has already been implemented.
Ahh, good find.:) If I'd read your comment first, I wouldn't have had to google for it...
The virtual gramophone is an awesome project and I hope someone with the requisite skills will pick it up and do something with it. There's a pile of 78's in the bottom of my grandma's victrola that I'd love to clean up nondestructively.
Since all this stuff is out of copyright, there should be no problem sharing it for everyone to enjoy. Consider two records of the same music, damaged and scratched in different areas. Could the recordings from both of them be lined up and used to fill in each other's gaps?
My favorite is on the side of the Wheaties box, there's a note above the ingredients list, "CAUTION: CONTAINS WHEAT"
It's reached a point now where I don't take any warning labels seriously, because I'd never get anything done. Someday I'm going to ignore one and hurt myself, and I'm going to sue every product liability lawyer in the country for dilluting the meaning of the word "warning".
I agree that the failure rate between rugged and standard hardware won't be different enough to justify the price delta. People will find a way to break the rugged stuff, if only to prove that it can be broken.
A padded neoprene sleeve with a vinyl window, slipped over a device that was designed well in the first place, should be plenty for most applications. If you really want rugged hardware, Google is much faster than Ask Slashdot, and it has a lower percentage of comments saying "you should've googled, idiot!".
Test screens for sunlight viewability. If beeps and boops are required, test speakers for audibility. Test batteries for operation over a wide temperature range, and make sure the manufacturer will replace any batteries that fail or degrade prematurely. Make sure dust and crud won't gum up the buttons or charging contacts.
Good luck getting wireless connectivity with no bits that stick out. Antennae don't last long, and they're hard to weatherproof.
The wireless part has me curious. How does your radio scheme deal with huge numbers of users in a small area? If you're trying to broadcast content to all of them at once, do the protocols support doing it with a single transmission? Does the application software?
Furthermore, for interactive applications where the user devices will be transmitting, does the system deal robustly and fairly with overload situations? Does efficiency plummet as utilization rises?
Given that the antennae integrated into handhelds are scarcely better than dummy loads, what sort of antenna gain do you need on the access point end to provide plenty of SNR? Consider the radio environment of the venue, given that metal surfaces will cause reflections and raise the noise floor. Also consider that reflections might make it difficult to divide the area into zones served by multiple access points, and devices might hop between points sporadically and quickly due to signal fluctuations.
Re:Unique? No... but legal questions?
on
Dual User Windows PC
·
· Score: 3, Informative
Yeah, it was called the Buddy, and the old incarnation sucked. It was a PCI board which was essentially a video card plus a PS/2 keyboard and mouse controller. The video capabilities were terrible (sync rate limited to 60Hz, IIRC), and the second keyboard was prone to random resets and other problems.
The new incarnation of Buddy seems to address all those problems by using standard hardware. Buddy and BeTwin (they look like the same software) appear to work with any PCI video board that'll coexist with other video hardware, and since they use USB keyboards and mice (and audio, if you want), the proprietary controller problems should be gone too.
The new Buddy doesn't stop at 2 stations, either. It'll happily run up to 5, which might have a chance of using some of the absurd CPU power available in a modern PC. They have a trial version up for download, I might have to check my hardware compatibility and tinker with it later.
Well, the important voltage of such a lamp is mostly determined by its length, so drop-in replacement should be 'close enough' for most applications. Don't save the whole screen for the lamp, just find a good sturdy box and some soft padding to store them in. The lamps have mercury in them and should be returned for recycling if you give up your packrat habit, or when one fails.
That pile of miscellaneous screws has saved my ass more times than I can count. Sometimes I need a dozen screws but only have one of the right kind, but being able to take that one to the hardware store makes it worth it.
Same here! Houses are so close together and all the big trees are gone, any noise in the neighborhood comes through clear as a bell. I rely on the noise of my machine to cover it up, and it's very hard to relax without that. I had the hardest time sleeping during the power outage.
"He said 'The streets were dark tonight, it was like another century, with dim lamps and candles lighting up the icy trees and the clouds and a covered moon.' She said 'What kind of people make a city where you can't see the sky and you can't feel the ground?'" -- Dar Williams
Tie it in with a pair of 3d monitor-glasses, and have it overlay the past over the present. Walk through NYC and see the shimmering LCD ghost of the twin towers superimposed on the skyline.
The trouble is that the whole system is AC and must be synchronized. All the shafts of all the turbines connected to the power grid are literally turning in sync. Adding distributed generation is tricky. The easiest way is to generate DC (or generate AC and rectify to DC) and then use a "grid-tie" inverter to make phase-correct AC.
Small grid-tie inverters use the grid itself to provide sync, and just match their output to it. The trouble with doing this on a large scale is that the whole system chases its tail, with no master clock. If grid-tie inverters had sync inputs, it would be trivial to connect them to a real reference clock.
Additionally, subtle phase changes on the grid itself are used as a signalling protocol between utilities, indicating an excess of capacity or demand. This would need to be overcome before synchronized distributed generation would be practical on a large scale.
Something similar exists; not a joke.
on
Usenet Audio
·
· Score: 1
There's streaming support in Freenet now, and the network has been behaving nicely enough lately that it almost ought to work. More nodes = better performance, so go play with it.
The trouble with usenet and Freenet is that they're both very high latency. Great for your everyday muzak feed, but not ideal if you want to run a breaking-news station.
Break one-nine for a clue check, anyone copy?
on
Mobile Wifi Backpack
·
· Score: 1
"every time you see a cop", this sounds a lot like CB radio.
Transmit messages between cars. "let me pass you, diesel dummy!" Also sounds a lot like CB.
Tell people what you're listening to. Ahh-yup. As if they'll care, but you can do that too.
Oh! And sending traffic reports by radio. That's surely a novel use for the technolo-- oh nevermind.
and they seem to be doing pretty well. I went looking for weird NIC hardware and came across Imagestream. They make big routers with Linux at the core, on x86 hardware in industrial form factors. Definitely worth a look.
Also on the thread of interface cards, try Mikrotik. If you're doing wireless, the MiniPCI carrier boards will make your day.
Full disclosure: I'm not related to or affiliated with either of those companies in any way. I've never even bought anything from either of them. I just came across them while searching and thought they were bookmark-worthy.
Into the few-megahertz range, twisted pair wire works remarkably well. This is the stuff we're all familiar with as phone lines and cat-5. The number of twists per unit of length determines how resistant it is to interference, hence cat-5 is much more tightly twisted than cat-3. Each pair in a multipair cable is twisted a slightly different amount, to prevent inductive coupling and crosstalk between pairs.
The signal sent down a twisted pair is bipolar and "balanced", so that the two wires are carrying mirror opposite signals. There's an excellent explanation of this. T-1 signals ride twisted pair for several kilofeet between repeaters. The N-carrier system (low rate analog multiplex) also used twisted pair, but I don't know how far it would go between terminals.
Above a few megahertz, twisted pair gets unacceptably lossy and noisy. Higher speed signals are carried on coaxial cable, which we all know and love for its role in television wiring. The characteristic impedance of coax is determined by the ratio of the center conductor diameter and the distance to the inner surface of the outer conductor. Very early coaxial lines were made by suspending thin rods inside sections of copper pipe, by means of cardboard disc insulators. Soon a method of manufacturing flexible cable was developed, and has remained largely unchanged.
Signals carried on coax are "unbalanced", where the outer conductor is grounded and the inner conductor carries an AC wave. The need for the ground reference means that coax runs between buildings can become part of a ground loop, and cause all sorts of electrical problems. T-3 circuits use coax, but only for very short runs. (A T-3 that leaves a building does so as a DS-3 carried on fiber.) The L-carrier system, which multiplexed several N-carrier signals together, used thick coaxial lines for long-haul runs across the countryside.
As you approach the gigahertz range, coax also becomes too lossy, and hollow waveguide becomes the obvious choice. Waveguide can be rectangular, ovoid, or circular in cross-section, which effects the polarization of the signals carried in it. The inner dimensions influence loss and frequency range. Personally I'm not familiar with the buried waveguide system, but the TD and TH microwave systems used waveguide to connect the antennae with the terminal equipment.
Agreed one hundred percent! Fiber is useless for home applications. Just install lots of conduit, and put double-gang boxes on the ends so you'll have places to mount jacks for all the wires you'll eventually have in the conduit.
Power is a more important consideration than signal cabling. Run LOTS of branch circuits, ideally set up so that any given corner of a room has 2 circuits accessible to it. You don't want your UPS sharing a circuit with any outlets you plug a vacuum cleaner into, or the power quality log files will fill your drive.
Get a whole-house surge protector and pay the electrician to install it at the service entry point. While you're at it, get a second ground rod a few feet away from the first, and make sure the connections are done with exothermic welds. (crimps corrode in weather, and screw-clamps loosen with thermal fluctuations.)
Run a conduit to the front porch, for the doorbell and door camera / intercom. Run two to the back patio, for your outdoor speakers and network/phone jacks in various places. Drop 20-amp outlets all over the garage, and plan for a few up on the wall near shelving units, so you can park the battery charger/maintainer up there and plug it right in.
Bad idea. PVC fittings are made for liquid, not cable. They don't have a smooth internal surface, so things will snag up as you pull them.
Secondly, the bend radius of small PVC fittings is so tight that pulling any moderately stiff wire will be awkward past just 2 or 3 bends. Conduit can be bent with much gentler sweeps for easier pulling and less cable damage.
The stiff blue flexible plastic conduit is ideal for this, because it automatically sweeps corners as you install it.
Handy installation hint: You can buy "cable lubricant" which makes long pulls go smoothly. Liquid hand soap works just as well.
This sounds like the logical extension of the L-carrier systems. Before digital encoding was invented, radio techniques (frequency division multiplexing) were used to shift the frequency of each voice channel, and pack dozens of channels into a wide-band signal which could ride a twisted pair, or itself be muxed into a still wider signal, which was transmitted on coaxial cable.
I'm guessing that the megahertz-range signals on the coax were then muxed into gigahertz-range signals to be transmitted down the tubes. Fascinating.
WinCE devices aren't designed to run solo, they have to have a "partnership" with a "host" to do things like install software. I had an iPaq for a week before the cradle arrived, and I was trying to install the drivers for the CF wireless card. I had the Dual-CF sleeve for it, with the ma701 in the top slot and a 64-meg flash card in the bottom slot.
I figured I could download the driver package and load it onto the CF card at a desktop machine with a reader, then move the CF over to the ipaq, unpack it, and install it. No such luck. The driver ships as a win32.exe file which then loads the drivers onto the portable via the sync cable. Whose assinine idea was this? (don't answer, I already know!)
Not that I would've wanted to use an iPaq as my primary machine anyway, even with a foldup keyboard, the screen's too small and the viewing angle is terrible. However, there are a pair of WinCE devices I could imagine using as an everyday machine: The IBM Workpad z50, and the Vadem Clio (a.k.a. Sharp Tripad) are both laptop-style PDAs, with a clamshell hinge and a full size keyboard. They both have real PCMCIA slots, CF slots, hardware serial ports, and VGA screens. Both of them get approximately 8 hours on a battery, and the z50 is available with a double-capacity battery which honestly, realistically does get 16 hours.
Because of the WinCE codependency problem mentioned above, they're both unable to survive without occasional connection to a desktop for certain tasks. The easy solution is to ditch WinCE and run the hpcmips port of NetBSD, which boots on both of them. With a decent-sized CF card you can have a full development environment and not even need someone to cross-compile for you. That's definitely enough to be self-sufficient.
Full keyboard, harware serial port, and 8 hours on a charge. That's a recipe for "portable serial console" if I ever heard one. Oh yeah, it also happens to be a full-fledged NetBSD machine.:)
Yeah, I've switched to compact fluorescent lights in parts of the house for efficiency reasons, but I try to pair them with incandescent bulbs in fixtures near reading areas. I figure, half the efficiency benefit is better than none, and the reduction in flicker is essential.
I try to let natural light into the room when I'm looking at the monitor, but I'm careful to balance the contrast and avoid glare. If it's nighttime and I need light other than the monitor, I use an incandescent on a resistive dimmer (never a triac dimmer, they flicker too).
The most important thing has got to be focus though. A friend of mine has a degenerative disease where her eyes can only change focus slowly, over the span of several seconds or minutes. Looking from the road to the instrument panel and back to the road is impossible for her, so she has to read the gauges with her peripheral vision. (Analog gauges are easy, because you can tell the positions of the needles. Digital readouts are next to impossible.) When she uses a computer, she has to turn away and focus on the room for a moment before standing up to go get something.
I'm determined not to let this happen to me. I use my music collection to help -- at the beginning of each track, I close my eyes for a few seconds, take a deep breath, open my eyes, and look at something across the room. It's relaxing in terms of general stress, and it gives my eyes a chance to change focus distance.
Think about it -- our eyes weren't designed to stand still for long periods. In a survival situation, you're focusing near and far constantly, and your eyes are all over the place. Try to give them a little workout now and then.
Leave them running for 2 days and see if the air conditioning system can keep up.:)
If the office includes backup power, make sure the air conditioners are part of the protected load. If commercial power fails and the generator kicks on, your server room will keep running but get very hot, very fast without proper cooling.
IIRC, Duke Nukem 3D was the first FPS to utilize stereo sound hardware for positional audio. It was the first one I played, anyway.
It was possible, though not easy, to make it past the first level with the monitor turned off. Since rockets had a finite travel speed and made noise on impact, they could be used like "sonar" to determine distance to walls.
When a tree was hit, it would burn and make a crackling sound. When a fire hydrant was hit, it would burst open and make a water-spraying sound. All of these sounds would continue for several moments, during which the player could use them for position feedback while moving through the map.
As a decided retro-geek with a fondness for old hardware, I'm glad there's finally a new game that doesn't require a 3-d graphics card!
Slashdot Mirror
Am I the only one who finds it obnoxious that a gig or two of drive space is locked up in the restore partition, only to be used during such a reinstall? It could just as easily be kept on CD, returning the drive space to active use.
Speaking of which, when is someone gonna write a virus that infects the system restore partition, so it can survive such "clean" reformats?
The trouble here is that I don't believe consciousness is easily measured in binary terms. A lot of the things that make it work are analog, by which I'm referring to things like the size of synaptic gaps, the positions of neurons, and the minute influence of things like external magnetic fields on the motion of chemicals in the brain.
To what resolution can you measure those things? What is the smallest change in a parameter that would affect the outcome of a thought?
If the universe were a pegboard and matter could only occupy definite locations, then you'd be able to describe the brain's layout in terms that would make it easy to quantify and store. But because position is analog, I submit that measuring the capacity for consciousness is folly.
It's quite a big assumption to say that a certain number of bits, or energy states, are required to cause a thought or a perception. Yes, the energy can be measured in discrete quantities, but the actions it generates have to cross analog distances, and it's those nuances that contain more detail than a simple quantified representation can hold.
IANAphysicist!
But... I got to talking about this with my brother a while back, and the way he was explaining the math, it seems that as you exceed light speed, yes everything does flip negative, but what happens as you break two or three times c? The way it worked in my head, you'd experience normal time if you were moving at 3c.
Personally, I'm more concerned with the poorly understood reaction between the DoE and actual science.
The way I see it, cold fusion is such a tremendously holy grail, and the Pons-Fleischman experiment was simple enough to replicate, it would've made more sense to throw some more experimental funding at it years ago. A handful of failed attempts to replicate the results are discouraging, yes, but the potential benefits should've justified a bit more tinkering back when it was announced.
Maybe I'm missing it, maybe the threshhold of debunking was passed and everyone gave up on it as a fluke. Maybe it still is a fluke, albeit a somewhat more convincing one.
Obviously not the whole scientific community gave up on the idea, or today's announcement never would've happened. What did these folks know that kept them working on it?
In other news, automated cameras have been installed, issuing tickets to those who run red lights in the middle of the night when there's no other traffic, including police, around.
Vehicles have been equipped with "black box" devices, recording operations without the driver's knowledge or consent.
Eavesdropping equipment has also been installed in new vehicles, giving the ability to listen in on the occupants at any time.
Law enforcement uses special equipment to "see through" walls and observe the occupants inside a building, without a warrant because it's observable from the street.
I started this post attempting to be sarcastic, but every terrifying example of surveillance I could come up with has already been implemented.
We have always been at war with Eurasia.
Ahh, good find. :) If I'd read your comment first, I wouldn't have had to google for it...
The virtual gramophone is an awesome project and I hope someone with the requisite skills will pick it up and do something with it. There's a pile of 78's in the bottom of my grandma's victrola that I'd love to clean up nondestructively.
Since all this stuff is out of copyright, there should be no problem sharing it for everyone to enjoy. Consider two records of the same music, damaged and scratched in different areas. Could the recordings from both of them be lined up and used to fill in each other's gaps?
My favorite is on the side of the Wheaties box, there's a note above the ingredients list, "CAUTION: CONTAINS WHEAT"
It's reached a point now where I don't take any warning labels seriously, because I'd never get anything done. Someday I'm going to ignore one and hurt myself, and I'm going to sue every product liability lawyer in the country for dilluting the meaning of the word "warning".
I agree that the failure rate between rugged and standard hardware won't be different enough to justify the price delta. People will find a way to break the rugged stuff, if only to prove that it can be broken.
A padded neoprene sleeve with a vinyl window, slipped over a device that was designed well in the first place, should be plenty for most applications. If you really want rugged hardware, Google is much faster than Ask Slashdot, and it has a lower percentage of comments saying "you should've googled, idiot!".
Test screens for sunlight viewability. If beeps and boops are required, test speakers for audibility. Test batteries for operation over a wide temperature range, and make sure the manufacturer will replace any batteries that fail or degrade prematurely. Make sure dust and crud won't gum up the buttons or charging contacts.
Good luck getting wireless connectivity with no bits that stick out. Antennae don't last long, and they're hard to weatherproof.
The wireless part has me curious. How does your radio scheme deal with huge numbers of users in a small area? If you're trying to broadcast content to all of them at once, do the protocols support doing it with a single transmission? Does the application software?
Furthermore, for interactive applications where the user devices will be transmitting, does the system deal robustly and fairly with overload situations? Does efficiency plummet as utilization rises?
Given that the antennae integrated into handhelds are scarcely better than dummy loads, what sort of antenna gain do you need on the access point end to provide plenty of SNR? Consider the radio environment of the venue, given that metal surfaces will cause reflections and raise the noise floor. Also consider that reflections might make it difficult to divide the area into zones served by multiple access points, and devices might hop between points sporadically and quickly due to signal fluctuations.
Yeah, it was called the Buddy, and the old incarnation sucked. It was a PCI board which was essentially a video card plus a PS/2 keyboard and mouse controller. The video capabilities were terrible (sync rate limited to 60Hz, IIRC), and the second keyboard was prone to random resets and other problems.
The new incarnation of Buddy seems to address all those problems by using standard hardware. Buddy and BeTwin (they look like the same software) appear to work with any PCI video board that'll coexist with other video hardware, and since they use USB keyboards and mice (and audio, if you want), the proprietary controller problems should be gone too.
The new Buddy doesn't stop at 2 stations, either. It'll happily run up to 5, which might have a chance of using some of the absurd CPU power available in a modern PC. They have a trial version up for download, I might have to check my hardware compatibility and tinker with it later.
Well, the important voltage of such a lamp is mostly determined by its length, so drop-in replacement should be 'close enough' for most applications. Don't save the whole screen for the lamp, just find a good sturdy box and some soft padding to store them in. The lamps have mercury in them and should be returned for recycling if you give up your packrat habit, or when one fails.
That pile of miscellaneous screws has saved my ass more times than I can count. Sometimes I need a dozen screws but only have one of the right kind, but being able to take that one to the hardware store makes it worth it.
Same here! Houses are so close together and all the big trees are gone, any noise in the neighborhood comes through clear as a bell. I rely on the noise of my machine to cover it up, and it's very hard to relax without that. I had the hardest time sleeping during the power outage.
"He said 'The streets were dark tonight, it was like another century, with dim lamps and candles lighting up the icy trees and the clouds and a covered moon.' She said 'What kind of people make a city where you can't see the sky and you can't feel the ground?'" -- Dar Williams
Tie it in with a pair of 3d monitor-glasses, and have it overlay the past over the present. Walk through NYC and see the shimmering LCD ghost of the twin towers superimposed on the skyline.
You could always play the sounds of an internal combustion engine through the speakers.
On a side note, has anyone considered mounting a handful of small speaker/mic pairs around the case, and doing active noise cancellation?
The trouble is that the whole system is AC and must be synchronized. All the shafts of all the turbines connected to the power grid are literally turning in sync. Adding distributed generation is tricky. The easiest way is to generate DC (or generate AC and rectify to DC) and then use a "grid-tie" inverter to make phase-correct AC.
Small grid-tie inverters use the grid itself to provide sync, and just match their output to it. The trouble with doing this on a large scale is that the whole system chases its tail, with no master clock. If grid-tie inverters had sync inputs, it would be trivial to connect them to a real reference clock.
Additionally, subtle phase changes on the grid itself are used as a signalling protocol between utilities, indicating an excess of capacity or demand. This would need to be overcome before synchronized distributed generation would be practical on a large scale.
There's streaming support in Freenet now, and the network has been behaving nicely enough lately that it almost ought to work. More nodes = better performance, so go play with it.
The trouble with usenet and Freenet is that they're both very high latency. Great for your everyday muzak feed, but not ideal if you want to run a breaking-news station.
"every time you see a cop", this sounds a lot like CB radio.
Transmit messages between cars. "let me pass you, diesel dummy!" Also sounds a lot like CB.
Tell people what you're listening to. Ahh-yup. As if they'll care, but you can do that too.
Oh! And sending traffic reports by radio. That's surely a novel use for the technolo-- oh nevermind.
1945 called. They want their idea back.
and they seem to be doing pretty well. I went looking for weird NIC hardware and came across Imagestream. They make big routers with Linux at the core, on x86 hardware in industrial form factors. Definitely worth a look.
Also on the thread of interface cards, try Mikrotik. If you're doing wireless, the MiniPCI carrier boards will make your day.
Full disclosure: I'm not related to or affiliated with either of those companies in any way. I've never even bought anything from either of them. I just came across them while searching and thought they were bookmark-worthy.
Into the few-megahertz range, twisted pair wire works remarkably well. This is the stuff we're all familiar with as phone lines and cat-5. The number of twists per unit of length determines how resistant it is to interference, hence cat-5 is much more tightly twisted than cat-3. Each pair in a multipair cable is twisted a slightly different amount, to prevent inductive coupling and crosstalk between pairs.
The signal sent down a twisted pair is bipolar and "balanced", so that the two wires are carrying mirror opposite signals. There's an excellent explanation of this. T-1 signals ride twisted pair for several kilofeet between repeaters. The N-carrier system (low rate analog multiplex) also used twisted pair, but I don't know how far it would go between terminals.
Above a few megahertz, twisted pair gets unacceptably lossy and noisy. Higher speed signals are carried on coaxial cable, which we all know and love for its role in television wiring. The characteristic impedance of coax is determined by the ratio of the center conductor diameter and the distance to the inner surface of the outer conductor. Very early coaxial lines were made by suspending thin rods inside sections of copper pipe, by means of cardboard disc insulators. Soon a method of manufacturing flexible cable was developed, and has remained largely unchanged.
Signals carried on coax are "unbalanced", where the outer conductor is grounded and the inner conductor carries an AC wave. The need for the ground reference means that coax runs between buildings can become part of a ground loop, and cause all sorts of electrical problems. T-3 circuits use coax, but only for very short runs. (A T-3 that leaves a building does so as a DS-3 carried on fiber.) The L-carrier system, which multiplexed several N-carrier signals together, used thick coaxial lines for long-haul runs across the countryside.
As you approach the gigahertz range, coax also becomes too lossy, and hollow waveguide becomes the obvious choice. Waveguide can be rectangular, ovoid, or circular in cross-section, which effects the polarization of the signals carried in it. The inner dimensions influence loss and frequency range. Personally I'm not familiar with the buried waveguide system, but the TD and TH microwave systems used waveguide to connect the antennae with the terminal equipment.
Agreed one hundred percent! Fiber is useless for home applications. Just install lots of conduit, and put double-gang boxes on the ends so you'll have places to mount jacks for all the wires you'll eventually have in the conduit.
Power is a more important consideration than signal cabling. Run LOTS of branch circuits, ideally set up so that any given corner of a room has 2 circuits accessible to it. You don't want your UPS sharing a circuit with any outlets you plug a vacuum cleaner into, or the power quality log files will fill your drive.
Get a whole-house surge protector and pay the electrician to install it at the service entry point. While you're at it, get a second ground rod a few feet away from the first, and make sure the connections are done with exothermic welds. (crimps corrode in weather, and screw-clamps loosen with thermal fluctuations.)
Run a conduit to the front porch, for the doorbell and door camera / intercom. Run two to the back patio, for your outdoor speakers and network/phone jacks in various places. Drop 20-amp outlets all over the garage, and plan for a few up on the wall near shelving units, so you can park the battery charger/maintainer up there and plug it right in.
Bad idea. PVC fittings are made for liquid, not cable. They don't have a smooth internal surface, so things will snag up as you pull them.
Secondly, the bend radius of small PVC fittings is so tight that pulling any moderately stiff wire will be awkward past just 2 or 3 bends. Conduit can be bent with much gentler sweeps for easier pulling and less cable damage.
The stiff blue flexible plastic conduit is ideal for this, because it automatically sweeps corners as you install it.
Handy installation hint: You can buy "cable lubricant" which makes long pulls go smoothly. Liquid hand soap works just as well.
This sounds like the logical extension of the L-carrier systems. Before digital encoding was invented, radio techniques (frequency division multiplexing) were used to shift the frequency of each voice channel, and pack dozens of channels into a wide-band signal which could ride a twisted pair, or itself be muxed into a still wider signal, which was transmitted on coaxial cable.
I'm guessing that the megahertz-range signals on the coax were then muxed into gigahertz-range signals to be transmitted down the tubes. Fascinating.
Lots more details at long-lines.net for the curious.
WinCE devices aren't designed to run solo, they have to have a "partnership" with a "host" to do things like install software. I had an iPaq for a week before the cradle arrived, and I was trying to install the drivers for the CF wireless card. I had the Dual-CF sleeve for it, with the ma701 in the top slot and a 64-meg flash card in the bottom slot.
.exe file which then loads the drivers onto the portable via the sync cable. Whose assinine idea was this? (don't answer, I already know!)
:)
I figured I could download the driver package and load it onto the CF card at a desktop machine with a reader, then move the CF over to the ipaq, unpack it, and install it. No such luck. The driver ships as a win32
Not that I would've wanted to use an iPaq as my primary machine anyway, even with a foldup keyboard, the screen's too small and the viewing angle is terrible. However, there are a pair of WinCE devices I could imagine using as an everyday machine: The IBM Workpad z50, and the Vadem Clio (a.k.a. Sharp Tripad) are both laptop-style PDAs, with a clamshell hinge and a full size keyboard. They both have real PCMCIA slots, CF slots, hardware serial ports, and VGA screens. Both of them get approximately 8 hours on a battery, and the z50 is available with a double-capacity battery which honestly, realistically does get 16 hours.
Because of the WinCE codependency problem mentioned above, they're both unable to survive without occasional connection to a desktop for certain tasks. The easy solution is to ditch WinCE and run the hpcmips port of NetBSD, which boots on both of them. With a decent-sized CF card you can have a full development environment and not even need someone to cross-compile for you. That's definitely enough to be self-sufficient.
Full keyboard, harware serial port, and 8 hours on a charge. That's a recipe for "portable serial console" if I ever heard one. Oh yeah, it also happens to be a full-fledged NetBSD machine.
Yeah, I've switched to compact fluorescent lights in parts of the house for efficiency reasons, but I try to pair them with incandescent bulbs in fixtures near reading areas. I figure, half the efficiency benefit is better than none, and the reduction in flicker is essential.
I try to let natural light into the room when I'm looking at the monitor, but I'm careful to balance the contrast and avoid glare. If it's nighttime and I need light other than the monitor, I use an incandescent on a resistive dimmer (never a triac dimmer, they flicker too).
The most important thing has got to be focus though. A friend of mine has a degenerative disease where her eyes can only change focus slowly, over the span of several seconds or minutes. Looking from the road to the instrument panel and back to the road is impossible for her, so she has to read the gauges with her peripheral vision. (Analog gauges are easy, because you can tell the positions of the needles. Digital readouts are next to impossible.) When she uses a computer, she has to turn away and focus on the room for a moment before standing up to go get something.
I'm determined not to let this happen to me. I use my music collection to help -- at the beginning of each track, I close my eyes for a few seconds, take a deep breath, open my eyes, and look at something across the room. It's relaxing in terms of general stress, and it gives my eyes a chance to change focus distance.
Think about it -- our eyes weren't designed to stand still for long periods. In a survival situation, you're focusing near and far constantly, and your eyes are all over the place. Try to give them a little workout now and then.
Leave them running for 2 days and see if the air conditioning system can keep up. :)
If the office includes backup power, make sure the air conditioners are part of the protected load. If commercial power fails and the generator kicks on, your server room will keep running but get very hot, very fast without proper cooling.
IIRC, Duke Nukem 3D was the first FPS to utilize stereo sound hardware for positional audio. It was the first one I played, anyway.
It was possible, though not easy, to make it past the first level with the monitor turned off. Since rockets had a finite travel speed and made noise on impact, they could be used like "sonar" to determine distance to walls.
When a tree was hit, it would burn and make a crackling sound. When a fire hydrant was hit, it would burst open and make a water-spraying sound. All of these sounds would continue for several moments, during which the player could use them for position feedback while moving through the map.
As a decided retro-geek with a fondness for old hardware, I'm glad there's finally a new game that doesn't require a 3-d graphics card!