I've stated my full opinion of this in an above post, but this warrants a response.
Yes, if something you do makes you happy, of course similar neurotransmitters and conditions will exist in the brain. But using that as a springboard into addiction==compulsive behavior is ludicrous.
The reason why people like myself fight so strenuously against this labeling is that it promotes a false sense of security in choosing treatment for someone. If all people "addicted" to a behavior (almost invariably actually compulsive behavior unless a drug is involved) really are addicted, they need chemical support to help them deal with the pain of withdrawl.
Giving someone a drug to help them quit Everquest is dodging the problem. The problem is that they don't want to be responsible for managing their time in a constructive manner. People like that need mental help, not chemical.
But what if the chemical helps them with their immediate problem? Big deal, as long as it never recurs. Giving someone drugs to counteract depression is useful only if accompanied with therapy to resolve the CAUSES of depression (assuming they are not genuinely abnormal in their brain chemistry). Most cases of diagnosed depression are caused by enviromental stimulus, such as stress, social problems, etc. This are causes that need to be addressed, not medicated to numbness.
No, you are not an addict. Your brain is capable of functioning perfectly normally if you are sufficiently distracted from the behavior that is compulsive. Someone who quits taking heroin feels REALLY BAD. Even comparing yourself to that shows an utter lack of understanding of the nature of addiction.
Addiction is not a mere want. If you were tempted with something juicy enough, you'd forget immediately about your Internet access. Someone addicted to heroin has no such luck. Underinformed pseudo-psychiatrists call these behavioral things "addiction" for culture bonus points... to make a name for themselves. Real scientists *LAUGH* at the idea of actually considering video games, TV, or the Internet to be addictive. The processes are NOT analagous. You like them both (hence the presence of certain neurotransmitters), but Internet withdrawl leaves you irritable because you don't have anything else to do. The symptoms you describe are exactly consistent with plain old boredom.
I'd suggest that you (or anyone who feels that they are "addicted" to, or dependant on the Internet) start reading some books, or find other hobbies that you can use in lieu of Internet access to add some other kind of value to your life.
Well, I'll give you that, but that would make this a slashback, or something like that. The headline "1024-bit RSA keys In Danger Of Compromise?" is old news.
Bcast2k in the word of Monty Python is "... not quite dead yet!"
If you want to check out the latest "backups," of the codebase, look for Cinelerra on sourceforge.
It is still being developed, but you can expect basically no support (compiling it can be problematic).
Bcast2k is an excellent program. I do feel it is unfortunate that the authors decided to withdraw it from public view.
Re:GNUStep? (GNUstep is not WindowMaker)
on
Penguin2Apple
·
· Score: 2
When I originally posted this, I was thinking that GNUStep would have aggregately fit the bill of everything that he wanted. However, I ended up just talking about WMaker, as it really has the key interface issues that he discussed. GNUStep would be the "whole environment" with the OpenStep compatibility, etc.
Re:The Real Problem With Cheap Laptops
on
Low-end Laptops?
·
· Score: 5, Interesting
If you can't get the batteries cheap, refurbish the ones you have.
Odds are that if you crack open the batteries, inside will be an array of standard or nearly standard NiMH or NiCad cells. Even if you have ones that look like elongated AA's, you can replace them with AA's and the machine should still work properly (charge, boot, etc), but the battery life will likely not be as long.
Note: do NOT use NiMH batteries in a NiCad notebook unless the MFG supported those. The charging circuit will probably not properly recognize the full charge (delta V method).
NiMH cells go for $2 a piece, and you should be able to get a working first try for less than $20.
As an ardent WindowMaker user (I couldn't stand the bloat of other desktops), I would say that half of this guy's problems would be solved if he were just to switch to WindowMaker, and learn about the middle mouse button. WindowMaker has a dock, you can collapse app's onto it, launch them from it, and even have neat dock apps. Adding apps to the dock is as simple as dragging their application icon (which is created for any application not already "docked") onto the dock.
I'll admit that the OSX dock is more graphically pleasing. And a little more flexible. But the big points are already there.
I also own an iBook (old clamshell), and wouldn't consider running OSX on it. OSX requires too much of your CPU and memory. WindowMaker under Linux runs as smooth on my iBook as it does on my Athlon.
You are using it for business. If you want the kind of service you'd expect for business purposes, you should pay for it.
I'm sure this is going against the grain of some here, who'd say that we should have perfect service on our cheap lines, or that you shouldn't have to pay additional for better service (customer service, not bandwidth). That is ridiculous. If everyone were to be prioritized the same, costs would increase (need more techs to handle faster response times) and your price would increase proportionately.
Shit happens, wear a helmet.
Re:Problem with the Netwinder
on
Netwinder is Back
·
· Score: 3, Informative
Power. The Netwinder uses a miniscule amount of power compared to the Athlon.
Foot print: The Netwinder takes up 1/4 the space of the smallest 1800 XP system
Already Built: Don't need a linux geek to set it up for you. Just use their basic web administration, and you don't even have to care that it is running linux.
I could list more, but these are the key points. I have played with one, and I wish I had one (even though it would replace my much more powerful 233MHz PII firewall, I just like the machine and how little power it drops/how small it is).
Nothing above 1GHz is needed right now for most users. However, history has shown us that every time we say this (processors are faster than they need to be, blah blah blah), someone comes out with killer apps that drive the need for faster machines. You know by the time Warcraft 4 (or its equivalent) comes out, 1GHz will be painful to run it on.
Until we have machines that can perform (near) perfect speech control/dictation, face recognition (in real time, reading expressions), and can make realistic holograms (ala STNG Holodeck), I will not even begin to believe that CPU's have come far enough.
As for you libertarians who seem to think liberals are the greatest threat to freedom, who are the ones currently taking away US freedoms in the old excuse of national security? It ain't the liberals anyway.
Yeah... it is both! Psychotic restriction of freedom is not a liberal or conservative tenant. It is a tenant of power.
Coming from someone who has done this in a professional environment (opened drives), I would say you are lucky, or did this to a relatively old drive. I have seen the effects that small amounts of dust can have on a drive.
Older drives have had fly heights higher than a particle size, meaning little effect from the presence of trace amounts of dust.
Current drives (last 4 years or so) have VERY low fly heights. They are designed to maintain an altitude over the drive platter that is generally smaller than a visible dust particle. A dust particle that becomes lodged like this on the disk head will draw cyclical patterns of dead/error sectors on the disk (yes, I have seen this, many times). In many cases, you will not have catastrophic drive failure, but you *will* have damaged sectors.
Even if the platter is "tough" enough to take this, the contamination is likely to accelerate corrosion (something that a disk head has no tolerance for), you risk damaging the head from particle impacts (at 7200+ RPM, that particle sticking to the drive surface can do some considerable damage).
Dust contamination may take weeks or months to develop problems (there is a small whirlwind going on in your HD, it is just a matter of time if there is any free dust in there). Taking a drive apart, then putting it back together, and watching it spin up would be an extremely naive method of calling it functional. If you are lucky, you didn't drop too much dust in there, and the filters in the HD would pick up most of the particles.
Also, some of the damage is not permanent. Reformatting the drive, or rewriting the sectors will clear the damage (really partial corruption). The heat from the particle being dragged across the surface of the disk may flip a few bits here and there.
Now, TVs, VCRs and power supplies are ALL DESIGNED TO BE OPEN to the air. You wouldn't crack open the TV tube, connect your Hoover to it and plug the hole with Fix-a-Flat, would you? The only part in a VCR that is really a problem to have dirty is the head (which is already exposed daily). That can be remedied with a solvent. You'd be an idiot to use any solvent on a modern HD surface, you'd be sure to crash the head then (film residue, most platters are coated).
As for the conspiracy theory regarding data recovery: it is your risk to take. If you can't afford the recovery fee, find as clean of a place as you can find to open the drive, and DO NOT TOUCH THE PLATTERS WITH YOUR FINGERS. Only touch them at the edges with clean plastic. Read the data off the drive, and consider the "new" drive dead. It will not live forever.
Also, this is only useful if the head dies in your drive. If the platter is scratched, it will destroy the head of the new drive as well (if it works at all).
Run mozilla with esddsp or artsdsp. It will redirect the sound requests to your favorite sound server. You have to be using esd or artsd for everything (xmms, xine, etc), but it will eliminate your problem.
e.g. change the execution string in your desktop icon from
Unfortunately, the change you ask for would destroy what Linux is: a Unix-a-like.
Linux is popular because it closely resembles the powerful Unix platform, not because it is user friendly. This is a truly unfortunate problem. The administration of a Linux machine can't really get much easier than it is already without throwing away the complicated Unix under pinning. And many of us would resist any attempt to change it.
The thing I loved about my good ol' Mac was that I could upgrade libraries by hand. On Linux, without an installation script, I am very nervous as there are typically several files per library, sometimes in many locations.
However, the flexibility I'm afforded for automation on Linux is utterly phenomenal compared to old Mac OS's (I'm unfortunately not that exposed to OSX, which may change my mind).
There are a number of problems with space colonization, but one of the killers is gravity.
But what about oxygen, food, etc you may ask?
Gravity takes care of the containment for you. Your gravity isn't going to spring a leak and start venting air away (assuming it is great enough to hold it in place at a proper pressure).
Further, without physical stress (ie: weight) bones/muscles deteriorate requiring more maintanence to keep the human colonists functional.
Food: A planet provides a MASSIVE surface area with which to grow crops (even if the soil is unarable, large hydroponics systems could deal with it). A self sustaining station would require massive amounts of materials to make a farm large enough to feed its inhabitants. Let alone if a small asteroid came along and broke the ceiling out of your greenhouse...
Assuming you aren't eating nutritional pills by then.
This is just short sighted. Application crashes are not ALWAYS the app developer's fault. In many circumstances, they are. But the app developer may be using libraries that the user can upgrade/interchange, that the app developer has nothing to do with implementing (such as glib). If there is a bug in the library they call which causes an obscure crash or the user changes the library to an unstable version, you can hardly blame the app developer.
Secondly, a surprising number of "crashes" aren't application crashes - they are hardware issues (conflicts, heat problems, failure, etc), driver issues (some call they make to a graphics routine tickles the video driver in a bad way), and OS problems (deadlock, Microsoft, bugs). What most people interpret as a crash (especially in OS's where a single application can easily take down the entire system, like Win9x), looks identical in these circumstances; the computer ceases to respond. Even experienced computer users can have a difficult time assigning blame.
And to be on topic: when selecting a documentation system, look for one that avoids redundant work. With lesser tools, you will have to update documentation in your code comments as well as in your design documents. Maintainence is everything here, and if you have to double commit all alterations, you will become frustrated very quickly. Hence why a source code parsing tool [like Doxygen] is probably the better choice, regardless of the features of other programs.
Err... what is the difference between impact and deceleration? WRT the physics of the shock? Deceleration occurs upon impact.... One is a cause, the other an effect....
Yes, the surface it is dropped onto is EVERYTHING when it has to do with the ac[de]celeration. A hard surface will tend to give a very sharp, high intensity force curve. A soft surface will give a long duration (relatively), with a lower intensity.
However, shock ratings can be misleading. The shape of the pulse is also extremely important. Different shapes mean different effective constituent frequencies. 1500G != 1500G in some cases because the impact shape may highlight resonance problems (especially when the head is not parked).
Think of it like jumping on a trampoline. When you shove as hard as you can, as fast as you can, you aren't in sync with the natural vibration of the trampoline. But if you jump a little slower bringing your stride into sync with it, you can jump higher and higher (to a point).
Anyway, drop from a desk onto carpet... all the shock tests I did were synthetics (ie: not drops onto carpet). So it is hard for me to talk precisely about this. However, I find it subjectively unlikely that a 3ft drop onto carpet could possibly create a 1500G 2ms force.
Warning, annoying obfuscation of details of my employment due to NDA...
As someone who may have done shock tests for a "small" company, I can tell you that is a bit misleading.
The shock rating typically is determined by placing a drive on an apparatus that drops the drive from a height onto a platform with a controlled shock response (how much it gives, the exact spring constant to give you a certain duration of a pulse). This distance IIRC (say for 800G at 2ms pulse width), is approximately 7ft. This pulse is idealized as a sinusoid with a maximum around the target shock (in this example, 800G).
1500G would be more, but not like firing it out of a gun at a brick. 15000m/s/s is a quick change in velocity. But a little mathematics would show that it isn't inconceivable for a quick stop from a relatively low speed.
I'm sure someone doing some karma enterprising could find some links to companies that develop industrial packaging testing equipment.
Why do you need multiple servers in one room?
on
Home Server Rooms?
·
· Score: 1
In my house, I have a single server. It shares my cable connection to the rest of the house, serves Samba shares, and until I got sick of walking that far, and realized it was ridiculous, was the print server as well. I can't believe that anyone setting up a *HOME* network would have any valid reason to have that much stuff running in one place (now, you can argue about your kitchen, living room, bathroom, deck, rec, and garage CPU's. Honestly, why the hell would you need all that much for just a house?
Anyway, my server is in a closet in a bedroom used for storage. It sits in that closet [with the door closed, GASP!], and has never crashed. This setup has been running continuously (other than power outages and upgrades) for well over a year now, and it never gets significantly warmer in the closet than the rest of the room/house. I really don't think there are any real concerns for you unless you are legally insane, and actually need to run seven computers to serve a house.
Now, if you are using your house for a server farm for your own employment, that is another issue. In which case, I'd retract the insanity and question your intellect about marketing (not a lot of people want to trust in someone who operates their ISP and/or web hosting out of a house. You can talk all you want about bottom line, but they are going to question your seriousness, and finacial stability. I know I wouldn't trust someone who didn't at least have a business presence that I could walk into to yell at someone when something wasn't getting done.
Raid doesn't address one of the other common modes of data loss: catastrophic failure/natural disaster. If your raid gets set on fire, shocked with 1MV lightning, or doused in water, it will probably be completely gone. Offline backups (such as removable, or offsite backups) are much more reliable (it isn't likely your house is going to burn down at the same time your bank is subsumed by a tidal wave).
Another thing that some of us are looking towards is finding a trustworthy friend to share capacity with. If you each buy the extra hard drive (or have space to spare), and rsync nightly with each other, you can get reasonable coverage, and offsite backup. Just pick one reasonably geographically far from you so your data doesn't get sucked up in the same tornado.
Or if you don't trust the "friend", use an encrypted filesystem, or crypt the files first.
No, what is really dangerous about the P.R. toy is
Power Rangers Time Force, "Virtual Reality World," Ban Dai,
Recommended Age: 4+
Description: A virtual reality headpiece, gloves and a leg sensor that allow children to re-enact scenarios from the Time Force Power Ranger television program
Do you want your five year old running around the house punching and kicking essentially blind?
I can't be sure that this is how it works, as I can't find the toy on Bandai's site or Toys-R-Us. But it certainly seems that way from the description....
Maybe I'm replying a bit late, but if you think about the magnetic fields around the phone wires for a minute, you'll probably understand why the voltage is largely irrelevant here. Current causes magnetic fields. Since there will be an equivalent current flowing back the other direction on a wire that is twisted in line with it, there will be only a small amount of actual inducing field.
This is one of the reasons why coax cables are so immune to causing interference. In an ideal cable (by ideal properties, not ideal as in an abstract sense), there is no induced magnetic field outside the conductor.
In non-twisted cable, yes, there is a significant chance something will go wrong.
And for the non-theoretical, I have, in my house, the ethernet line that supplies the connection to my linux server combined with the phone line to the room it is in. If there was a problem, I'd notice it in a hurry (if my shoutcast streams broke whenever someone called, or I had other problems when on the phone. For a while, it was the modem line while on the internet as well. No problems then either).
This is not to discount evidence others have cited as bunk, but chances are, telephones were not the sole problem in their environment.
Slashdot Mirror
I've stated my full opinion of this in an above post, but this warrants a response.
Yes, if something you do makes you happy, of course similar neurotransmitters and conditions will exist in the brain. But using that as a springboard into addiction==compulsive behavior is ludicrous.
The reason why people like myself fight so strenuously against this labeling is that it promotes a false sense of security in choosing treatment for someone. If all people "addicted" to a behavior (almost invariably actually compulsive behavior unless a drug is involved) really are addicted, they need chemical support to help them deal with the pain of withdrawl.
Giving someone a drug to help them quit Everquest is dodging the problem. The problem is that they don't want to be responsible for managing their time in a constructive manner. People like that need mental help, not chemical.
But what if the chemical helps them with their immediate problem? Big deal, as long as it never recurs. Giving someone drugs to counteract depression is useful only if accompanied with therapy to resolve the CAUSES of depression (assuming they are not genuinely abnormal in their brain chemistry). Most cases of diagnosed depression are caused by enviromental stimulus, such as stress, social problems, etc. This are causes that need to be addressed, not medicated to numbness.
No, you are not an addict. Your brain is capable of functioning perfectly normally if you are sufficiently distracted from the behavior that is compulsive. Someone who quits taking heroin feels REALLY BAD. Even comparing yourself to that shows an utter lack of understanding of the nature of addiction.
Addiction is not a mere want. If you were tempted with something juicy enough, you'd forget immediately about your Internet access. Someone addicted to heroin has no such luck. Underinformed pseudo-psychiatrists call these behavioral things "addiction" for culture bonus points... to make a name for themselves. Real scientists *LAUGH* at the idea of actually considering video games, TV, or the Internet to be addictive. The processes are NOT analagous. You like them both (hence the presence of certain neurotransmitters), but Internet withdrawl leaves you irritable because you don't have anything else to do. The symptoms you describe are exactly consistent with plain old boredom.
I'd suggest that you (or anyone who feels that they are "addicted" to, or dependant on the Internet) start reading some books, or find other hobbies that you can use in lieu of Internet access to add some other kind of value to your life.
Well, I'll give you that, but that would make this a slashback, or something like that. The headline "1024-bit RSA keys In Danger Of Compromise?" is old news.
Here?
Bcast2k in the word of Monty Python is "... not quite dead yet!"
If you want to check out the latest "backups," of the codebase, look for Cinelerra on sourceforge.
It is still being developed, but you can expect basically no support (compiling it can be problematic).
Bcast2k is an excellent program. I do feel it is unfortunate that the authors decided to withdraw it from public view.
When I originally posted this, I was thinking that GNUStep would have aggregately fit the bill of everything that he wanted. However, I ended up just talking about WMaker, as it really has the key interface issues that he discussed. GNUStep would be the "whole environment" with the OpenStep compatibility, etc.
If you can't get the batteries cheap, refurbish the ones you have.
Odds are that if you crack open the batteries, inside will be an array of standard or nearly standard NiMH or NiCad cells. Even if you have ones that look like elongated AA's, you can replace them with AA's and the machine should still work properly (charge, boot, etc), but the battery life will likely not be as long.
Note: do NOT use NiMH batteries in a NiCad notebook unless the MFG supported those. The charging circuit will probably not properly recognize the full charge (delta V method).
NiMH cells go for $2 a piece, and you should be able to get a working first try for less than $20.
As an ardent WindowMaker user (I couldn't stand the bloat of other desktops), I would say that half of this guy's problems would be solved if he were just to switch to WindowMaker, and learn about the middle mouse button. WindowMaker has a dock, you can collapse app's onto it, launch them from it, and even have neat dock apps. Adding apps to the dock is as simple as dragging their application icon (which is created for any application not already "docked") onto the dock.
I'll admit that the OSX dock is more graphically pleasing. And a little more flexible. But the big points are already there.
I also own an iBook (old clamshell), and wouldn't consider running OSX on it. OSX requires too much of your CPU and memory. WindowMaker under Linux runs as smooth on my iBook as it does on my Athlon.
You are using it for business. If you want the kind of service you'd expect for business purposes, you should pay for it.
I'm sure this is going against the grain of some here, who'd say that we should have perfect service on our cheap lines, or that you shouldn't have to pay additional for better service (customer service, not bandwidth). That is ridiculous. If everyone were to be prioritized the same, costs would increase (need more techs to handle faster response times) and your price would increase proportionately.
Shit happens, wear a helmet.
Power. The Netwinder uses a miniscule amount of power compared to the Athlon.
Foot print: The Netwinder takes up 1/4 the space of the smallest 1800 XP system
Already Built: Don't need a linux geek to set it up for you. Just use their basic web administration, and you don't even have to care that it is running linux.
I could list more, but these are the key points. I have played with one, and I wish I had one (even though it would replace my much more powerful 233MHz PII firewall, I just like the machine and how little power it drops/how small it is).
Nothing above 1GHz is needed right now for most users. However, history has shown us that every time we say this (processors are faster than they need to be, blah blah blah), someone comes out with killer apps that drive the need for faster machines. You know by the time Warcraft 4 (or its equivalent) comes out, 1GHz will be painful to run it on.
Until we have machines that can perform (near) perfect speech control/dictation, face recognition (in real time, reading expressions), and can make realistic holograms (ala STNG Holodeck), I will not even begin to believe that CPU's have come far enough.
In the meantime, AMD rides the gravy train.
So, would running windows on your cell phone increase the probability of having a fatal car crash while dialing?
As for you libertarians who seem to think liberals are the greatest threat to freedom, who are the ones currently taking away US freedoms in the old excuse of national security? It ain't the liberals anyway.
Yeah... it is both! Psychotic restriction of freedom is not a liberal or conservative tenant. It is a tenant of power.
Coming from someone who has done this in a professional environment (opened drives), I would say you are lucky, or did this to a relatively old drive. I have seen the effects that small amounts of dust can have on a drive.
Older drives have had fly heights higher than a particle size, meaning little effect from the presence of trace amounts of dust.
Current drives (last 4 years or so) have VERY low fly heights. They are designed to maintain an altitude over the drive platter that is generally smaller than a visible dust particle. A dust particle that becomes lodged like this on the disk head will draw cyclical patterns of dead/error sectors on the disk (yes, I have seen this, many times). In many cases, you will not have catastrophic drive failure, but you *will* have damaged sectors.
Even if the platter is "tough" enough to take this, the contamination is likely to accelerate corrosion (something that a disk head has no tolerance for), you risk damaging the head from particle impacts (at 7200+ RPM, that particle sticking to the drive surface can do some considerable damage).
Dust contamination may take weeks or months to develop problems (there is a small whirlwind going on in your HD, it is just a matter of time if there is any free dust in there). Taking a drive apart, then putting it back together, and watching it spin up would be an extremely naive method of calling it functional. If you are lucky, you didn't drop too much dust in there, and the filters in the HD would pick up most of the particles.
Also, some of the damage is not permanent. Reformatting the drive, or rewriting the sectors will clear the damage (really partial corruption). The heat from the particle being dragged across the surface of the disk may flip a few bits here and there.
Now, TVs, VCRs and power supplies are ALL DESIGNED TO BE OPEN to the air. You wouldn't crack open the TV tube, connect your Hoover to it and plug the hole with Fix-a-Flat, would you? The only part in a VCR that is really a problem to have dirty is the head (which is already exposed daily). That can be remedied with a solvent. You'd be an idiot to use any solvent on a modern HD surface, you'd be sure to crash the head then (film residue, most platters are coated).
As for the conspiracy theory regarding data recovery: it is your risk to take. If you can't afford the recovery fee, find as clean of a place as you can find to open the drive, and DO NOT TOUCH THE PLATTERS WITH YOUR FINGERS. Only touch them at the edges with clean plastic. Read the data off the drive, and consider the "new" drive dead. It will not live forever.
Also, this is only useful if the head dies in your drive. If the platter is scratched, it will destroy the head of the new drive as well (if it works at all).
Maybe a little late for public notice, but...
If another program takes exclusive access, flash will still freeze, even if it is using ESD.
The point was to avoid having anything but ESD actually talking to the sound device.
Run mozilla with esddsp or artsdsp. It will redirect the sound requests to your favorite sound server. You have to be using esd or artsd for everything (xmms, xine, etc), but it will eliminate your problem.
e.g. change the execution string in your desktop icon from
mozilla (or mozilla-bin)
to
esddsp mozilla
Unfortunately, the change you ask for would destroy what Linux is: a Unix-a-like.
Linux is popular because it closely resembles the powerful Unix platform, not because it is user friendly. This is a truly unfortunate problem. The administration of a Linux machine can't really get much easier than it is already without throwing away the complicated Unix under pinning. And many of us would resist any attempt to change it.
The thing I loved about my good ol' Mac was that I could upgrade libraries by hand. On Linux, without an installation script, I am very nervous as there are typically several files per library, sometimes in many locations.
However, the flexibility I'm afforded for automation on Linux is utterly phenomenal compared to old Mac OS's (I'm unfortunately not that exposed to OSX, which may change my mind).
Gravity.
There are a number of problems with space colonization, but one of the killers is gravity.
But what about oxygen, food, etc you may ask?
Gravity takes care of the containment for you. Your gravity isn't going to spring a leak and start venting air away (assuming it is great enough to hold it in place at a proper pressure).
Further, without physical stress (ie: weight) bones/muscles deteriorate requiring more maintanence to keep the human colonists functional.
Food: A planet provides a MASSIVE surface area with which to grow crops (even if the soil is unarable, large hydroponics systems could deal with it). A self sustaining station would require massive amounts of materials to make a farm large enough to feed its inhabitants. Let alone if a small asteroid came along and broke the ceiling out of your greenhouse...
Assuming you aren't eating nutritional pills by then.
This is just short sighted. Application crashes are not ALWAYS the app developer's fault. In many circumstances, they are. But the app developer may be using libraries that the user can upgrade/interchange, that the app developer has nothing to do with implementing (such as glib). If there is a bug in the library they call which causes an obscure crash or the user changes the library to an unstable version, you can hardly blame the app developer.
Secondly, a surprising number of "crashes" aren't application crashes - they are hardware issues (conflicts, heat problems, failure, etc), driver issues (some call they make to a graphics routine tickles the video driver in a bad way), and OS problems (deadlock, Microsoft, bugs). What most people interpret as a crash (especially in OS's where a single application can easily take down the entire system, like Win9x), looks identical in these circumstances; the computer ceases to respond. Even experienced computer users can have a difficult time assigning blame.
And to be on topic: when selecting a documentation system, look for one that avoids redundant work. With lesser tools, you will have to update documentation in your code comments as well as in your design documents. Maintainence is everything here, and if you have to double commit all alterations, you will become frustrated very quickly. Hence why a source code parsing tool [like Doxygen] is probably the better choice, regardless of the features of other programs.
Err... what is the difference between impact and deceleration? WRT the physics of the shock? Deceleration occurs upon impact.... One is a cause, the other an effect....
Yes, the surface it is dropped onto is EVERYTHING when it has to do with the ac[de]celeration. A hard surface will tend to give a very sharp, high intensity force curve. A soft surface will give a long duration (relatively), with a lower intensity.
However, shock ratings can be misleading. The shape of the pulse is also extremely important. Different shapes mean different effective constituent frequencies. 1500G != 1500G in some cases because the impact shape may highlight resonance problems (especially when the head is not parked).
Think of it like jumping on a trampoline. When you shove as hard as you can, as fast as you can, you aren't in sync with the natural vibration of the trampoline. But if you jump a little slower bringing your stride into sync with it, you can jump higher and higher (to a point).
Anyway, drop from a desk onto carpet... all the shock tests I did were synthetics (ie: not drops onto carpet). So it is hard for me to talk precisely about this. However, I find it subjectively unlikely that a 3ft drop onto carpet could possibly create a 1500G 2ms force.
Warning, annoying obfuscation of details of my employment due to NDA...
As someone who may have done shock tests for a "small" company, I can tell you that is a bit misleading.
The shock rating typically is determined by placing a drive on an apparatus that drops the drive from a height onto a platform with a controlled shock response (how much it gives, the exact spring constant to give you a certain duration of a pulse). This distance IIRC (say for 800G at 2ms pulse width), is approximately 7ft. This pulse is idealized as a sinusoid with a maximum around the target shock (in this example, 800G).
1500G would be more, but not like firing it out of a gun at a brick. 15000m/s/s is a quick change in velocity. But a little mathematics would show that it isn't inconceivable for a quick stop from a relatively low speed.
I'm sure someone doing some karma enterprising could find some links to companies that develop industrial packaging testing equipment.
In my house, I have a single server. It shares my cable connection to the rest of the house, serves Samba shares, and until I got sick of walking that far, and realized it was ridiculous, was the print server as well. I can't believe that anyone setting up a *HOME* network would have any valid reason to have that much stuff running in one place (now, you can argue about your kitchen, living room, bathroom, deck, rec, and garage CPU's. Honestly, why the hell would you need all that much for just a house?
Anyway, my server is in a closet in a bedroom used for storage. It sits in that closet [with the door closed, GASP!], and has never crashed. This setup has been running continuously (other than power outages and upgrades) for well over a year now, and it never gets significantly warmer in the closet than the rest of the room/house. I really don't think there are any real concerns for you unless you are legally insane, and actually need to run seven computers to serve a house.
Now, if you are using your house for a server farm for your own employment, that is another issue. In which case, I'd retract the insanity and question your intellect about marketing (not a lot of people want to trust in someone who operates their ISP and/or web hosting out of a house. You can talk all you want about bottom line, but they are going to question your seriousness, and finacial stability. I know I wouldn't trust someone who didn't at least have a business presence that I could walk into to yell at someone when something wasn't getting done.
Raid doesn't address one of the other common modes of data loss: catastrophic failure/natural disaster. If your raid gets set on fire, shocked with 1MV lightning, or doused in water, it will probably be completely gone. Offline backups (such as removable, or offsite backups) are much more reliable (it isn't likely your house is going to burn down at the same time your bank is subsumed by a tidal wave).
Another thing that some of us are looking towards is finding a trustworthy friend to share capacity with. If you each buy the extra hard drive (or have space to spare), and rsync nightly with each other, you can get reasonable coverage, and offsite backup. Just pick one reasonably geographically far from you so your data doesn't get sucked up in the same tornado.
Or if you don't trust the "friend", use an encrypted filesystem, or crypt the files first.
No, what is really dangerous about the P.R. toy is
Power Rangers Time Force, "Virtual Reality World," Ban Dai, Recommended Age: 4+ Description: A virtual reality headpiece, gloves and a leg sensor that allow children to re-enact scenarios from the Time Force Power Ranger television program
Do you want your five year old running around the house punching and kicking essentially blind? I can't be sure that this is how it works, as I can't find the toy on Bandai's site or Toys-R-Us. But it certainly seems that way from the description....
Maybe I'm replying a bit late, but if you think about the magnetic fields around the phone wires for a minute, you'll probably understand why the voltage is largely irrelevant here. Current causes magnetic fields. Since there will be an equivalent current flowing back the other direction on a wire that is twisted in line with it, there will be only a small amount of actual inducing field.
This is one of the reasons why coax cables are so immune to causing interference. In an ideal cable (by ideal properties, not ideal as in an abstract sense), there is no induced magnetic field outside the conductor.
In non-twisted cable, yes, there is a significant chance something will go wrong.
And for the non-theoretical, I have, in my house, the ethernet line that supplies the connection to my linux server combined with the phone line to the room it is in. If there was a problem, I'd notice it in a hurry (if my shoutcast streams broke whenever someone called, or I had other problems when on the phone. For a while, it was the modem line while on the internet as well. No problems then either).
This is not to discount evidence others have cited as bunk, but chances are, telephones were not the sole problem in their environment.