It's pretty obvious what happened, the Linux kernel has problems with the hardware in question. It's not like that's a unique situation. Even if Linux works with 95% of machines one person in 4000 will have consecutive failures on two different machines. And they'll sure as hell make some noise about it when it happens.
It's not the same situation. "Linux" (as in the GNU/Linux/KDE/Mozilla/blah/blah/blah system) as a whole doesn't have a boss who can tell people to fix something. Bill Gates can tell the Windows team to streeamline some process and they'll (try to) do it, because if they say no they're liable to get canned. There is nobody in overall charge of the "Linux" OS. Even they myrid components which make up a Linux system rarely have more than a leader who can make suggestsions and accept or reject patches. I think a large part of Ubuntu's success is that there is somebody in charge of a bunch of programmers who will tweak things on command.
Why would he not buy it? By buying through normal channels it he gets to experience what his users experience. It's not like he, or Microsoft, can't afford to burn a few bucks on usability testing.
Linux distros, in common with Apple, can bundle whatever they like because they don't have a monopoly on the desktop operating system market. It would be interesting to see what would happen if Ubuntu or another Linux distriubtion did become dominant. Do you think commerical vendors (they'd all be selling stuff for Linux by that point) wouldn't press for an anti-trust suit? Canonical is a for-profit corporation after all. Dominance on of the desktop would help their business, even if only 10% of users ever paid them anything.
That's dangerously bad advice. In addition to the headline power rating, PSUs are rated per rail (5V, 12V etc.). It's extremely unlikely that your computer would draw power from the various supply lines in exactly the proportions the supply can provide. Most notably, older systems drew CPU power from the 5V rail. Boards for the P4 and other more modern boards use the 12V rail for CPU power. (In both cases there's a switching regulator which coverts this to the CPU's actual supply voltage.) PSUs, at least those sold to home users, are designed to handle either situation, so you may approach the limits for one rail or the other, but rarely both at the same time. Unless you're an OEM who can control the entire design, you are very unlikely to be able to run a system which requires 200W from a "220W" PSU - you'll be over the rating one one rail or another.
All PSUs are not created equal. Those gold 550W PSUs which were popular a few years ago provided less power on the 5V line than the 300W Antec I went for in the end. If you were overclocking an Athlon (as I was) that difference was make-or-break.
Only if you use the peak-to-peak voltage. For AC, the voltage everybody quotes is the RMS voltage. For a sine wave, the RMS value is 1/(root 2), about 0.707, times the peak value. With the RMS voltage it's just P=VI, which is why it's the RMS value which is quoted. The reason for the VA rating is more to do with power factors - how close to a resistive load the actual load is. A very basic switching supply will only draw any current at all close to the peak voltage and will draw nothing the rest of the time. If you're designing generators, inverters and transformers it matters a lot if your 5A average load is spread out over the cycle or if the load is zero 99% of the time and 500A 1% of the time.
So does your computer, powersupplies get less efficeint the warmer the room is. So while your useing only 200 watts, at 70 degrees, at 85 degrees, it's probably past 250.
Resistance does increase with temperature and a thermally controlled fan will spin faster and draw more current. But enough for a 25% rise in consumption from a 15 degree (in unspecified units, I guess you mean Fahrenheit) temperature rise? That's seems like a hell of a lot for a fairly modest rise in temperature.
For an 80% efficient power supply, an increase of 25% overall consumption is more than double the power loss. The reality is very complex, but we can pick out a few relevant numbers to get a feel for the magnitudes involved. Empirical testing would be easier than an analysis, but here's some food for thought:
For copper, the resistance rises by about 0.4% per degree Celsius rise. Your roughly 7 Celsius rise would increase it by a whopping 2.8%. You'll have melted the insulation well before even a 50% rise in the resistance of your copper wire.
If you look inside a power supply, you'll see a big fat heatsink. Attached to that are rectifiers and switches - diodes and FETs. That's where a big proportion of your power supply's inefficiency comes from. Looking at the first power FET datasheet I have to hand (for a Fairchild HUF75337P3), the on resistance increases by something like 1% per degree Celsius rise. For diodes because the forward voltage drop actually decreases with increasing junction temperature - they get more efficient. For an International Rectifier 12CWQ03FN it looks to be about 0.2% lower per degree Celsius rise.
The YS-Tech 80mm fans in this box next to me consume 0.84W at full speed. That's a slow fan though, I wouldn't be surprised if more typical ones used 2-3W at full speed.
Hardly a complete analysis, but just can't see where you're getting this additional 50W from. I think you're out by an order of magnitude.
I'd say the [unemployment] figure (if you include things like people who were previously "professionals" of one stripe or another who are now doing things like phone support and credit card telemarketing) would be more like 20-25%.
So, if you include people who are employed in the unemployment figures the unemployment figures get higher? I am in absolute awe of your ability to distort your perception of reality. It's just utterly mind-boggling. Even politicians wouldn't be so utterly brazen. Here comes the clue stick. Written on the side in really big letters is this: people with jobs are not unemployed.
You do pretty much the same thing. It's really not very different, just imagine the class definition and constructor being rolled into one.
You break up the source code however you like, but you often get many classes in one file because a) JavaScript apps and thus classes tend to be smaller than those in C++ or Java and b) you're usually loading it over HTTP so keeping the number of requests down helps. You can fetch new code and add things (eg. new methods) to objects and "classes" of objects at runtime, which can influence how you lay files out - you might have a "class" for a dynamic element which has collision physics in a separate file so you're not carrying that baggage when you're not using it for a game, but you use the same class elsewhere for simpler dynamic content. Working that way tends to flatten you "class" hierarchy - rather than Manager and Technician inheriting from Employee you can just have Employee and add Manager or Technician (or both) capabilities to it as and when required. That's basically how you do multiple inheritance, but you'd only do it at object creation time if you wanted a more class-based feel. You can design in a very similar style to rigid class-based languages if you want to, but you'll be missing half the fun.
That kind of flexibility makes it easy to write code that's bug-prone and impossible to follow, let alone debug. Combined with the lack of strong typing, namespaces and the kind of debuggers other languages enjoy you need some discipline to write medium to large JavaScript apps - object checking, type checking, naming conventions and exception handling are essential tools.
Google Gears will let you do threads in JavaScript, so you can make responsive apps which do something approaching real work. The lack of threads is a bitch though, you have to not do anything too strenuous, live with an unresponsive GUI or do something really dirty like insert SetTimeout calls in your heavy lifting code to give the UI a chance (and hope the OS/browser can manage a short enough timeout that performance doesn't turn entirely to poop).
"IP world" is in no small way confusing because there's term which lumps three distinct things together with a name which implies some correlation with physical property. If people just talked about copyright, patents and trademarks there would be far less confusion.
You want to get rid of squatters? Simple:
1) Elimintate "tasting" completely.
2) Require an actual site (not just a page of ads) go live at any give address within 30 days.
1 makes a lot of sense, but 2 ignores the fact that not all domains are used for web sites and it is impractical to enforce. Remember that a domain only costs a few dollars, so your mechanism has to cost a few cents per domain, which totally eliminates the possibility of human interpretation as normal practice. Any automated system would be "spoofed" in short order.
I'm looking for a job. £42000 for 42 days (you get £3k a day for the last 14 days if not convicted) would make it by a good margin the best contract job I'd ever had. Being interrogated by MI5 for six weeks surely can't be as mentally damaging as fixing some jerk's shoddy PHP.
A night in the cells and some community service (in the UK) is as close as I've been to prison. Community service usually means spending a fair amount of time with people with first-hand experience of prison. Anyway, your question is a false dichotomy - there is a whole spectrum between direct experience and mere assumptions. First-hand experience isn't required to gain knowledge about things. Conditions in prisons are reported in newspapers, on the news and in documentaries. Books and academic journals too, though I've never had the urge to delve quite that deeply into this subject.
Don't lump the rest of us Anglos-Saxons in with the Americans. UK prisons may not the most pleasant in the West (though they are currently overcrowded), but they're a damn sight more civilised than those in the USA.
You're reading/. and you've never even read an O'Reilly book? Wow. I though everyone with with a reasonable level of interest in computers would own one, or at least have had one thrust at them when they asked a colleague for help. I've got all of them. Well, sort of - I've got a Safari subscription. Half a dozen or so paper ones too. Safari is awesome, by the way. The web is nice and all, good for specific answers, but when you need a properly structured introduction or detailed reference you can't beat instant access to an enormous library of books. Having the option to buy an electronic copy I can be assured will work with all my future devices will make it even nicer.
Not sure if you can edit the database manually, but it looks like pkgutil's --edit-pkg and --learn options might do the trick to update the package receipts Repair Permissions uses.
Does ARD continue to work after you've changed the permissions? If it doesn't you might as well just remove it.
Your question is a red herring. It doesn't matter if the person intended to configure their router as requiring a password or not. If it's not yours and you're not sure you have permission to use the network you do not have permission to use it. It's really that simple. The default position for dealing with other people's property is that if you don't know for certain it's OK for you to use it's not OK for you to use. Why do you think different rules apply to wireless networks than to everything else? Why do you think an inadequately secured wireless network constitutes permission to use it? You're implying that it's OK taking things from people without permission if they are ignorant of or incompetent at adequately securing their stuff. That's just pure bullshit. It's not OK either legally or morally. It's not yours so you don't get to make the decision whether it's OK or not.
This is the key: AC power is available everywhere.
Everywhere has AC power? You really don't get out much, do you? I honestly wonder if you've ever left the house. A few places I use my laptop that do not have AC power: Trains, aeroplanes, buses. Cafes, pubs, restaurants. The park, my garden, in fact outside in general. My bathroom. Some of those places do in fact have AC power, but do not have it in a convenient place. If the socket is on the wall of the cafe and the only free table is in the middle of the room you're screwed.
A machine cannot grant you permission, it's just a machine. All a machine can do is grant you access. A human being has to grant you the permission. Argument by analogy just doesn't work for this - it's not a house, or a car, or a sprinkler. It's a wireless network. If it's not your network you don't get to decide who has permission to connect. Incidentally, the "sign" doesn't say "please come in", it says only says "here is network X".
Better analogy: you knock on the door, a 5 year-old answers, you ask them if you can come in and they say yes. While the child and the router are capable of granting access, nether one can grant you the necessary permission.
Capacitors don't pass current at DC, but they will pass AC current. An analogue signal (which POTS is) is by definition AC. What you need is high impedance. Any old op-amp will have an input impedance of 10^5 ohms or better (often an order of magnitude better), which would cause a negligible voltage drop and be virtually undetectable.
Only if your definition of "serious" only includes law enforcement or government. Industrial espionage is pretty serious and shady private detectives can't just wave a court order at the telecoms company, they have to do it the old-fashioned way. But as far as clicking and the like goes, it would take a pretty clumsily designed device to be detectable that easily.
Slashdot Mirror
It's Mr, not Sir. He aint British, so he just gets to append "KBE" to his name, not style himself as "Sir Bill".
It's pretty obvious what happened, the Linux kernel has problems with the hardware in question. It's not like that's a unique situation. Even if Linux works with 95% of machines one person in 4000 will have consecutive failures on two different machines. And they'll sure as hell make some noise about it when it happens.
It's not the same situation. "Linux" (as in the GNU/Linux/KDE/Mozilla/blah/blah/blah system) as a whole doesn't have a boss who can tell people to fix something. Bill Gates can tell the Windows team to streeamline some process and they'll (try to) do it, because if they say no they're liable to get canned. There is nobody in overall charge of the "Linux" OS. Even they myrid components which make up a Linux system rarely have more than a leader who can make suggestsions and accept or reject patches. I think a large part of Ubuntu's success is that there is somebody in charge of a bunch of programmers who will tweak things on command.
Why would he not buy it? By buying through normal channels it he gets to experience what his users experience. It's not like he, or Microsoft, can't afford to burn a few bucks on usability testing.
Linux distros, in common with Apple, can bundle whatever they like because they don't have a monopoly on the desktop operating system market. It would be interesting to see what would happen if Ubuntu or another Linux distriubtion did become dominant. Do you think commerical vendors (they'd all be selling stuff for Linux by that point) wouldn't press for an anti-trust suit? Canonical is a for-profit corporation after all. Dominance on of the desktop would help their business, even if only 10% of users ever paid them anything.
That's dangerously bad advice. In addition to the headline power rating, PSUs are rated per rail (5V, 12V etc.). It's extremely unlikely that your computer would draw power from the various supply lines in exactly the proportions the supply can provide. Most notably, older systems drew CPU power from the 5V rail. Boards for the P4 and other more modern boards use the 12V rail for CPU power. (In both cases there's a switching regulator which coverts this to the CPU's actual supply voltage.) PSUs, at least those sold to home users, are designed to handle either situation, so you may approach the limits for one rail or the other, but rarely both at the same time. Unless you're an OEM who can control the entire design, you are very unlikely to be able to run a system which requires 200W from a "220W" PSU - you'll be over the rating one one rail or another.
All PSUs are not created equal. Those gold 550W PSUs which were popular a few years ago provided less power on the 5V line than the 300W Antec I went for in the end. If you were overclocking an Athlon (as I was) that difference was make-or-break.
Only if you use the peak-to-peak voltage. For AC, the voltage everybody quotes is the RMS voltage. For a sine wave, the RMS value is 1/(root 2), about 0.707, times the peak value. With the RMS voltage it's just P=VI, which is why it's the RMS value which is quoted. The reason for the VA rating is more to do with power factors - how close to a resistive load the actual load is. A very basic switching supply will only draw any current at all close to the peak voltage and will draw nothing the rest of the time. If you're designing generators, inverters and transformers it matters a lot if your 5A average load is spread out over the cycle or if the load is zero 99% of the time and 500A 1% of the time.
Resistance does increase with temperature and a thermally controlled fan will spin faster and draw more current. But enough for a 25% rise in consumption from a 15 degree (in unspecified units, I guess you mean Fahrenheit) temperature rise? That's seems like a hell of a lot for a fairly modest rise in temperature.
For an 80% efficient power supply, an increase of 25% overall consumption is more than double the power loss. The reality is very complex, but we can pick out a few relevant numbers to get a feel for the magnitudes involved. Empirical testing would be easier than an analysis, but here's some food for thought:
For copper, the resistance rises by about 0.4% per degree Celsius rise. Your roughly 7 Celsius rise would increase it by a whopping 2.8%. You'll have melted the insulation well before even a 50% rise in the resistance of your copper wire.
If you look inside a power supply, you'll see a big fat heatsink. Attached to that are rectifiers and switches - diodes and FETs. That's where a big proportion of your power supply's inefficiency comes from. Looking at the first power FET datasheet I have to hand (for a Fairchild HUF75337P3), the on resistance increases by something like 1% per degree Celsius rise. For diodes because the forward voltage drop actually decreases with increasing junction temperature - they get more efficient. For an International Rectifier 12CWQ03FN it looks to be about 0.2% lower per degree Celsius rise.
The YS-Tech 80mm fans in this box next to me consume 0.84W at full speed. That's a slow fan though, I wouldn't be surprised if more typical ones used 2-3W at full speed.
Hardly a complete analysis, but just can't see where you're getting this additional 50W from. I think you're out by an order of magnitude.
You do pretty much the same thing. It's really not very different, just imagine the class definition and constructor being rolled into one.
You break up the source code however you like, but you often get many classes in one file because a) JavaScript apps and thus classes tend to be smaller than those in C++ or Java and b) you're usually loading it over HTTP so keeping the number of requests down helps. You can fetch new code and add things (eg. new methods) to objects and "classes" of objects at runtime, which can influence how you lay files out - you might have a "class" for a dynamic element which has collision physics in a separate file so you're not carrying that baggage when you're not using it for a game, but you use the same class elsewhere for simpler dynamic content. Working that way tends to flatten you "class" hierarchy - rather than Manager and Technician inheriting from Employee you can just have Employee and add Manager or Technician (or both) capabilities to it as and when required. That's basically how you do multiple inheritance, but you'd only do it at object creation time if you wanted a more class-based feel. You can design in a very similar style to rigid class-based languages if you want to, but you'll be missing half the fun.
That kind of flexibility makes it easy to write code that's bug-prone and impossible to follow, let alone debug. Combined with the lack of strong typing, namespaces and the kind of debuggers other languages enjoy you need some discipline to write medium to large JavaScript apps - object checking, type checking, naming conventions and exception handling are essential tools.
Google Gears will let you do threads in JavaScript, so you can make responsive apps which do something approaching real work. The lack of threads is a bitch though, you have to not do anything too strenuous, live with an unresponsive GUI or do something really dirty like insert SetTimeout calls in your heavy lifting code to give the UI a chance (and hope the OS/browser can manage a short enough timeout that performance doesn't turn entirely to poop).
"IP world" is in no small way confusing because there's term which lumps three distinct things together with a name which implies some correlation with physical property. If people just talked about copyright, patents and trademarks there would be far less confusion.
1 makes a lot of sense, but 2 ignores the fact that not all domains are used for web sites and it is impractical to enforce. Remember that a domain only costs a few dollars, so your mechanism has to cost a few cents per domain, which totally eliminates the possibility of human interpretation as normal practice. Any automated system would be "spoofed" in short order.
I'm looking for a job. £42000 for 42 days (you get £3k a day for the last 14 days if not convicted) would make it by a good margin the best contract job I'd ever had. Being interrogated by MI5 for six weeks surely can't be as mentally damaging as fixing some jerk's shoddy PHP.
A night in the cells and some community service (in the UK) is as close as I've been to prison. Community service usually means spending a fair amount of time with people with first-hand experience of prison. Anyway, your question is a false dichotomy - there is a whole spectrum between direct experience and mere assumptions. First-hand experience isn't required to gain knowledge about things. Conditions in prisons are reported in newspapers, on the news and in documentaries. Books and academic journals too, though I've never had the urge to delve quite that deeply into this subject.
Don't lump the rest of us Anglos-Saxons in with the Americans. UK prisons may not the most pleasant in the West (though they are currently overcrowded), but they're a damn sight more civilised than those in the USA.
You're reading /. and you've never even read an O'Reilly book? Wow. I though everyone with with a reasonable level of interest in computers would own one, or at least have had one thrust at them when they asked a colleague for help. I've got all of them. Well, sort of - I've got a Safari subscription. Half a dozen or so paper ones too. Safari is awesome, by the way. The web is nice and all, good for specific answers, but when you need a properly structured introduction or detailed reference you can't beat instant access to an enormous library of books. Having the option to buy an electronic copy I can be assured will work with all my future devices will make it even nicer.
Not sure if you can edit the database manually, but it looks like pkgutil's --edit-pkg and --learn options might do the trick to update the package receipts Repair Permissions uses.
Does ARD continue to work after you've changed the permissions? If it doesn't you might as well just remove it.
Your question is a red herring. It doesn't matter if the person intended to configure their router as requiring a password or not. If it's not yours and you're not sure you have permission to use the network you do not have permission to use it. It's really that simple. The default position for dealing with other people's property is that if you don't know for certain it's OK for you to use it's not OK for you to use. Why do you think different rules apply to wireless networks than to everything else? Why do you think an inadequately secured wireless network constitutes permission to use it? You're implying that it's OK taking things from people without permission if they are ignorant of or incompetent at adequately securing their stuff. That's just pure bullshit. It's not OK either legally or morally. It's not yours so you don't get to make the decision whether it's OK or not.
Everywhere has AC power? You really don't get out much, do you? I honestly wonder if you've ever left the house. A few places I use my laptop that do not have AC power: Trains, aeroplanes, buses. Cafes, pubs, restaurants. The park, my garden, in fact outside in general. My bathroom. Some of those places do in fact have AC power, but do not have it in a convenient place. If the socket is on the wall of the cafe and the only free table is in the middle of the room you're screwed.
A machine cannot grant you permission, it's just a machine. All a machine can do is grant you access. A human being has to grant you the permission. Argument by analogy just doesn't work for this - it's not a house, or a car, or a sprinkler. It's a wireless network. If it's not your network you don't get to decide who has permission to connect. Incidentally, the "sign" doesn't say "please come in", it says only says "here is network X".
Better analogy: you knock on the door, a 5 year-old answers, you ask them if you can come in and they say yes. While the child and the router are capable of granting access, nether one can grant you the necessary permission.
Capacitors don't pass current at DC, but they will pass AC current. An analogue signal (which POTS is) is by definition AC. What you need is high impedance. Any old op-amp will have an input impedance of 10^5 ohms or better (often an order of magnitude better), which would cause a negligible voltage drop and be virtually undetectable.
Only if your definition of "serious" only includes law enforcement or government. Industrial espionage is pretty serious and shady private detectives can't just wave a court order at the telecoms company, they have to do it the old-fashioned way. But as far as clicking and the like goes, it would take a pretty clumsily designed device to be detectable that easily.