How so? False negative versus false positive rate. With a background check, you risk missing something. With a polygraph, there is the additional risk that you fire someone who did nothing wrong.
in commercial Civ games ramping up the difficulty usually means giving bonuses and cheats to the AI.
It's at least 10 years ago that I last played Freeciv, but I recall that back then, Freeciv AI was also cheating at the higher difficulty levels, by getting resources cheaper and by being able to peek into enemy cities. It didn't matter that much to me, I never got good enough to bother with the higher difficulty levels - and it was still quite a few all-nighters that I pulled playing it.
However, from a brief look at Freeciv AI documentation, I get it that nowadays the AI is not cheating, and just gets handicaps at lower difficulty levels.
I use xterm (with a small bitmap font), but a disadvantage is the old-style mouse-button copy/paste. Now that fewer and fewer laptops have a middle mouse button, it gets hard to paste text. I sometimes run linux in Virtualbox (Windows host) and I never managed to get middle-button emulation working there.
A single book has 10^5 words. The passphrase is a sequence of 5 consecutive words from the book, so there are (10^5 - 5) possible 5-word passphrases that you can draw from this book. Much less if passphrases must start after a period/comma/semicolon/etc. A bit more if you also allow 4- or 6-word pasphrases. Much less than if you draw 5 random words from a dictionary or book, but that's much more difficult to remember (at least, I won't remember tens of correct-battery-horse-staple passphrases).
"Thus he would know which students have a feel for the numbers and answers and who blindly follow the procedures"
Eehm, so which students did he consider good? The ones with the mathematically correct answer, or the realistic answer?
The soot ends up in the soot filter, which is automatically cleaned every now and then by burning it into CO2. Hence, carbon emission = fuel consumption x conversion factor.
The reason they are talking about CO2 emissions is because in Europe, cars are typically taxed based on CO2 emissions, not on mpg or l/100 km. Fuel consumption values are too much apples and oranges between diesel, LPG, gasoline/petrol, and electric.
Note that CO2 values are a bit higher than you'd expect from chemistry; some of the energy use during production are accounted for as well.
The keyspace is only large if the attacker doesn't know or suspect how your password is constructed. Otherwise: 10^3 possible famous books, 10^5 words (starting positions) per book, 5 possible key lengths, 2 for with/without spaces. This gives you a key space of 10^9 that can easily be brute-forced.
And if they/NSA see you look up the book on your browser, you're definitely done.
Technically correct. But kJ in urine means you have diabetes; kJ in feces in any form that can be processed by bacteria would give you tremendous intestonal upset. Think of people with lactose intolerance; because they don't digest/absorb milk sugar in the upper intestine, bacteria will do so in the lower intestine.
A more important difference is in the "kJ expended", which involves more than physical exercise. The digestion process itself wastes energy (generates heat) and brown fat in the body continually convert food kJ to heat kJ, with large individual variation and generally more in children/young people than in older people (40+).
"I realize not every pill has the same manufacturing cost, but they are at least within an order of magnutude of each other for the most part."For some value of "most". Ibuprofen is a small molecule, separating stereoisomers is not necessary, and it has a huge economy of scale. Granted, daraprin doesn't look too difficult either (http://www.rxlist.com/script/main/mobileart-rx.asp?drug=daraprim&monotype=rx-desc&monopage=1 ).
Re uranium and mercury in coal: I call BS. AFAIK modern coal plants use scrubbers to remove the vast majority of stuff that is not air or CO2 from their exhaust, so that's only relevant for ancient plants. Moreover, the story about U in coal is based on coal mined in one particular location that is not representative. Finally, radioactivity of natural uranium is a non-issue with a half life of a billion years. It's the fission products that are produced in a reactor that are the problem, especially in case of broken containment.
By the way, I bought a solid-state gamma ray detector (Pocket geiger, made in japan near Fukushima, plugs into a smartphone) and was quite disappointed with the apparent lack of radioactivity in things like ceramics and fertilizer. Turned out that SS gamma detectors are actually quite insensitive.
He meant: F=(dm/dt)v for thrust and P=(dm/dt)v^2/2 for power. Alternatively p=mv for momentum and KE=mv^2/2.
Your F=ma is not very relevant either for analyzing propulsion mechanisms.
"I seem to recall from a while ago that a surprising number of passive things go on while the phone is off. The OS may not be running, but it's not the only thing tracking you."
Back when the Snowden thing started, he mentioned things like that, which was then vehemently denied by the manufacturers. They saw no way how the radio could be powered with the phone switched off. Of course, he was in a good position to be paranoid about phones with the firmware modified by TLAs.
There are phones that can wake up from power off to sound an alarm. Presumably, they boot some lightweight OS image that's just enough to play the alarm audio and display a snooze button. A phone like that could be modified to wake up with the screen off and the radios on (and remember the unlock code for the SIM). But that doesn't mean that phones work like that out of the box. There would be trouble with the FCC and FAA if phones did that during air travel, not to mention unexpected data roaming fees.
I've a Brother color laser/led printer, had it set into toner save mode and rarely printed anything in color. After a few hundred pages, it claimed that the black was finished and the color toner at 50%. Toner saver my ass; apparently the cartridges have some kind of mechanical counter and when the mechanism turns at the power-on self test, it counts, even if zero toner is used. If you use the printer mostly for one-page jobs, that counts. I was able to reset the counter, but it was a pain.
"this [4G off state] might result in data-checking apps (e.g. your e-mail app) not getting notifications that an e-mail has arrived."
You got it. Or rather, apps stop polling for data if there is no internet connection. With the 4g radio in standby (only listening for incoming calls) it doesn't consume much power.
"So perhaps it could turn on the data for 5 minutes every hour to allow background apps to pull data. I wonder just how much battery life an app like this would save."
There are apps (e.g. juice defender) that can disable data sync or even mobile data (iirc) and briefly enable it at set intervals, so that all apps wake up and sync at the same time. You could also try to do that manually (it's the widget with two half-circle arrows). I tried JD in the past, but found it too much a pain that Whatsapp messages came in with a delay and that I had to wait for mobile data to become active whenever I wanted to look up something.
"Battery doctor already does this, I've had it on my phone for a year or two, it simply terminates one or two dozen apps that somehow run themselves for no good reason - has a whitelist too."
BD seems like snake oil to me. From the app description:
- Disable unnecessary apps that drain your battery! Task Killer kills tasks with one click! -- The ones that drain the battery are background services and you can't really kill them without root; the OS will restart them after a while. You can disable the service manually in the Android settings if you want, although i'm not sure that it survives a reboot. Killing "tasks" (recently opened apps) won't help the battery at all since those don't run but just sit idly in memory until the memory is needed for something else. After that, the "task" just a shortcut icon in the recent-apps list. Killing an app is only useful if it's stuck in an unresponsive state. Killing a "task" generally won't kill the background service if the app has one.
- Unique 3 Stage Charging system! -- Unique, WTF, that's how every Li-ion charger works. No way that an app can influence the charging circuitry of the phone.
"AES instructions are included by default in almost every single processor produced in the last 5 years."
Not in the i3-3xxx mobile cpus (released 2013), celeron N29xx (released 2014), Pentium N35xx (2014), and so on. I.e. my laptop and my SO's... (we're more interested in battery life and compactness)
And ssh on my phone (ARM) isn't particularly fast even if the hardware supports it. Can't tell whether dropbear and ssh client actually use AES instructions.
"in order to calculate which pages are linked to by good pages, and are therefore also good. It's recursive across the whole internet"
You speak in the present tense, but I think it's widely believed that today, the original pagerank algorithm plays only a minor role. The original algorithm was very easy to game by building a site with a million auto-generated pages, all linking to each other and to the main page. How they actually do it today is a closely guarded secret, although it's likely that links between sites and internal links play a role.
OK, I was reading and posting from a phone screen while commuting. From a laptop screen and based on your screenshots, I agree. The Hack-i looks like a tiny backslash, quite disrupting the flow of reading. The Hack-a at your rendering looks like a small 8 when viewed from a distance, unlike the Dejavu-a.
Next time you post screenshots, make sure that they will fit in a browser window with menu bar on a 1366x768 sceen, though. The browser was messing them up by an additional scaling.
"When setting up an access point, it should be possible to designate it as "expensive", and by default devices should adhere to this and try to limit unnecessary data usage"
Android has a feature (settings / data usage / menu / mobile hotspots) to do exactly that. Android also seems to detect if it is tethered to another Android phone but I'm not sure how that works. iPhones certainly don't recognize Android hotspots, as a I learned when my friend's iPhone downloaded 50 MB roaming data in 3 minutes when she just wanted to check her email.
"These tests should have been part of the product test and qualification plan."
I suspect that they don't test the device in their final assembled form because of the tight development and release schedule for thus kind of devices. Redesigning the spring mechanism and setting up the production lines could mean months of delay.
They need to trust their engineers to catch this kind of problem while the design only exists as a CAD model. Unfortunately that didn't happen this time...
Slashdot Mirror
How so? False negative versus false positive rate. With a background check, you risk missing something. With a polygraph, there is the additional risk that you fire someone who did nothing wrong.
"Polygraphs are not 100% pointless, they *do* work for *some* people so there is actually a benefit to their use...."
The problem is that you have to know whether a person being tested is one of those "some people".
It's at least 10 years ago that I last played Freeciv, but I recall that back then, Freeciv AI was also cheating at the higher difficulty levels, by getting resources cheaper and by being able to peek into enemy cities. It didn't matter that much to me, I never got good enough to bother with the higher difficulty levels - and it was still quite a few all-nighters that I pulled playing it.
However, from a brief look at Freeciv AI documentation, I get it that nowadays the AI is not cheating, and just gets handicaps at lower difficulty levels.
I use xterm (with a small bitmap font), but a disadvantage is the old-style mouse-button copy/paste. Now that fewer and fewer laptops have a middle mouse button, it gets hard to paste text. I sometimes run linux in Virtualbox (Windows host) and I never managed to get middle-button emulation working there.
A single book has 10^5 words. The passphrase is a sequence of 5 consecutive words from the book, so there are (10^5 - 5) possible 5-word passphrases that you can draw from this book. Much less if passphrases must start after a period/comma/semicolon/etc. A bit more if you also allow 4- or 6-word pasphrases. Much less than if you draw 5 random words from a dictionary or book, but that's much more difficult to remember (at least, I won't remember tens of correct-battery-horse-staple passphrases).
"Almost no election fraud involves coercing votes one individual at a time."
Maybe that's because elections are set up to make coersion very difficult. It doesn't necessarily stay so once you offer the technical capability.
"Thus he would know which students have a feel for the numbers and answers and who blindly follow the procedures" Eehm, so which students did he consider good? The ones with the mathematically correct answer, or the realistic answer?
The soot ends up in the soot filter, which is automatically cleaned every now and then by burning it into CO2. Hence, carbon emission = fuel consumption x conversion factor.
The reason they are talking about CO2 emissions is because in Europe, cars are typically taxed based on CO2 emissions, not on mpg or l/100 km. Fuel consumption values are too much apples and oranges between diesel, LPG, gasoline/petrol, and electric.
Note that CO2 values are a bit higher than you'd expect from chemistry; some of the energy use during production are accounted for as well.
The keyspace is only large if the attacker doesn't know or suspect how your password is constructed. Otherwise: 10^3 possible famous books, 10^5 words (starting positions) per book, 5 possible key lengths, 2 for with/without spaces. This gives you a key space of 10^9 that can easily be brute-forced.
And if they/NSA see you look up the book on your browser, you're definitely done.
"KJ food input = KJ stored + KJ expended (effort) + KJ expelled (urine + feces)"
Technically correct. But kJ in urine means you have diabetes; kJ in feces in any form that can be processed by bacteria would give you tremendous intestonal upset. Think of people with lactose intolerance; because they don't digest/absorb milk sugar in the upper intestine, bacteria will do so in the lower intestine.
A more important difference is in the "kJ expended", which involves more than physical exercise. The digestion process itself wastes energy (generates heat) and brown fat in the body continually convert food kJ to heat kJ, with large individual variation and generally more in children/young people than in older people (40+).
"I realize not every pill has the same manufacturing cost, but they are at least within an order of magnutude of each other for the most part."For some value of "most". Ibuprofen is a small molecule, separating stereoisomers is not necessary, and it has a huge economy of scale. Granted, daraprin doesn't look too difficult either (http://www.rxlist.com/script/main/mobileart-rx.asp?drug=daraprim&monotype=rx-desc&monopage=1 ).
But not all pharma compounds are like that. For example, taxol as an extreme counter example: https://en.m.wikipedia.org/wik... .
Re uranium and mercury in coal: I call BS. AFAIK modern coal plants use scrubbers to remove the vast majority of stuff that is not air or CO2 from their exhaust, so that's only relevant for ancient plants. Moreover, the story about U in coal is based on coal mined in one particular location that is not representative. Finally, radioactivity of natural uranium is a non-issue with a half life of a billion years. It's the fission products that are produced in a reactor that are the problem, especially in case of broken containment. By the way, I bought a solid-state gamma ray detector (Pocket geiger, made in japan near Fukushima, plugs into a smartphone) and was quite disappointed with the apparent lack of radioactivity in things like ceramics and fertilizer. Turned out that SS gamma detectors are actually quite insensitive.
He meant: F=(dm/dt)v for thrust and P=(dm/dt)v^2/2 for power. Alternatively p=mv for momentum and KE=mv^2/2. Your F=ma is not very relevant either for analyzing propulsion mechanisms.
There are only a few crops that can reach that kind (6%) of efficiency. 1% is a fairly typical number...
"I seem to recall from a while ago that a surprising number of passive things go on while the phone is off. The OS may not be running, but it's not the only thing tracking you."
Back when the Snowden thing started, he mentioned things like that, which was then vehemently denied by the manufacturers. They saw no way how the radio could be powered with the phone switched off. Of course, he was in a good position to be paranoid about phones with the firmware modified by TLAs.
There are phones that can wake up from power off to sound an alarm. Presumably, they boot some lightweight OS image that's just enough to play the alarm audio and display a snooze button. A phone like that could be modified to wake up with the screen off and the radios on (and remember the unlock code for the SIM). But that doesn't mean that phones work like that out of the box. There would be trouble with the FCC and FAA if phones did that during air travel, not to mention unexpected data roaming fees.
I've a Brother color laser/led printer, had it set into toner save mode and rarely printed anything in color. After a few hundred pages, it claimed that the black was finished and the color toner at 50%. Toner saver my ass; apparently the cartridges have some kind of mechanical counter and when the mechanism turns at the power-on self test, it counts, even if zero toner is used. If you use the printer mostly for one-page jobs, that counts. I was able to reset the counter, but it was a pain.
"this [4G off state] might result in data-checking apps (e.g. your e-mail app) not getting notifications that an e-mail has arrived."
You got it. Or rather, apps stop polling for data if there is no internet connection. With the 4g radio in standby (only listening for incoming calls) it doesn't consume much power.
"So perhaps it could turn on the data for 5 minutes every hour to allow background apps to pull data. I wonder just how much battery life an app like this would save."
There are apps (e.g. juice defender) that can disable data sync or even mobile data (iirc) and briefly enable it at set intervals, so that all apps wake up and sync at the same time. You could also try to do that manually (it's the widget with two half-circle arrows). I tried JD in the past, but found it too much a pain that Whatsapp messages came in with a delay and that I had to wait for mobile data to become active whenever I wanted to look up something.
"Battery doctor already does this, I've had it on my phone for a year or two, it simply terminates one or two dozen apps that somehow run themselves for no good reason - has a whitelist too."
BD seems like snake oil to me. From the app description:
- Disable unnecessary apps that drain your battery! Task Killer kills tasks with one click! -- The ones that drain the battery are background services and you can't really kill them without root; the OS will restart them after a while. You can disable the service manually in the Android settings if you want, although i'm not sure that it survives a reboot. Killing "tasks" (recently opened apps) won't help the battery at all since those don't run but just sit idly in memory until the memory is needed for something else. After that, the "task" just a shortcut icon in the recent-apps list. Killing an app is only useful if it's stuck in an unresponsive state. Killing a "task" generally won't kill the background service if the app has one.
- Unique 3 Stage Charging system! -- Unique, WTF, that's how every Li-ion charger works. No way that an app can influence the charging circuitry of the phone.
"AES instructions are included by default in almost every single processor produced in the last 5 years."
Not in the i3-3xxx mobile cpus (released 2013), celeron N29xx (released 2014), Pentium N35xx (2014), and so on. I.e. my laptop and my SO's... (we're more interested in battery life and compactness)
And ssh on my phone (ARM) isn't particularly fast even if the hardware supports it. Can't tell whether dropbear and ssh client actually use AES instructions.
"in order to calculate which pages are linked to by good pages, and are therefore also good. It's recursive across the whole internet"
You speak in the present tense, but I think it's widely believed that today, the original pagerank algorithm plays only a minor role. The original algorithm was very easy to game by building a site with a million auto-generated pages, all linking to each other and to the main page. How they actually do it today is a closely guarded secret, although it's likely that links between sites and internal links play a role.
OK, I was reading and posting from a phone screen while commuting. From a laptop screen and based on your screenshots, I agree. The Hack-i looks like a tiny backslash, quite disrupting the flow of reading. The Hack-a at your rendering looks like a small 8 when viewed from a distance, unlike the Dejavu-a.
Next time you post screenshots, make sure that they will fit in a browser window with menu bar on a 1366x768 sceen, though. The browser was messing them up by an additional scaling.
"For example, the line in the zero, the changes to the i and a -- all are horrible."
Could yo explain.why the new 'i' is horrible but not the original 'l' (ell)? And I don't quite see any difference in 'a' in these two comparisons.
http://gfycat.com/SomberUnited...
http://i.imgur.com/8SqL6mT.gif
"When setting up an access point, it should be possible to designate it as "expensive", and by default devices should adhere to this and try to limit unnecessary data usage"
Android has a feature (settings / data usage / menu / mobile hotspots) to do exactly that. Android also seems to detect if it is tethered to another Android phone but I'm not sure how that works. iPhones certainly don't recognize Android hotspots, as a I learned when my friend's iPhone downloaded 50 MB roaming data in 3 minutes when she just wanted to check her email.
"These tests should have been part of the product test and qualification plan."
I suspect that they don't test the device in their final assembled form because of the tight development and release schedule for thus kind of devices. Redesigning the spring mechanism and setting up the production lines could mean months of delay.
They need to trust their engineers to catch this kind of problem while the design only exists as a CAD model. Unfortunately that didn't happen this time...
Normally, "variety" in the context of RSI refers to changing your posture and activities, though.