What makes you think Linux's RNG is 'best practice'. There are many of us in the huge and diverse RNG designer community that think it stinks.
I probably didn't express myself well, but by "best practice" I meant: for programmers with only a little understanding of design of random number generators (like me), it's probably better to assume that smarter people have done a better effort than I could ever do, designing/dev/random and/dev/urandom behaviour for the Linux kernel (/dev/urandom good enough for computer games and/dev/random for crypto applications and Monte-Carlo optimizations).
If, as you say, "it stinks", then what do you recommend? FreeBSD? Could you please elaborate why "it stinks"?
But, if it's a CPU instruction, you'd have to write code for it yourself in assembler. How many people would go that far, to bypass "best practice" default Linux kernel randomness, in order to only use this one CPU maker's code? It would look very suspicious:
"Our security program runs only on that specific line of Intel CPUs, because.. well.. that's more secure than using anything else! just trust us!"
Of course it would matter, it would significantly lower entropy and make entropy estimates incorrect.
I don't understand this: suppose you have this suspicious RdRand entropy source, and another source that you trust much better, say an external thermometer, and you XOR the two. Suppose the RdRand output is just 0 bits all the time (worst case scenario); this would still give the same entropy after the XOR as if you just used the thermometer only, i.o.w. how has the entropy been lowered? Mixing two random sources that rely on differing physical phenomena would probably improve the randomness, but how can it go down if you XOR it with whatever imperfect randomness?
I get it, this is just a sexist comment to make us believe that soccer moms don't have any inclination to study C and Linux kernel programming.
It's probably likely that only a small percentage of the soccer moms have any interest in system-level programming, however you're being terribly stigmatizing to that small percentage.
I'm a man, so I can't comment, but I'd ask any soccer mom Slashdot reading programmers to give their input here on whether they think Linus acts like a 12-year-old dumbass or whatever.
I don't know.. wouldn't military people feel an urge to "win", to "excel", to "do something (exciting)" wouldn't you rather have people that have an urge to "gently extract the soy bean sprouts from the red sand/astronaut-turd mixture and sprinkle them with water, for the rest of my life-- whether my death from starvation comes in 2 years or in 6 years".
Talk to this guy about building a prototype Mars base in Poland, maybe.
Plants don't need the full spectrum, so maybe it is possible to use solar panels that are space-rated to produce electricity, put the gardens underground protected from solar flares, and use LED lamps that don't have green to grow the plants. I've seen a picture somewhere. Aren't the Russians (IMBP) doing exactly the same thing (what did they eat on the Mars 500 experiment? IIRC it was a little bit of fresh vegetables as well as the contents of 1 large food storage module. Who knows, maybe the Chinese have got good results with hydroponic paksoi by now. There is an extremely steep "learning curve" but for all of humanity there should be a tremendous pay-off in learning "this is how complicated it is to create an artificial biosphere. So don't fuck up the one fully functional biosphere we have."
There are probably extremely important scientific results that are needed for this but are not accessible enough (e.g. just imagine that some Tchadian agricultural institute studied "how to grow salad and earth worms in sharp sand", I'm just making this example up).
Exactly the same issues came up in the BBC documentary series "Castaway 2000" on Taransay. And that was only for 1 year. It was truly fascinating television to watch.
Granted, on Mars they'd need something more advanced than a polytunnel for their veggies...
First of all, thank you for your thoughtful posting.
A few loose comments:
We're stuck wandering around in the uncanny valley between one sigma and five sigma.
Very well put; although according to a previous poster, w.r.t. arctic ice we're at two sigma already, so (if I'm not making a stupid mistake and the data has a normal distribution) it's 4.6% likely that this year's values are normal, assuming a long-term stable arctic ice extent.
Proto-facts are the new king. Everyone line up to call the other side dunderheads for not bowing to your side's self-evident truths. Then complain when the public dials out with their hands over their ears.
Well, there is a whole guild of people who specialize in determining the statistical pay-off between high-damage low-probability events (black swan?) and low-reward high-probability events (maximum squeezable insurance premium), and they're called actuaries. I'd like to know if, say, Lloyds can insure you against global warming affecting your business, and how much it would cost qua premium. What do the actuaries think?
Lastly a loose comment (not based on your essay): one of the global warming denialist talking points is "the necessary effort to stop global warming would bring our society back to the stone age, which is what YOU tree-hugging hippies want". You mentioned "putting the long view out of mind"; the long view is that, 300 years ago, we were before the "age of petroleum" and we didn't live in the stone age, either. (Although maybe in the whale oil age, I don't know). Who knows what the 21st century may bring technologically; maybe a mixture of high-tech low-energy-use equipment and low-tech low-energy-use 19th century technologies.
Yeah, and if Hitler were alive today, he'd probably be a "Warmist" as well, what with his vegetarianism and long walks in the woods.
Give me a break.
This whole Slashdot discussion today, based on a Daily Mail article, seems to be mental preparation of the public so that they're properly revved up for global warming denialism, before the next IPCC report gets published in a few weeks.
So that on 2013-09-27, Joe Public will say to Jane Public: "but it's all rubbish; wasn't that in the newspaper a few weeks ago?".
Yes, but this is exactly the problem. You quote "minimum concentration", but there is also a maximum effective concentration, which is when the "CO2 hole" in the spectrum is closed.
Any CO2 after that is basically just food for plants. There is some argument that CO2 is already at this level, and further CO2 doesn't matter. (other than greening the deserts)
That's an interesting argument, and apparently it took until 1956 until more work was done on that. I googled "CO2 extinction coefficient" and got: http://www.skepticalscience.com/saturated-co2-effect-advanced.htm. In short: the CO2 IR spectrum is not monochromatic, and there are plenty of wavelengths where the IR radiation still comes through and is dependent on concentration, so no.
You'll need to consult a real IR spectroscopist to study more, because this stuff is not trivial.
I do remember vaguely that there was this enormous C-O double bond stretch vibration band around 1800? reciproke centimeters. But it was a bit different in -COOH and aldehydes than in CO2 itself.
So, you hate socialists. Fine. That's your opinion.
But then you go on about Obama and "leftist cabal" and "the socialist lot are pissing on the Constitution".
To me, this means that you are talking about USA politics and name-calling the USA Democrat party as "socialist".
That doesn't make sense in the normal way the word "socialist" is used. The USA Democrat party is very right-wing. The USA Republican party is "bat-shit crazy" extreme right-wing. We outside the USA almost never hear about the left-wing or socialist parties and politics of the USA.
I have heard that you have a Green Party, used to be chaired by the famous Ralph "Seat Belt" Nader, now run by a lady named Cynthia McKinney. She's probably left-wing.
But if you want to see real socialist parties in action (4.5 % of the parliament), read here:
I wasn't laughing at his/her nuclear terminology; on the Internet it's difficult to see if people are joking sometimes (it's called Poe's law) and especially on Slashdot people can sometimes say quite sick things. I thought he/she was referring to Hiroshima and Nagasaki. That is also a likely interpretation of the sentence "We have had a lot of experience with nukes", and my first reaction was "what a sick bastard!".
Do you like Harry Potter? I'll try to elaborate:
In "the Goblet of Fire", there is a scene in ch. 14 where the faux Mad-Eye Moody talks to Neville Longbottom in class:
"'Pain,' said Moody softly. 'You don't need thumbscrews or knives to torture someone if you can perform the Cruciatus curse... that one was very popular once, too.' Right.... anyone know any others?"
No, I meant the clever Odysseus / Ballmer (ok.. bit difficult analogy here) couldn't destroy the walls of Troy / non-MS mobile ecosystem at Nokia, so he sent a gift wooden horse / ex-Microsoftie Elop to open up the gates / burn the existing platform, sell the essential GSM patents, reduce the value of Nokia so the rest can be bought up and sold in pieces. Then the triumphant Greeks went back to their own army.
I'm still a bit shocked that that story is not at the end of the Iliad (I really really thought it was), but ch. 24 ends with the mutilation, sale of the corpse to his dad, and burial of Hektor.
Do you know the scam game with three cups and a pea? Where you win money if you guess right under which cup the pea is after some shuffling?
The "Nokia basic essential mobile telecom patents" pea is no longer under the Nokia cup for some time now. It's now under the cup labelled "Vringo or I/P Engine". Watch where that cup moves and when the pea gets exchanged.
Here's an older article on Groklaw: http://www.groklaw.net/articlebasic.php?story=20130601134450374
The patents have already been sold apparently, to a "ringtone advertisement" company called Vringo Inc. which is now sueing mobile phone companies like ZTE for billions. So who's going to buy Vringo? My guess would be Intellectual Ventures.
Here's a scary thought: maybe most of their trained personnel has already received the maximum lifetime dose and has been given their retirement already..
Maybe your sentence should be split in two; we know that basic research is a government task because it's almost certainly not profitable. On the other hand "opening new frontiers" could be profitable (Hilton hotel on Mars for the super-rich) but it carries so much risk that the taxpayers might revolt, e.g. European Green Parties's complaint against the ITER reactor is not that it is an abomination and the Ents should release the Rhône and flood and extinguish it, but rather that "nuclear fusion is still too far away a goal to spend taxpayer money on at the moment" (note the large difference in position towards nuclear fusion and nuclear fission reactors, here!).
Slashdot Mirror
I probably didn't express myself well, but by "best practice" I meant: for programmers with only a little understanding of design of random number generators (like me), it's probably better to assume that smarter people have done a better effort than I could ever do, designing /dev/random and /dev/urandom behaviour for the Linux kernel (/dev/urandom good enough for computer games and /dev/random for crypto applications and Monte-Carlo optimizations) .
If, as you say, "it stinks", then what do you recommend? FreeBSD? Could you please elaborate why "it stinks"?
But, if it's a CPU instruction, you'd have to write code for it yourself in assembler. How many people would go that far, to bypass "best practice" default Linux kernel randomness, in order to only use this one CPU maker's code? It would look very suspicious:
"Our security program runs only on that specific line of Intel CPUs, because.. well.. that's more secure than using anything else! just trust us!"
I don't understand this: suppose you have this suspicious RdRand entropy source, and another source that you trust much better, say an external thermometer, and you XOR the two. Suppose the RdRand output is just 0 bits all the time (worst case scenario); this would still give the same entropy after the XOR as if you just used the thermometer only, i.o.w. how has the entropy been lowered? Mixing two random sources that rely on differing physical phenomena would probably improve the randomness, but how can it go down if you XOR it with whatever imperfect randomness?
I get it, this is just a sexist comment to make us believe that soccer moms don't have any inclination to study C and Linux kernel programming.
It's probably likely that only a small percentage of the soccer moms have any interest in system-level programming, however you're being terribly stigmatizing to that small percentage.
I'm a man, so I can't comment, but I'd ask any soccer mom Slashdot reading programmers to give their input here on whether they think Linus acts like a 12-year-old dumbass or whatever.
I don't know.. wouldn't military people feel an urge to "win", to "excel", to "do something (exciting)" wouldn't you rather have people that have an urge to "gently extract the soy bean sprouts from the red sand/astronaut-turd mixture and sprinkle them with water, for the rest of my life-- whether my death from starvation comes in 2 years or in 6 years".
Talk to this guy about building a prototype Mars base in Poland, maybe.
Plants don't need the full spectrum, so maybe it is possible to use solar panels that are space-rated to produce electricity, put the gardens underground protected from solar flares, and use LED lamps that don't have green to grow the plants. I've seen a picture somewhere. Aren't the Russians (IMBP) doing exactly the same thing (what did they eat on the Mars 500 experiment? IIRC it was a little bit of fresh vegetables as well as the contents of 1 large food storage module. Who knows, maybe the Chinese have got good results with hydroponic paksoi by now. There is an extremely steep "learning curve" but for all of humanity there should be a tremendous pay-off in learning "this is how complicated it is to create an artificial biosphere. So don't fuck up the one fully functional biosphere we have."
There are probably extremely important scientific results that are needed for this but are not accessible enough (e.g. just imagine that some Tchadian agricultural institute studied "how to grow salad and earth worms in sharp sand", I'm just making this example up).
Exactly the same issues came up in the BBC documentary series "Castaway 2000" on Taransay. And that was only for 1 year. It was truly fascinating television to watch.
Granted, on Mars they'd need something more advanced than a polytunnel for their veggies...
yet.. http://www.youtube.com/watch?v=neoUi4poCXI (WARNING: advertisement)
A few loose comments:
Very well put; although according to a previous poster, w.r.t. arctic ice we're at two sigma already, so (if I'm not making a stupid mistake and the data has a normal distribution) it's 4.6% likely that this year's values are normal, assuming a long-term stable arctic ice extent.
Well, there is a whole guild of people who specialize in determining the statistical pay-off between high-damage low-probability events (black swan?) and low-reward high-probability events (maximum squeezable insurance premium), and they're called actuaries . I'd like to know if, say, Lloyds can insure you against global warming affecting your business, and how much it would cost qua premium. What do the actuaries think?
Lastly a loose comment (not based on your essay): one of the global warming denialist talking points is "the necessary effort to stop global warming would bring our society back to the stone age, which is what YOU tree-hugging hippies want". You mentioned "putting the long view out of mind"; the long view is that, 300 years ago, we were before the "age of petroleum" and we didn't live in the stone age, either. (Although maybe in the whale oil age, I don't know). Who knows what the 21st century may bring technologically; maybe a mixture of high-tech low-energy-use equipment and low-tech low-energy-use 19th century technologies.
Yeah, and if Hitler were alive today, he'd probably be a "Warmist" as well, what with his vegetarianism and long walks in the woods.
Give me a break.
This whole Slashdot discussion today, based on a Daily Mail article, seems to be mental preparation of the public so that they're properly revved up for global warming denialism,
before the next IPCC report gets published in a few weeks.
So that on 2013-09-27, Joe Public will say to Jane Public: "but it's all rubbish; wasn't that in the newspaper a few weeks ago?".
That's an interesting argument, and apparently it took until 1956 until more work was done on that. I googled "CO2 extinction coefficient" and got: http://www.skepticalscience.com/saturated-co2-effect-advanced.htm. In short: the CO2 IR spectrum is not monochromatic, and there are plenty of wavelengths where the IR radiation still comes through and is dependent on concentration, so no.
You'll need to consult a real IR spectroscopist to study more, because this stuff is not trivial.
I do remember vaguely that there was this enormous C-O double bond stretch vibration band around 1800? reciproke centimeters. But it was a bit different in -COOH and aldehydes than in CO2 itself.
It's difficult to understand your rant.
... LOL!
So, you hate socialists. Fine. That's your opinion.
But then you go on about Obama and "leftist cabal" and "the socialist lot are pissing on the Constitution".
To me, this means that you are talking about USA politics and name-calling the USA Democrat party as "socialist".
That doesn't make sense in the normal way the word "socialist" is used. The USA Democrat party is very right-wing. The USA Republican party is "bat-shit crazy" extreme right-wing. We outside the USA almost never hear about the left-wing or socialist parties and politics of the USA.
I have heard that you have a Green Party, used to be chaired by the famous Ralph "Seat Belt" Nader, now run by a lady named Cynthia McKinney. She's probably left-wing.
But if you want to see real socialist parties in action (4.5 % of the parliament), read here:
http://www.guengl.eu/group/delegations (I was going to send a link to SYRIZA.gr but my Greek is so poor I couldn't even find an english language link)
Oh yeah, and "Slashdot the echo chamber of socialism"
Obviously, they want to block you from uploading those advanced ritual yoga videos...
I wasn't laughing at his/her nuclear terminology; on the Internet it's difficult to see if people are joking sometimes (it's called Poe's law) and especially on Slashdot people can sometimes say quite sick things. I thought he/she was referring to Hiroshima and Nagasaki. That is also a likely interpretation of the sentence "We have had a lot of experience with nukes", and my first reaction was "what a sick bastard!".
Do you like Harry Potter? I'll try to elaborate:
In "the Goblet of Fire", there is a scene in ch. 14 where the faux Mad-Eye Moody talks to Neville Longbottom in class:
No, I meant the clever Odysseus / Ballmer (ok.. bit difficult analogy here) couldn't destroy the walls of Troy / non-MS mobile ecosystem at Nokia, so he sent a gift wooden horse / ex-Microsoftie Elop to open up the gates / burn the existing platform, sell the essential GSM patents, reduce the value of Nokia so the rest can be bought up and sold in pieces. Then the triumphant Greeks went back to their own army.
I'm still a bit shocked that that story is not at the end of the Iliad (I really really thought it was), but ch. 24 ends with the mutilation, sale of the corpse to his dad, and burial of Hektor.
Wow.. what a mind-fuck.. I just browsed through my copy of the Iliad / Odyssey, and you're right--It's not mentioned.
Now I feel a bit stupid...
It's not at the beginning of the Odyssey, either.
I was thinking more about the Dead Kennedys song "Kinky Sex Makes the World go Round".
Mind you that song was from about a decade before the Iraq War.
That's right; we can't allow the Jolla-rowboat to escape Nokia's fate!
Sheesh, don't you know your classics? Re-read the Iliad, see what the clever Odysseus did. Hint: it required a lot of carpentry.
Do you know the scam game with three cups and a pea? Where you win money if you guess right under which cup the pea is after some shuffling?
The "Nokia basic essential mobile telecom patents" pea is no longer under the Nokia cup for some time now. It's now under the cup labelled "Vringo or I/P Engine". Watch where that cup moves and when the pea gets exchanged.
Here's an older article on Groklaw: http://www.groklaw.net/articlebasic.php?story=20130601134450374
The patents have already been sold apparently, to a "ringtone advertisement" company called Vringo Inc. which is now sueing mobile phone companies like ZTE for billions.
So who's going to buy Vringo? My guess would be Intellectual Ventures.
Here's a scary thought: maybe most of their trained personnel has already received the maximum lifetime dose and has been given their retirement already..
No, I meant more like: what if spent fuel storage pool 4 stuff leaks into the ground water before cleanup is ready for it. I wasn't being clear.
Maybe your sentence should be split in two; we know that basic research is a government task because it's almost certainly not profitable. On the other hand "opening new frontiers" could be profitable (Hilton hotel on Mars for the super-rich) but it carries so much risk that the taxpayers might revolt, e.g. European Green Parties's complaint against the ITER reactor is not that it is an abomination and the Ents should release the Rhône and flood and extinguish it, but rather that "nuclear fusion is still too far away a goal to spend taxpayer money on at the moment" (note the large difference in position towards nuclear fusion and nuclear fission reactors, here!).
Did that make sense? I'm tired..