The smoke in San Diego (Pacific Beach, for locals) this afternoon was just the right density to attenuate the sun just the right amount to let you see the two giant sunspots with the naked eye. Now that makes you feel small;) Now let me get back to hacking up a lung...
I dunno, I hit 70mph at around 1800rpm in fifth, though unless I'm cruising, I'll keep it in fourth (~2600rpm) so I have some quick power if I need it without downshifting. Fifth is only really much good above 90.
The care provider can require the patient to consent to their information being shared only "for treatment, payment, and health care operations". There's no way the care provider can fall back on a consent to protect them if patient info gets out into the wild.
Any other transfer of the information for any purpose other than treatment or payment requires an authorization that explicitly says what info is being given out, to whom, why, and how the receiver will safeguard the info. For the most part, you can't refuse treatment if a patient won't sign an authorization form.
For more info, please click on answer number four in this list. (sorry, the actual answer has a giant php address and would probably time out...)
"I never heard of Edisson so much as since I connect to slashdot." "One can argue that Edisson might be underestimated for a reason or another, but this is usually not the way of science, which is usually not country-centrist. I am really wondering at time if Edisson is not simply an overblown legend from the US, a bit like billy the kid, or whatever."
Perhaps the reson you've never heard of him is that you've been googling with a mis-spelled name?
1) Put into their terms of service that the user agrees to use the phone company as an agent for their $500 Title 47 small claims court actions against telemarketers, giving the phone company a 50% cut of any awarded damages
2) Give users a special dial code to call immediately after receiving a telemarketing call, like you can use *57 for harassing calls
3) Deliver the telemarketing companies a weekly invoice for their calls to cell phones
4) No, this isn't the stupid joke you thought it was, move along.
I doubt this'll be settled out of court (unless the idiots just drop it...) In the press release, they name the student and Princeton University as their targets. Princeton U.'s a private institution with more resources and lawyers than 10 SunnComms put together. Hope to see an IBM sized bitchslap headed their way;>
Yeah, but if you haven't notified the infringer, and can't in some other way show they wilfully infringed, you can only get your real damages-- if you notify them and they continue to infringe, you'll have an easy time showing wilfulness, and get treble damages plus attorney's fees. Plus, if they can show that you knew about their infringment but waited to inform them, they can reduce their damages further by showing that you failed to take reasonable steps to mitigate your loss. Overall not a good idea to know somebody's infringeing your patent and not at least serve them notice. You don't have to sue right away, but at least send a letter...
It's called the coefficient of thermal expansion, and given the difference between the daytime temp on Mars vs. Mercury or Venus, it'd have a measurable effect on the sundial reading (depending on what the dial was made of, 'course)
Sorry, I shouldn't have included CNS on that reference-- I assumed that the parallel routines had worked their way back in (I remember Kay Diederichs saying she had some || routines on cnsbb last year). CNX definitly uses MP very well on linux machines. I'd rather use the free version as well, but Accelrys would sue me into the stone age if I tried;>
Anyway, traditional refinement/ manual fitting loops aren't amenable to distributed computing, but you could do something along the line of giving each client a current model and F's, having it calculate a map, do an automated real space fit on a selected small part of the model (along the lines of what Quanta does), dump it into SA refinement, calculate R's and Rfrees, along with real space fit of the modified portion of the model, and report the changes to the model along with the Rs back to the server. Every few days, pick 10 of the best new models and present them and their maps to the crystallographer who would accept or reject the models.
I know these ideas have been kicked around before, but the biggest challenge is getting people to distribute their F's and initial models;>. Maybe a secure client that encrypts everything, once Palladium rolls out....
(just kidding, for the humor impaired mods. And $Diety help you if you're reading this deep into this thread and not a crystallographer...)
"These structures were all at around 3-Angstrom resolution, and once you're able to collect and properly phase X-ray data of that quality, the global structure is obvious."
Having fitted a number of experimental 3A maps, I can unequivocally say that the global structure is far from obvious;>. The phases are usually crap, you can't see sidechains, or tell which direction the chain is running. Anything that's not helix looks like a sausage with dysentary. Give me a nice 1A map of a small soluble enzyme any day (of course, membrane protein work is what pays the bills, but what the hey...)
the software that does this is single-processor...
CNS/CNX (the current versions of the molecular dynamics refinement program) works very well on multiprocessor systems, using OpenMP. As long as you don't run out of memory, SHELX (a non-MD based refinement program that uses non-linear least squares) uses multiple processors very well-- the computation time scales inversly with the number of procs you use. And refinement of large structures is still slow, mainly because resolution on these keeps getting better, and bigger proteins are being crystallized (in bigger unit cells) so you have more data points. And when you add in ADPs, the number of parameters you're fitting doubles, so that makes it a bit of a pain, too;>
Even with the most advanced, powerful folding techniques, I've yet to hear of a single case where a de novo computationally folded protein was close enough to fit the x-ray data in molecular replacement, which is probably the best test of whether the theoretical structure matches the observed structure-- a quick pub med search found nothing, either. Even NMR structures rarely work in molecular replacement (an NMR structure is basically a theoretical folding experiment driven by a massive number of experimentally determined atom-atom distance and angle constraints).
As far as I can tell, the molecular structure of a protein is pretty much exactly how that protein fills up 3-dimensional space. Of course, people draw all sorts of unfounded conclusions from structures that have nothing to do with the observed data. Sorry to disappoint, but it sounds like your friend worked on integrins (the proteins most involved in binding/anchoring cells to the extracelluar matrix), and as far as I know MacKinnon's lab hasn't worked on that family.
This isn't to say that de novo folding experiments like folding@home aren't important, they are. But, by and large, not to experimental structure determination work.
Have you tried using WinXP on a mobile that's hopping wireless networks several times a day? My one complaint with Win2k was the pitiful way laptops handled bouncing around different networks. XP has no problems moving around; that alone has made it worth switching for mobiles.
Don't you own and operate your site at the sufferance of your ISP/upstream, and according to their terms of service? By your definition, it seems that the 1st amendment doesn't apply to you, either.
The 1st amendment only restricts activities by the government, not private individuals or corporations. It protects you and those using your mailserver equally. Of course, post patriot-act, you've got equally zero protection to your speach, but hey...
The parent didn't say it was about artist's rights, just that it's not about money. More correctly, it's not about the settlement money. It's about running a few kids into bankruptcy (and hence about a decade of financial ruin) and getting those cases well publicized. Their hope is that this'll scare their peers into not sharing-- and when enough users stop sharing, the system breaks down.
Of course, you have to question how their logic bridges the gap between stopping sharing and getting college students to each shell out $20 for all the CDs they want. It's pretty much the legendary step 2: ???
They could care less how much money they actually get from the student(s) they target. If they destroy a few kids' lives for uploading (not downloading), far more people will turn into leeches on the p2p networks. By increasing the number of leeches and simultaneously flooding the networks with false files, they're in effect causing a massive DOS on the p2p networks, making them far less useful than they had been. Their goal is to make it cost more (in terms of time and frustration) to download the song than it does to buy the album-- if you're making $10 an hour and it takes you more than 2 hours to download an album, then they've won.
Have you seen Goodman's paper from last year? "The serotonergic system and mysticism: could LSD and the nondrug-induced mystical experience share common neural mechanisms?" J Psychoactive Drugs. 2002 Jul-Sep;34(3):263-72.
A bit wacky, but a good read if you're into behavioral neurobiology...
Slashdot Mirror
As long as the other phone's set to be discoverable.... which is why mine isn't.
Come on, everybody knows dinosaurs got cancer from smoking. Duh.
The smoke in San Diego (Pacific Beach, for locals) this afternoon was just the right density to attenuate the sun just the right amount to let you see the two giant sunspots with the naked eye. Now that makes you feel small ;) Now let me get back to hacking up a lung...
Since it's exponential, Mach 5 would be warp -16. (e^-16=0.000000112). Under the TOS scale. Gawd what geeks we are.
I dunno, I hit 70mph at around 1800rpm in fifth, though unless I'm cruising, I'll keep it in fourth (~2600rpm) so I have some quick power if I need it without downshifting. Fifth is only really much good above 90.
;).
And that's in a car with a cast-iron block
I wonder how many folks here would be bitching out LG if it was XP that was trashing the hardware?
The care provider can require the patient to consent to their information being shared only "for treatment, payment, and health care operations". There's no way the care provider can fall back on a consent to protect them if patient info gets out into the wild.
Any other transfer of the information for any purpose other than treatment or payment requires an authorization that explicitly says what info is being given out, to whom, why, and how the receiver will safeguard the info. For the most part, you can't refuse treatment if a patient won't sign an authorization form.
For more info, please click on answer number four in this list. (sorry, the actual answer has a giant php address and would probably time out...)
Don't you think step two should precede, rather than follow, step one? I mean, ouch...
Probably true, but $50 mil is a rounding error for a company with a market cap of 154 billion. Remember, total assets:
SCO << $Diety < IBM
IBM still wins
"I never heard of Edisson so much as since I connect to slashdot."
"One can argue that Edisson might be underestimated for a reason or another, but this is usually not the way of science, which is usually not country-centrist. I am really wondering at time if Edisson is not simply an overblown legend from the US, a bit like billy the kid, or whatever."
Perhaps the reson you've never heard of him is that you've been googling with a mis-spelled name?
1) Put into their terms of service that the user agrees to use the phone company as an agent for their $500 Title 47 small claims court actions against telemarketers, giving the phone company a 50% cut of any awarded damages
2) Give users a special dial code to call immediately after receiving a telemarketing call, like you can use *57 for harassing calls
3) Deliver the telemarketing companies a weekly invoice for their calls to cell phones
4) No, this isn't the stupid joke you thought it was, move along.
Right here.
I doubt this'll be settled out of court (unless the idiots just drop it...) In the press release, they name the student and Princeton University as their targets. Princeton U.'s a private institution with more resources and lawyers than 10 SunnComms put together. Hope to see an IBM sized bitchslap headed their way ;>
Yeah, but if you haven't notified the infringer, and can't in some other way show they wilfully infringed, you can only get your real damages-- if you notify them and they continue to infringe, you'll have an easy time showing wilfulness, and get treble damages plus attorney's fees. Plus, if they can show that you knew about their infringment but waited to inform them, they can reduce their damages further by showing that you failed to take reasonable steps to mitigate your loss. Overall not a good idea to know somebody's infringeing your patent and not at least serve them notice. You don't have to sue right away, but at least send a letter...
It's called the coefficient of thermal expansion, and given the difference between the daytime temp on Mars vs. Mercury or Venus, it'd have a measurable effect on the sundial reading (depending on what the dial was made of, 'course)
"Kay Diederichs saying she had some || routines on cnsbb last year"
;>
Damn typos.
"Kay Diederichs saying he had some || routines on cnsbb last year"
Sorry Kay
Sorry, I shouldn't have included CNS on that reference-- I assumed that the parallel routines had worked their way back in (I remember Kay Diederichs saying she had some || routines on cnsbb last year). CNX definitly uses MP very well on linux machines. I'd rather use the free version as well, but Accelrys would sue me into the stone age if I tried ;>
;>. Maybe a secure client that encrypts everything, once Palladium rolls out....
Anyway, traditional refinement/ manual fitting loops aren't amenable to distributed computing, but you could do something along the line of giving each client a current model and F's, having it calculate a map, do an automated real space fit on a selected small part of the model (along the lines of what Quanta does), dump it into SA refinement, calculate R's and Rfrees, along with real space fit of the modified portion of the model, and report the changes to the model along with the Rs back to the server. Every few days, pick 10 of the best new models and present them and their maps to the crystallographer who would accept or reject the models.
I know these ideas have been kicked around before, but the biggest challenge is getting people to distribute their F's and initial models
(just kidding, for the humor impaired mods. And $Diety help you if you're reading this deep into this thread and not a crystallographer...)
"These structures were all at around 3-Angstrom resolution, and once you're able to collect and properly phase X-ray data of that quality, the global structure is obvious."
;>. The phases are usually crap, you can't see sidechains, or tell which direction the chain is running. Anything that's not helix looks like a sausage with dysentary. Give me a nice 1A map of a small soluble enzyme any day (of course, membrane protein work is what pays the bills, but what the hey...)
...
;>
Having fitted a number of experimental 3A maps, I can unequivocally say that the global structure is far from obvious
the software that does this is single-processor
CNS/CNX (the current versions of the molecular dynamics refinement program) works very well on multiprocessor systems, using OpenMP. As long as you don't run out of memory, SHELX (a non-MD based refinement program that uses non-linear least squares) uses multiple processors very well-- the computation time scales inversly with the number of procs you use. And refinement of large structures is still slow, mainly because resolution on these keeps getting better, and bigger proteins are being crystallized (in bigger unit cells) so you have more data points. And when you add in ADPs, the number of parameters you're fitting doubles, so that makes it a bit of a pain, too
I'd take that bet.
Even with the most advanced, powerful folding techniques, I've yet to hear of a single case where a de novo computationally folded protein was close enough to fit the x-ray data in molecular replacement, which is probably the best test of whether the theoretical structure matches the observed structure-- a quick pub med search found nothing, either. Even NMR structures rarely work in molecular replacement (an NMR structure is basically a theoretical folding experiment driven by a massive number of experimentally determined atom-atom distance and angle constraints).
As far as I can tell, the molecular structure of a protein is pretty much exactly how that protein fills up 3-dimensional space. Of course, people draw all sorts of unfounded conclusions from structures that have nothing to do with the observed data. Sorry to disappoint, but it sounds like your friend worked on integrins (the proteins most involved in binding/anchoring cells to the extracelluar matrix), and as far as I know MacKinnon's lab hasn't worked on that family.
This isn't to say that de novo folding experiments like folding@home aren't important, they are. But, by and large, not to experimental structure determination work.
Have you tried using WinXP on a mobile that's hopping wireless networks several times a day? My one complaint with Win2k was the pitiful way laptops handled bouncing around different networks. XP has no problems moving around; that alone has made it worth switching for mobiles.
Don't you own and operate your site at the sufferance of your ISP/upstream, and according to their terms of service? By your definition, it seems that the 1st amendment doesn't apply to you, either.
The 1st amendment only restricts activities by the government, not private individuals or corporations. It protects you and those using your mailserver equally. Of course, post patriot-act, you've got equally zero protection to your speach, but hey...
The parent didn't say it was about artist's rights, just that it's not about money. More correctly, it's not about the settlement money. It's about running a few kids into bankruptcy (and hence about a decade of financial ruin) and getting those cases well publicized. Their hope is that this'll scare their peers into not sharing-- and when enough users stop sharing, the system breaks down.
Of course, you have to question how their logic bridges the gap between stopping sharing and getting college students to each shell out $20 for all the CDs they want. It's pretty much the legendary step 2: ???
They could care less how much money they actually get from the student(s) they target. If they destroy a few kids' lives for uploading (not downloading), far more people will turn into leeches on the p2p networks. By increasing the number of leeches and simultaneously flooding the networks with false files, they're in effect causing a massive DOS on the p2p networks, making them far less useful than they had been. Their goal is to make it cost more (in terms of time and frustration) to download the song than it does to buy the album-- if you're making $10 an hour and it takes you more than 2 hours to download an album, then they've won.
Of course, MDMA is a TON more neurotoxic than either LSD or mystical experiences (e.g. Green et al., Pharmacol Rev. 2003 Sep;55(3):463-508).
;>
So I guess the question of which is worse: flashbacks or neuronal degeneration
Have you seen Goodman's paper from last year?
"The serotonergic system and mysticism: could LSD and the nondrug-induced mystical experience share common neural mechanisms?" J Psychoactive Drugs. 2002 Jul-Sep;34(3):263-72.
A bit wacky, but a good read if you're into behavioral neurobiology...