Threads are supposed to have better performance, due to less cost when doing a context switch
The difference, if any, is minimal. A lot of the theory about content switch performance relies on outdated models of processors; note that from Core2 / Barcelona onwards, both Intel and AMD processors can store page table entries for multiple processes in their TLB and therefore do not need to flush them when the process changes. This means that theoretically, two small processes can perform as well as two threads in a single large process.
Where things *do* get interesting is in communication between those threads. Typically, a threaded model involves *more* communication (locking, etc), but each individual communication is more efficient (because it can use shared memory, rather than invoking kernel code).
Yeah. The problem is, how do you tell how much free RAM there actually is? The issue is clouded by two cases:
1. There is another program on the system that uses RAM because it's available (e.g. Photoshop) 2. There is a program on the system that relies on memory being available that isn't reported as in use (e.g. anything that uses memory mapped files, which includes most database systems)
So how hard is it to PDF the decision and upload it somewhere?
Right. Here in the UK, our supreme court publishes all of its decisions on its own web site, and they are typically up within a day or two. Why do other courts not do the same thing?
Yep. GNOME3 is horrible. I'm not a regular GNOME user, but just yesterday I happened to need a Linux system quick for some maintenance and the only thing handy was an Ubuntu 11 live CD. I hate to admit it, but I had to google to figure out how to get a terminal. As far as I could tell, there is no menu of applications, just a search interface...? What the hell happened to discoverability?
every company in the MMORPG business is scrambling to stay relevant or minimize losses. SWG will fold December 10
I thought it was Dec 15. But never mind, the reason isn't that they're scrambling to stay relevant (there's no way they can) or minimize losses (SWG is, from all accounts, still very profitable). The problem is that Lucas has pulled the IP licence rug out from under them, so they *aren't allowed to continue running it*.
Rift dropped the initial price of the game to a ridiculous $4.99
Probably less to do with competition from SWTOR than it is to try to remain relevant in a market that's steadily losing out to free-to-play colossi (LOTRO, DDO, Champions Online, soon Star Trek Online, Fallen Earth, APB Reloaded, etc.). When I can get another similar high-quality game and play large chunks of it for free, with only one-off purchases required to unlock the rest of the content in most cases, it seems penny-pinching for Trion to charge both an upfront fee for their second-tier game and for the time I play it. I'm honestly amazed they aren't offering the game free. Possibly with a month's subscription included. It would attract new subscribers, and subscribers is where their money comes from, not box sales.
No. The Scoville scale is not linear because it's based solely on human senses (which tend to be logarithmic).
So Bhut Jolokia may have 1/16th the Scoville Units but that does not necessarily translate to 1/16th the capsaicin content.
Actually, the Scoville scale is linear, because it is based on amount of dilution required before the flavour is not detectable (which removes the nonlinearity of human sensation).
1ppm capsaicin = 1 ASTA pungency unit. The conversion from there to Scovilles is explained in TFA, if you actually bothered reading it, but it *is* linear.
What English speaking country/region uses the long scale today?
While the wikipedia article may reflect current media & government use, unofficially the long scale has been more popular up until very recently in the UK. It was the method taught in schools up until (at least) the early 90s, and as such most British people are more comfortable with it than short scale.
Note, specifically, that the Oxford English Dictionary still suggests the long scale is more common:
(from the definition of 'billion')
In the 19th century, the U.S. adopted the French convention, but Britain retained the original and etymological use (to which France reverted in 1948). Since 1951 the U.S. value, a thousand millions, has been increasingly used in Britain, especially in technical writing and, more recently, in journalism; but the older sense ‘a million millions’ is still common
That's from the 4.0 CDROM edition, i.e. one that contains revisions up to 2008.
2^43, ie. far worse than any standard encryption algorithm in use today.
The best published cryptanalysis on AES256 requires 2^99.5 steps. While this will undoubtedly come down in future, it is unlikely to halve. Also, each of those steps is somewhat more complex than calculating a bit of PI and running one modular addition per bit of message.
I think we can be pretty sure that somebody with a PhD in astrophysics has at least a vague understanding of the fact that pi is transcendental. However, I see no reason to believe that simply because a number is transcendental it cannot contain some message. In fact, as far as I can see it, the set of transcendental numbers being infinite, it seems that there must therefore be at least one transcendental number that encodes any given message within any given number of digits of its beginning in any given base.
Anyway how would you determine, when looking at an infinitely long string of "random" numbers, what is a "message"?
For any given "message" you can calculate the probability of something containing at least that much order occuring within any given number of digits. If that probability is too low for what you have found in the region in which you found it, then it can be reasonably inferred that it is actually not a random occurrence.
No, that is not the case at all. If it were, this would be a different story.
We're talking here about third-party cookies.
Every browser I've used since some time around 2005 or so can be configured not to send third party cookies. Most come preconfigured not to do so; I know that some time around 2006 I had to rewrite some of my sites that used third party cookies (to track users across a single site that operated on multiple domains) because it simply didn't work for the majority of my visitors. Preventing yourself from being tracked by third party cookies is *trivially easy*.
If we're just talking about the referer[sic] header, then that's a different matter entirely, but I fail to see how it amounts to interception of communications. That's like saying that *any* web site that hosts images that other web sites are using is intercepting communications, because exactly the same thing is happening. OK, so they're ignoring them when they get to the other end (possibly; many sites log referrers automatically), but the law in question doesn't differentiate based on what you do with the information once you have it.
The word "infamous" is often used by scientists to describe problems that a large number of people have attempted to solve and have failed. I suspect that this is the sense the submitter was using it in.
Because in Europe, such a picture would be considered personal data pretty much everywhere, and thus it would fall under data protection laws that explicitly provide certain rights to individuals regarding personal data about them and certain limits on what anyone else may do with that data.
Yes. But those rights are limited: you can basically only prevent processing that is actively harmful to you. Which is why I phrased my question as I did. European law *does not* prevent Facebook storing information about you if they want to, as long as they do not use that information in a way that prejudices your legitimate interests. Kenja seemed to be arguing that they *should* be prevented from storing and/or processing an image of you without your permission, and I'd still like to know why...
There are plenty of fully functional 3D engines for Java [wikipedia.org].
You know what I don't see in that list: one of those engines that has been used for commercially successful games. With one notable exception: minecraft. A game renowned for the low quality of its graphics.
There may be a reason for this omission: they may not be good enough. (I don't know, I haven't actually tried using any of these, although I look at the descriptions of some of them and see that they are Java ports of C++ systems that I knew were second-rate systems five years ago).
That's probably only true if they bought it. If they silently obtain it from elsewhere (the producers don't even know about it), then I don't see how that is.
According to the *IAA, the latter people are doing the "industry" irreperable harm and will eventually kill it.
Good point. We should all pirate as much kiddie porn as we can, because doing so will eventually kill the kiddie porn industry...
Probably, yes. Unity is a fully-functional 3d engine. Implementing the features of such an engine in any of the above environments is at least hundreds of hours of work, and the result is unlikely to be as good. It would likely have tripled the required budget for this project, which was probably done on a shoestring.
Erm... no. They took DNA, a natural structure which is half of a system that can replicate (the other half being a collection of enzymes that can transcribe the DNA to RNA, and a ribosome that can take RNA templates and make enzymes, some of which can produce more DNA or ribosomes) and rearranged it into an entirely new structure that doesn't require the assistance of a ribosome to replicate itself.
Slashdot Mirror
Threads are supposed to have better performance, due to less cost when doing a context switch
The difference, if any, is minimal. A lot of the theory about content switch performance relies on outdated models of processors; note that from Core2 / Barcelona onwards, both Intel and AMD processors can store page table entries for multiple processes in their TLB and therefore do not need to flush them when the process changes. This means that theoretically, two small processes can perform as well as two threads in a single large process.
Where things *do* get interesting is in communication between those threads. Typically, a threaded model involves *more* communication (locking, etc), but each individual communication is more efficient (because it can use shared memory, rather than invoking kernel code).
Yeah. The problem is, how do you tell how much free RAM there actually is? The issue is clouded by two cases:
1. There is another program on the system that uses RAM because it's available (e.g. Photoshop)
2. There is a program on the system that relies on memory being available that isn't reported as in use (e.g. anything that uses memory mapped files, which includes most database systems)
So how hard is it to PDF the decision and upload it somewhere?
Right. Here in the UK, our supreme court publishes all of its decisions on its own web site, and they are typically up within a day or two. Why do other courts not do the same thing?
Yep. GNOME3 is horrible. I'm not a regular GNOME user, but just yesterday I happened to need a Linux system quick for some maintenance and the only thing handy was an Ubuntu 11 live CD. I hate to admit it, but I had to google to figure out how to get a terminal. As far as I could tell, there is no menu of applications, just a search interface...? What the hell happened to discoverability?
every company in the MMORPG business is scrambling to stay relevant or minimize losses. SWG will fold December 10
I thought it was Dec 15. But never mind, the reason isn't that they're scrambling to stay relevant (there's no way they can) or minimize losses (SWG is, from all accounts, still very profitable). The problem is that Lucas has pulled the IP licence rug out from under them, so they *aren't allowed to continue running it*.
Rift dropped the initial price of the game to a ridiculous $4.99
Probably less to do with competition from SWTOR than it is to try to remain relevant in a market that's steadily losing out to free-to-play colossi (LOTRO, DDO, Champions Online, soon Star Trek Online, Fallen Earth, APB Reloaded, etc.). When I can get another similar high-quality game and play large chunks of it for free, with only one-off purchases required to unlock the rest of the content in most cases, it seems penny-pinching for Trion to charge both an upfront fee for their second-tier game and for the time I play it. I'm honestly amazed they aren't offering the game free. Possibly with a month's subscription included. It would attract new subscribers, and subscribers is where their money comes from, not box sales.
No. The Scoville scale is not linear because it's based solely on human senses (which tend to be logarithmic).
So Bhut Jolokia may have 1/16th the Scoville Units but that does not necessarily translate to 1/16th the capsaicin content.
Actually, the Scoville scale is linear, because it is based on amount of dilution required before the flavour is not detectable (which removes the nonlinearity of human sensation).
1ppm capsaicin = 1 ASTA pungency unit. The conversion from there to Scovilles is explained in TFA, if you actually bothered reading it, but it *is* linear.
What English speaking country/region uses the long scale today?
While the wikipedia article may reflect current media & government use, unofficially the long scale has been more popular up until very recently in the UK. It was the method taught in schools up until (at least) the early 90s, and as such most British people are more comfortable with it than short scale.
Note, specifically, that the Oxford English Dictionary still suggests the long scale is more common:
(from the definition of 'billion')
That's from the 4.0 CDROM edition, i.e. one that contains revisions up to 2008.
2^43, ie. far worse than any standard encryption algorithm in use today.
The best published cryptanalysis on AES256 requires 2^99.5 steps. While this will undoubtedly come down in future, it is unlikely to halve. Also, each of those steps is somewhat more complex than calculating a bit of PI and running one modular addition per bit of message.
I think we can be pretty sure that somebody with a PhD in astrophysics has at least a vague understanding of the fact that pi is transcendental. However, I see no reason to believe that simply because a number is transcendental it cannot contain some message. In fact, as far as I can see it, the set of transcendental numbers being infinite, it seems that there must therefore be at least one transcendental number that encodes any given message within any given number of digits of its beginning in any given base.
Anyway how would you determine, when looking at an infinitely long string of "random" numbers, what is a "message"?
For any given "message" you can calculate the probability of something containing at least that much order occuring within any given number of digits. If that probability is too low for what you have found in the region in which you found it, then it can be reasonably inferred that it is actually not a random occurrence.
How irrational of me.
Get real.
No, that is not the case at all. If it were, this would be a different story.
We're talking here about third-party cookies.
Every browser I've used since some time around 2005 or so can be configured not to send third party cookies. Most come preconfigured not to do so; I know that some time around 2006 I had to rewrite some of my sites that used third party cookies (to track users across a single site that operated on multiple domains) because it simply didn't work for the majority of my visitors. Preventing yourself from being tracked by third party cookies is *trivially easy*.
If we're just talking about the referer[sic] header, then that's a different matter entirely, but I fail to see how it amounts to interception of communications. That's like saying that *any* web site that hosts images that other web sites are using is intercepting communications, because exactly the same thing is happening. OK, so they're ignoring them when they get to the other end (possibly; many sites log referrers automatically), but the law in question doesn't differentiate based on what you do with the information once you have it.
Have to say, that's the weirdest fucking chain mail I've ever read.
It's made of approximately 1,000 tonnes of lead with photodetectors dispersed through it.
The word "infamous" is often used by scientists to describe problems that a large number of people have attempted to solve and have failed. I suspect that this is the sense the submitter was using it in.
They have to make money in order to keep the business going dumbass.
You're confused. They need to make money in order to keep the business going. Profit is not a right, it's a reward.
You seem to be drawing a distinction between the phrases "need to [do something]" and "have to [do something]" which I don't think exists.
(source: OED CD-ROM edition 4.0)
Those are *extremely* similar, as far as I can see, and I really don't understand what you're trying to say by differentiating between them.
Because in Europe, such a picture would be considered personal data pretty much everywhere, and thus it would fall under data protection laws that explicitly provide certain rights to individuals regarding personal data about them and certain limits on what anyone else may do with that data.
Yes. But those rights are limited: you can basically only prevent processing that is actively harmful to you. Which is why I phrased my question as I did. European law *does not* prevent Facebook storing information about you if they want to, as long as they do not use that information in a way that prejudices your legitimate interests. Kenja seemed to be arguing that they *should* be prevented from storing and/or processing an image of you without your permission, and I'd still like to know why...
There are plenty of fully functional 3D engines for Java [wikipedia.org].
You know what I don't see in that list: one of those engines that has been used for commercially successful games. With one notable exception: minecraft. A game renowned for the low quality of its graphics.
There may be a reason for this omission: they may not be good enough. (I don't know, I haven't actually tried using any of these, although I look at the descriptions of some of them and see that they are Java ports of C++ systems that I knew were second-rate systems five years ago).
And many of them are open source, which would have saved NASA tons of cash in licensing costs alone.
Unity is free-as-in-beer, as long as you don't need certain advanced features (e.g. mobile clients).
That's probably only true if they bought it. If they silently obtain it from elsewhere (the producers don't even know about it), then I don't see how that is.
According to the *IAA, the latter people are doing the "industry" irreperable harm and will eventually kill it.
Good point. We should all pirate as much kiddie porn as we can, because doing so will eventually kill the kiddie porn industry...
And how would the company have found out about that anyway?
Theoretically, if he had, his requests would be in their access logs...
Probably, yes. Unity is a fully-functional 3d engine. Implementing the features of such an engine in any of the above environments is at least hundreds of hours of work, and the result is unlikely to be as good. It would likely have tripled the required budget for this project, which was probably done on a shoestring.
Also FTFA:
no biological components, particularly enzymes, are used in its execution
So no, not PCR or PCR-like, as such processes require enzymes.
Except, at least as I read the press release, it appears to be self-catalyzing.
Erm... no. They took DNA, a natural structure which is half of a system that can replicate (the other half being a collection of enzymes that can transcribe the DNA to RNA, and a ribosome that can take RNA templates and make enzymes, some of which can produce more DNA or ribosomes) and rearranged it into an entirely new structure that doesn't require the assistance of a ribosome to replicate itself.