You're asking a couple distinct, and reasonable, questions. About "blind testing" -- I don't know the details for this particular result, but particle physicists put quite a bit of effort into making sure that they aren't fooling themselves. One of the best ways of doing it is so-called "blind analysis". The idea there is to define your entire data analysis strategy based solely on simulated data. There are pretty good simulations available of both the expected backgrounds, and of the process you are trying to actually find (the signal). So you define all of the methods you are going to use using these simulations before you look at the data. This ensures that you don't bias yourself into "finding something" in the data that isn't really there. (I don't know if a strict blinding procedure was used for this analysis, but likely something similar was done.)
The formal peer review system will come into effect now that the result is submitted to a journal. The paper will be distributed to some anonymous referees who will try to judge the merits of the physics and decide whether it merits publication. But I should note that the peer review process in modern particle physics actually starts long before the result is made public. Although there are only 3 or 4 main analysts, the paper is signed by the entire 3000 person CMS Collaboration (of which I am a member). So we have a very stringent internal review process to ensure that the result is sound before we release it with 3000 names taking responsibility. That doesn't mean that particle physics collaborations never make mistakes, but it does mean that results are scrutinized by a number of more or less unbiased eyes before they are made public.
I agree with the AC that it is wrong to call it a "statistical" effect. I don't think it has well defined statistical properties. The margin of error on a poll is a well-defined quantity based on the sample size of a poll. Here they're trying to quantify poorly-understood effects like how often people make dumb mistakes (write unclearly, misplace a stack of ballots, etc). This is more akin to a systematic uncertainty than a statistical one.
Now let the real fight, over the 2012 budget, start.
The problem with all of these proposals is that nobody can get over ideology enough to actually hammer out how to solve fiscal problems. This fight got stuck on a piddling amount of money for Planned Parenthood. There was also a bunch of wrangling about the EPA.
I expect more of the same when it comes to arguing about the Ryan plan. He's started things in the wrong direction already, by wanting to cut taxes on the rich and turn Medicare into a block grant program. And all of the really ugly details ala Planned Parenthood and the EPA aren't even in his proposal. He just says that discretionary spending is going to be cut, but doesn't say how. So there will be more fights like we just saw.
Sad indeed (to the point where I feel guilty for using xemacs....). But it doesn't strike me as something that somebody would say because they haven't been appreciated enough. Rather, it sounds like he's clinically depressed. When you're that sad, it's not for a logical reason....
Somebody should mod the parent up (red=ECAL, blue=HCAL). The color legend is confusing because some colors are repeated. I think that's what got the guy confused who originally posted the link.
The bundling together of unrelated discoveries is weird, and I feel like it diminishes the impact of the prizes a bit in the public eye. (Instead of explaining one seminal discovery to the public, you have to explain two, and make it clear that they are not even related.) If anything, Nambu should have received his own prize, and then KM could have shared one with Cabibbo. But there are only a finite number of years, and particle physics only gets a prize at all every few years, so it is hard to reward all of the deserving subjects (if not people) without this type of bundling.
2. digital backs for medium format cameras in the 22-35MP range
Cost? Compared to buying and storing pricy film for 30 years, I think the answer is obvious.
In 5 years we'll have disposible 16MP cameras and the pro cameras will be 100MP.
The number of pixels is (nearly) irrelevant to image quality. What matters is the physical size sensor, and unfortunately Moore's law doesn't help you there. Squeezing more and more pixels into the same size sensor eventually reaches a point where the main source of noise is photon statistics, and there is little point to squeezing still more pixels into that same area.
"This Slate article is crap" -- yep, the author's former advisor (Nima Arkani-Hamed at Harvard) has now come out and said so: From the author's blog entry on the article:
Nima Arkani-Hamed says: June 6th, 2007 at 2:52 am
Hi James,
Someone pointed out your slate article to me, and I have to say as your ex-undergrad advisor I was very disappointed. You got the physics 100% wrong, (which is all the more surprising to me given that I know you understand some subtle physics rather well). As some of the commenters above indicated, if there is anything at all to the D0 rumor, there is absolutely no way it can be the standard model higgs. It would instead be a remarkable indication of physics *beyond* the standard model, and would lead us to expect much more at the LHC, not less. One possible explanation might indeed be an MSSM higgs, though until there is an actual D0 paper and the details of the analysis are known one can only speculate; there is still a real chance that (like other significant excesses in the past) it will go away. BTW you say above that an MSSM higgs at 180 GeV isn't plausible, but that is because you misunderstand: it is the SECOND Higgs of the MSSM, the one that is not the Standard Model Higgs, and 180 GeV is certainly OK for it's mass. Your assertions to the contrary don't change this fact; saying something with confidence doesn't make it true.
To add insult to (your) injury, if the D0 rumor turns out to be real, not only would we think that the LHC is more likely to see lots of extra new physics, even beyond confirming this second higgs, the LHC would also still be set to see the actual Standard Model higgs itself! So the entire logical structure of your article is completely wrong, indeed almost exactly the opposite of the truth.
Of course in this bloggy, postmodern day and age, where people routinely pontificate on things they know nothing about, I suppose a little wild inaccuracy about physics in slate is a drop in the bucket. It is nonetheless irksome to me that the work of hundreds of amazingly talented experimental particle physicists gets characterized in such a profoundly incorrect way. They are charged now more than ever with moving fundamental physics forward, and are working around the clock to make it happen. They deserve a little fact-checking when you write about them. I strongly encourage you to do everything you can to set the record straight in this matter.
Nima
PS I will not be checking back here; I detest the blogosphere for reasons that this little fiasco make completely self-evident, and I have already wasted more time on this than it deserves. However I did feel the need to write something about it, especially since you invoked me as your ex-advisor, lest anyone get the impression that I condone this type of shoddy work in any way.
I don't understand the D40/D40x bashing due to the lack of an autofocus motor. The market for the camera is not people who have a closet of Nikon lenses, nor is it people who will scour eBay for old lenses. The market is people who want to upgrade from point-and-shoot cameras because they want the option of changing lenses and they want to get away from the limitations of a tiny image sensor. There are plenty of AF-S lenses to satisfy this market.
The exception is in prime lenses, notably the cheap and useful 50mm f/1.8D. Nikon clearly thinks consumers are uninterested in these and I think they are making a mistake. But I don't think this issue alone makes the D40 a crappy camera.
How can two Nikons rank higher than the one Canon they put on the list? They obviously forgot about the 1D MkIII. Seriously, a camera that shoots 10 10-megapixel pictures in one second deserves to be on the list.
I don't think the 1D MkIII (a pro-level camera) is the market that PC World is interested in. I am surprised to see that the Canon 30D made the list but the 5D did not. (Since the 5D has brought a full-frame sensor to the "consumer" market.) They also put the Nikon D40x above the Nikon D80. This is arguably reasonable but indicates strongly that price plays a big role in their listing. So clearly their list is aimed at the electronics consumer who wants a spiffy new dSLR, not the hobbyist photographer. Hobbyist photographers don't read PC World for camera advice.
...that robot/space telescope exploration gets you a lot more bang for the buck than trying to put a man back on the moon. Hopefully the next President will kill off this return to the moon business and start putting money into stuff like this again.
Actually there is a blatant mistake in the NYTimes article. It says that collisions wil happen 30 million times a second.
"Again and again and again -- 30 million times a second, in fact."
Nope. The LHC runs at 40MHz.... A number that is absolutely hard-coded into the design and can not be changed... Wrote an e-mail to the NYTimes. They are generally pretty good with correcting in due time.
Are you sure this is wrong? Not every bucket has to be filled, right? An LHC PDF document says that there are 2808 bunches in the 26659 meter ring at peak luminosity.
Using Google as my calculator:
1 / (((26 658.883 m) / c) / 2 808) = 31.5773629 megahertz
or about 30 million collisions per second. But I didn't think about this too hard, so maybe you can point out my mistake.
But I don't think that this is the end of particle physics. It just might delay it.
As a young person in the field, I would agree. But it would be a very depressing delay.....
If LHC fails to find physics beyond the Standard Model (or worse, does not even find the Higgs), then the ILC is on shaky ground. If the LHC really plays out as a disappointment (due to Nature or technical failure) and the ILC is cancelled, then particle physicists will run to astrophysics and neutrino physics in droves. A lot of attention would shift to learning about dark matter from space instead of from colliders. Collider physics would slow down for a long time, and the future might depend on how quickly really advanced accelerator technologies can be developed.
Both the NYT and New Yorker articles were pretty good. As an experimentalist, I was slightly perturbed at the rant from Nima Arkani-Hamed in the New Yorker piece. Paraphrasing, he basically says, "The experimentalists don't let us see their raw data," and then later says, "Theorists do the important work anyway by drawing broad conclusions rather than just looking at individual observations." But in fairness, the author of the piece also included some anti-theorist quotes from Sam Ting and Leon Lederman. So I guess it balances out.
Anyway, it is good to see experimental particle physics get some media attention.
White dwarves were aged around 16 Gyr, so my impression is that the dating technique was off by 3 Gyr at least, very iffy to say the least.
I'm sure you're right about this. Traditionally a lot of these dating methods had big errors (they probably still do). At the level I understand it, an uncertainty as small as a billion years (~10%) is good by astronomy standards. Also, I seem to recall from my cosmology class that the measurements of the Hubble parameter fluctuated a lot over the years. But the key change is the interpretation of the measurements--how you translate them to age depends critically on what values you assume for Omega_Lambda, etc. People used to leave dark energy out of the calculation, which gave a very different answer than we get today.
Now a bit of a tangent, while not straying from the subject:
I remember in the mid-nineties there was a strange snafu in the world of astrophysics, as the apparent age of globular clusters made them older than the Universe itself! As it turned out, of course, the star dating technique was wrong and off by a couple of billion years.
Just how fined-tuned is this new dating technique?
Actually, I think the resolution of this problem was on the other side. The old estimates of 8 or 9 billion years for the age of the universe came from calculations where it was assumed that the universe was matter-dominated. This was indeed a problem since the globular cluster measurements were in the >10 billion year range. Thanks to WMAP and others we now know that the universe is dark-energy dominated, and the age estimate of the universe has been nailed down with relative precision at 13.7 billion years (with an error of ~2%).
From the paper, it looks like this age measurement has O(10%) uncertainties. The authors don't even try to create an overall error on the measurement, instead giving a large table of various uncertainties. (Note: I'm not an astrophysicist, so I'm not used to looking at this style of paper.) From the paper: "Despite their large uncertainties the age limits provided by HE 1523-0901 and CS 31082-001 are in good agreement with the WMAP result of 13.7 Gyr for the age of the Universe." I would have phrased it replacing "Despite" with "Within", but that's just semantics.
From the ApJ website: Discovery of a Ring-Like Dark Matter Structure in the Core of the Galaxy Cluster CL0024+17 M. J. Jee, H. C. Ford, G. D. Illingworth, R. L. White, T. J. Broadhurst, D. A. Coe, G. R. Meurer, A. van der Wel, N. Benitez, J. P. Blakeslee, R. J. Bouwens, L. D. Bradley, R. Demarco, N. L. Homeier, A. R. Martel, And S. Mei Received: 06 Sep 2006 Accepted: 02 Mar 2007 Dr. Myungkook Jee, Department of Physics and Astronomy, John Hopkins University, 3400 North Charles Street, Baltimore, MD 21218-2686, USA (mkjee@pha.jhu.edu)
There's no link to a preprint and no preprint on the arxiv (that I can find). I'm guessing that they're embargoing it until it gets published?
I'm fairly certain this all just the NASA PR machine anyway. Another little piece of evidence to support what we all know: non-baryonic dark matter exists. I'm sure it's a nice piece of science, but I'm nots sure why it's on slashdot.
I'm a current BaBarian, so let me try to fix this up a bit....
The essence of the question that BaBar and Belle were trying to answer is "Is CP violated in strong interactions?".
Not quite. We are investigating CP violation in the weak interaction. CP is conserved in the strong interaction.
It generally takes several years of university physics just to understand the question.
I hope not. In one paragraph: There is matter and there is antimatter. Astronomers tell us that the universe is completely made of matter. Particle physicists have a very well tested theory (The Standard Model) predicting how matter and antimatter behave. Using this theory, we predict that there should have been about the same amount of antimatter as matter made in the Big Bang, leading to a whole lot of nothing (i.e. we wouldn't exist). This is a bad contradiction, so we want to see whether the Standard Model's predictions about differences between matter and antimatter (CP violation) hold up.
Incidentally, the answer is "yes, maximally".
Although C and P individually are maximally violated by the weak interaction, the combination CP is nearly conserved. And the CP violation we measure at BaBar and Belle is exactly what is predicted by The Standard Model (i.e. not enough to explain the matter-dominated universe).
So we continue to search for new sources of CP violation, because something doesn't add up.
If you download the linked paper, on the second page they talk about the Breit-Wigner (Cauchy) density psi, and later they claim that their score process has zero expectation. Now, everyone knows that the Breit-Wigner does not *have* an expectation, and it's often used as an example where the asymptotic normal (Gaussian) distribution approximation doesn't hold. But still, they derive all sorts of distribution formulas involving a chi squared and a Gaussian process, as if there was no problem at all with the Breit-Wigner tails.
They use a Breit-Wigner because that's often a realistic model of the signal distribution, when one is talking about resonance production in a particle physics experiment. (My copy is at work, but I know this is discussed, for example, in Sakurai's Modern Quantum Mechanics.) I don't think this paper nearly lived up to the press release, and certainly isn't germane to Slashdot, but I don't think the use of a BW has anything to do with it.
On the other hand, I'm merely a particle physics grad student, and I didn't even attempt to read the center of the paper. If they really did come up with something that has more power than chi^2 (at least for an extremely simple fit) then that is notable. What would be really interesting would be for someone to come up with a real goodness-of-fit statistic for unbinned fits.
> > A favoured theory is that the dark matter
> > consists of Wimps (weakly interacting massive
> > particles) about a thousand times more massive
> > than a proton
> My training in physics is quite elementary, but
> I was led to believe the proton is relatively
> massive on the atomic level, especially when
> compared to an electron. How could a wimp be so
> large and yet unnoticed?
The key is the "weakly interacting" in the name. At the microscopic level, these particles (if they exist) can only interact via the weak force, which is both weak and short-range.
In particle physics the size of a particle has no relation to a physical size or a particle's mass. It is defined in terms of how strongly a particle interacts with other matter. (See the definition of cross section at PhysicsWorld.) So since the WIMP particle interacts only weakly, it is by definition "small," even if it is massive.
If the WIMP hypothesis is correct, then the WIMPs have hardly been "unnoticed." One of the chief motivations for looking for them is to explain the rotation of various galaxies which appear to be much more massive than can be calculated by adding up the mass of all the stars and dust in them. So if this missing mass does consist of WIMPs, then they have already been noticed!
Slashdot Mirror
You're asking a couple distinct, and reasonable, questions. About "blind testing" -- I don't know the details for this particular result, but particle physicists put quite a bit of effort into making sure that they aren't fooling themselves. One of the best ways of doing it is so-called "blind analysis". The idea there is to define your entire data analysis strategy based solely on simulated data. There are pretty good simulations available of both the expected backgrounds, and of the process you are trying to actually find (the signal). So you define all of the methods you are going to use using these simulations before you look at the data. This ensures that you don't bias yourself into "finding something" in the data that isn't really there. (I don't know if a strict blinding procedure was used for this analysis, but likely something similar was done.)
The formal peer review system will come into effect now that the result is submitted to a journal. The paper will be distributed to some anonymous referees who will try to judge the merits of the physics and decide whether it merits publication. But I should note that the peer review process in modern particle physics actually starts long before the result is made public. Although there are only 3 or 4 main analysts, the paper is signed by the entire 3000 person CMS Collaboration (of which I am a member). So we have a very stringent internal review process to ensure that the result is sound before we release it with 3000 names taking responsibility. That doesn't mean that particle physics collaborations never make mistakes, but it does mean that results are scrutinized by a number of more or less unbiased eyes before they are made public.
I agree with the AC that it is wrong to call it a "statistical" effect. I don't think it has well defined statistical properties. The margin of error on a poll is a well-defined quantity based on the sample size of a poll. Here they're trying to quantify poorly-understood effects like how often people make dumb mistakes (write unclearly, misplace a stack of ballots, etc). This is more akin to a systematic uncertainty than a statistical one.
Because the results are not binding anyway, there's no need for a recount, or so the NYTimes says:
http://thecaucus.blogs.nytimes.com/2012/01/04/no-need-for-recount-in-iowa-caucus/?scp=1&sq=iowa%20recount&st=cse
Not sure about your other points, but the point about eliminating the Bush tax cuts is on the money:
http://www.washingtonpost.com/blogs/ezra-klein/post/how-to-halve-the-deficit-by-doing-nothing/2011/03/25/AFXb0RoB_blog.html
Now let the real fight, over the 2012 budget, start.
The problem with all of these proposals is that nobody can get over ideology enough to actually hammer out how to solve fiscal problems. This fight got stuck on a piddling amount of money for Planned Parenthood. There was also a bunch of wrangling about the EPA.
I expect more of the same when it comes to arguing about the Ryan plan. He's started things in the wrong direction already, by wanting to cut taxes on the rich and turn Medicare into a block grant program. And all of the really ugly details ala Planned Parenthood and the EPA aren't even in his proposal. He just says that discretionary spending is going to be cut, but doesn't say how. So there will be more fights like we just saw.
NPR got some experts to use various techniques to analyze the flow. They came up with numbers around a factor of 10 higher than the 5000 bpd estimate.
http://www.npr.org/templates/story/story.php?storyId=126809525
Sad indeed (to the point where I feel guilty for using xemacs....). But it doesn't strike me as something that somebody would say because they haven't been appreciated enough. Rather, it sounds like he's clinically depressed. When you're that sad, it's not for a logical reason....
A Google search on:
gives exactly the bogus answer that Microsoft used in the top hit.
Unfortunately for Microsoft, a bing search gives the same top hit.
Somebody should mod the parent up (red=ECAL, blue=HCAL). The color legend is confusing because some colors are repeated. I think that's what got the guy confused who originally posted the link.
I disagree
The bundling together of unrelated discoveries is weird, and I feel like it diminishes the impact of the prizes a bit in the public eye. (Instead of explaining one seminal discovery to the public, you have to explain two, and make it clear that they are not even related.) If anything, Nambu should have received his own prize, and then KM could have shared one with Cabibbo. But there are only a finite number of years, and particle physics only gets a prize at all every few years, so it is hard to reward all of the deserving subjects (if not people) without this type of bundling.
The number of pixels is (nearly) irrelevant to image quality. What matters is the physical size
sensor, and unfortunately Moore's law doesn't help you there. Squeezing more and more pixels
into the same size sensor eventually reaches a point where the main source of noise is photon
statistics, and there is little point to squeezing still more pixels into that same area.
From the author's blog entry on the article:
I don't understand the D40/D40x bashing due to the lack of an autofocus motor. The market for the camera is not people who have a closet of Nikon lenses, nor is it people who will scour eBay for old lenses. The market is people who want to upgrade from point-and-shoot cameras because they want the option of changing lenses and they want to get away from the limitations of a tiny image sensor. There are plenty of AF-S lenses to satisfy this market.
The exception is in prime lenses, notably the cheap and useful 50mm f/1.8D. Nikon clearly thinks consumers are uninterested in these and I think they are making a mistake. But I don't think this issue alone makes the D40 a crappy camera.
I don't think the 1D MkIII (a pro-level camera) is the market that PC World is interested in. I am surprised to see that the Canon 30D made the list but the 5D did not. (Since the 5D has brought a full-frame sensor to the "consumer" market.) They also put the Nikon D40x above the Nikon D80. This is arguably reasonable but indicates strongly that price plays a big role in their listing. So clearly their list is aimed at the electronics consumer who wants a spiffy new dSLR, not the hobbyist photographer. Hobbyist photographers don't read PC World for camera advice.
...that robot/space telescope exploration gets you a lot more bang for the buck than trying to put a man back on the moon. Hopefully the next President will kill off this return to the moon business and start putting money into stuff like this again.
Using Google as my calculator:
1 / (((26 658.883 m) / c) / 2 808) = 31.5773629 megahertz
or about 30 million collisions per second. But I didn't think about this too hard, so maybe you can point out my mistake.
As a young person in the field, I would agree. But it would be a very depressing delay.....
If LHC fails to find physics beyond the Standard Model (or worse, does not even find the Higgs), then the ILC is on shaky ground. If the LHC really plays out as a disappointment (due to Nature or technical failure) and the ILC is cancelled, then particle physicists will run to astrophysics and neutrino physics in droves. A lot of attention would shift to learning about dark matter from space instead of from colliders. Collider physics would slow down for a long time, and the future might depend on how quickly really advanced accelerator technologies can be developed.
Let's hope for a rosier scenario.
Both the NYT and New Yorker articles were pretty good. As an experimentalist, I was slightly perturbed at the rant from Nima Arkani-Hamed in the New Yorker piece. Paraphrasing, he basically says, "The experimentalists don't let us see their raw data," and then later says, "Theorists do the important work anyway by drawing broad conclusions rather than just looking at individual observations." But in fairness, the author of the piece also included some anti-theorist quotes from Sam Ting and Leon Lederman. So I guess it balances out.
Anyway, it is good to see experimental particle physics get some media attention.
I'm sure you're right about this. Traditionally a lot of these dating methods had big errors (they probably still do). At the level I understand it, an uncertainty as small as a billion years (~10%) is good by astronomy standards. Also, I seem to recall from my cosmology class that the measurements of the Hubble parameter fluctuated a lot over the years. But the key change is the interpretation of the measurements--how you translate them to age depends critically on what values you assume for Omega_Lambda, etc. People used to leave dark energy out of the calculation, which gave a very different answer than we get today.
From the paper, it looks like this age measurement has O(10%) uncertainties. The authors don't even try to create an overall error on the measurement, instead giving a large table of various uncertainties. (Note: I'm not an astrophysicist, so I'm not used to looking at this style of paper.) From the paper: "Despite their large uncertainties the age limits provided by HE 1523-0901 and CS 31082-001 are in good agreement with the WMAP result of 13.7 Gyr for the age of the Universe." I would have phrased it replacing "Despite" with "Within", but that's just semantics.
From the ApJ website:
Discovery of a Ring-Like Dark Matter Structure in the Core of the Galaxy Cluster CL0024+17
M. J. Jee, H. C. Ford, G. D. Illingworth, R. L. White, T. J. Broadhurst, D. A. Coe, G. R. Meurer, A. van der Wel, N. Benitez, J. P. Blakeslee, R. J. Bouwens, L. D. Bradley, R. Demarco, N. L. Homeier, A. R. Martel, And S. Mei
Received: 06 Sep 2006
Accepted: 02 Mar 2007
Dr. Myungkook Jee, Department of Physics and Astronomy, John Hopkins University, 3400 North Charles Street, Baltimore, MD 21218-2686, USA (mkjee@pha.jhu.edu)
There's no link to a preprint and no preprint on the arxiv (that I can find). I'm guessing that they're embargoing it until it gets published?
I'm fairly certain this all just the NASA PR machine anyway. Another little piece of evidence to support
what we all know: non-baryonic dark matter exists. I'm sure it's a nice piece of science, but I'm nots
sure why it's on slashdot.
They use a Breit-Wigner because that's often a realistic model of the signal distribution, when one is talking about resonance production in a particle physics experiment. (My copy is at work, but I know this is discussed, for example, in Sakurai's Modern Quantum Mechanics.) I don't think this paper nearly lived up to the press release, and certainly isn't germane to Slashdot, but I don't think the use of a BW has anything to do with it.
On the other hand, I'm merely a particle physics grad student, and I didn't even attempt to read the center of the paper. If they really did come up with something that has more power than chi^2 (at least for an extremely simple fit) then that is notable. What would be really interesting would be for someone to come up with a real goodness-of-fit statistic for unbinned fits.
> > consists of Wimps (weakly interacting massive
> > particles) about a thousand times more massive
> > than a proton
> My training in physics is quite elementary, but
> I was led to believe the proton is relatively
> massive on the atomic level, especially when
> compared to an electron. How could a wimp be so
> large and yet unnoticed?
The key is the "weakly interacting" in the name. At the microscopic level, these particles (if they exist) can only interact via the weak force, which is both weak and short-range.
In particle physics the size of a particle has no relation to a physical size or a particle's mass. It is defined in terms of how strongly a particle interacts with other matter. (See the definition of cross section at PhysicsWorld.) So since the WIMP particle interacts only weakly, it is by definition "small," even if it is massive.
If the WIMP hypothesis is correct, then the WIMPs have hardly been "unnoticed." One of the chief motivations for looking for them is to explain the rotation of various galaxies which appear to be much more massive than can be calculated by adding up the mass of all the stars and dust in them. So if this missing mass does consist of WIMPs, then they have already been noticed!