Slashdot Mirror

You don't need to link to Roy Spencer (if you don't want to). You can link to peer reviewed studies. There are plenty of studies showing they're wrong. This too.

We find that the proxies do not predict temperature significantly better than random series generated independently of temperature.

Anyone who genuinely wants to debunk global warming should start here, trust me, climate scientists will respond with collective sigh of releif should anyone succeed.

RealClimate.org maybe wouldn't be the best place to start. There's a lot of very aggressively close minded chaps dominating the forums. I know, who'd have thought that could happen on an internet forum?

Real climate is also co-founded by Michael Mann, whom I really take some issue with. Tell me I'm making a mountain out of a mole hill, but his paleo reconstructions of temperature have really bothered me in the past. Nothing to do with the results, not as much to do even with his methodology now that his later work is addressing and correcting problems. The presentation and usage of the 'hide the decline' trick in graphs is just disgusting. When your paleo reconstruction ends around 1900, just end the graph there. If your paleo reconstruction doesn't show the same temperature rise since 1900 as instrumental, then show that too. What you DO NOT DO, is paste in the instrumental record with a thick enough line to hide the paleo reconstruction since 1900. Even further, don't point to the overlapped instrumental part of the graph as startling and clear evidence of an abrupt trend in the data starting at 1900.

If your wanting to have an open and honest discussion about the evidence, that's a difficult environment. Even scientists with a decent publishing record within the field like Lindzen are put under a microscope for criticism for not conforming to the 'consensus'. Even researchers widely embraced and accepted like Mauritsen have their results heavily disputed and interpreted there. When statisticians like McShane and Wyner take issue with the statistical methods in Mann and others work, Mann takes to his blog for the 'final' word while leaving out any response to their real and legitimate questions and arguments. I'm not anticipating that it's going to be a particularly receptive audience as you seem to believe.

Oh how I wish people would stop quoting skepticalscience as if a blog is a scientific resource.

It's kind of sad that you can make so many errors in one sentence. I referenced Skeptical Science because they have articles explaining in more detail exactly what I was explaining. I quoted one of the authors of the paper used by the parent to that post explicitly contradicting the view presented based on that's author's paper. How I wish you wouldn't ignore things that were inconvenient to you.

Skeptical science says exactly would you did, and most of what they say is sourced against another blog(RealClimate.org) which was at least started by a pair of actual scientists, but is still itself not subject to peer review either and really does not belong in your exhibit of evidences.

Isn't this just an ad hominem attack?

This is is what is WRONG with the whole 'debate'.

I would agree that your behaviour is exactly what is "WRONG" with this whole debate.

it very clearly shows temperatures as measured by proxy records matched or exceeded todays temperatures on multiple occasions in the last 2k years.

That would have been interesting if you had backed it up with actual evidence to support the claim. However, all I see is hand-waving and quote-mining. Looking at the graphs in the actual article shows one proxy in the Southern Hemisphere that appears to rival current temperatures at exactly one point.

Look for yourself here. there was considerable discussion, including on on RealClimate in which you'll likely find the assessment agreeable.

It's interesting, but I don't see the relevance here. It does not address the actual issue which is that no actual reconstructions show warming to actually have been higher in the past, so the dismissal of that claim as bogus is reasonable and expected, and the claim that the "current pause" indicates an end to global warming involves ignoring the oceans and wishful thinking.

Everything else you wrote is just pompous puffery. As you indicated above, we could do with less of that.

Those claims would be more interesting if some references were provided. For example, I seem to remember some people who are often referred to as statisticians (actually a minerals prospector and an economist) doing something similar, but it turns out instead of "proving" that the hockey stick wasn't real, they proved that they couldn't follow the documented procedures.

McIntyre & McKitrick aren't statisticians at all, so no argument there. Science shouldn't be about credentials, but if it is...

McShane and Wyner ARE statisticians and they published this paper in The Annals of Applied Statistics regarding Mann's statistic uasge for proxy reconstruction methods. The abstract follows, mostly because it pretty much speaks for itself:
Predicting historic temperatures based on tree rings, ice cores, and other natural proxies is a difficult endeavor. The relationship between proxies and temperature is weak and the number of proxies is far larger than the number of target data points. Furthermore, the data contain complex spatial and temporal dependence structures which are not easily captured with simple models.

In this paper, we assess the reliability of such reconstructions and their statistical significance against various null models. We find that the proxies do not predict temperature significantly better than random series generated independently of temperature. Furthermore, various model specifications that perform similarly at predicting temperature produce extremely different historical backcasts. Finally, the proxies seem unable to forecast the high levels of and sharp run-up in temperature in the 1990s either in-sample or from contiguous holdout blocks, thus casting doubt on their ability to predict such phenomena if in fact they occurred several hundred years ago.

We propose our own reconstruction of Northern Hemisphere average annual land temperature over the last millennium, assess its reliability, and compare it to those from the climate science literature. Our model provides a similar reconstruction but has much wider standard errors, reflecting the weak signal and large uncertainty encountered in this setting.

Mann et al. of course filed a rebuttal, which more or less amounts to declaring that statisticians know nothing about handling climate data, and furthermore that McShane and Wyner used completely inappropriate statistical methods.

I've read the rebuttals and McShane and crew seem to be the most on the ball in the exchange from my reading, with Mann et al's arguments seeming to be tangential to the central meat of the article and concerns identified, but go read it yourself.

Climate science is probably the most scrutinized field of science right now. And despite people saying the whole field is a crock, nothing of substance is found wrong.

Obviously the whole thing isn't a crock, there is just a lot of noise in the field now largely owing to it being such a hot topic and gold mine for grants and publicity. The basics like the instrumental record warming for a century, CO2 measurements increasing for a century and the fact CO2 contributes to the greenhouse effect are all thoroughly solid. That doesn't mean a horde of soft science hasn't been piled on speculating about the social impacts of potential speculative future changes brought on by this all good or might mean getting published too, often with not much more underpinning scientific work than something by Asimov.

Even some of the harder science is problematic, and yes in particular that includes Mann's work. His whole hockey stick temperature reconstruction has been thoroughly rebuked by The Annals of Applied Statistics. The two most blatant criticisms being failing to provide any and later accurate error margins, and the blatant attachment of disparate datasets to create a desired impression by appending the instrumental record onto his reconstructed temperature. The lack of error bars and the fact the reconstruction was calibrated to that instrumental data the close fit the instrumental gives the false impression of much greater accuracy than is present.

Mann is thin skinned because his early work was thin, bordering on deliberate dishonesty, or more generously, ignorance of proper statistical practices and methods.

>whether dedicated hardware generators

Those would be the ones that cell phone chip companies buy from IP providers right? Like this one..
http://www.discretix.com/accel...

Go look at the the picture. It's a ring oscillator that is sub sampled, to turn the accumulated phase noise of the RO into random bits. The process introduces serial correlation. This is normal.

Then it goes into a "Von-Neumann" whitener. This is a bias correction algorithm. Put in biased bits, get out unbiased bits. Lovely, it doesn't say anything about auto-correlation. The link to the Von Neumann paper is broken, but the Yuval Peres paper is just as illuminating. I quote..

1. Introduction. A source produces independent biased random bits {x_sub_i}^n,i=1 with p = Pr[x_sub_i = 0] ne 1/2, q = Pr[x_sub_i = 1].

So the Von Neumann balancer and the iterated Yuval Peres version have undefined properties for dependent inputs. They require independent inputs. But that isn't what the circuit is generating. They're most definitely auto-correlated.

However I've done the test and I can confirm that a VN whitener will make the result worse. not better when you put auto-correlated data into it.

This is how most of the published IP designs work. They are all wrong by design. /dev/random isn't magic. It needs its input to be entropic. Synchronous embedded systems regularly don't meet this requirement. PCs regularly don't meet this requirement early in the boot cycle.

Good RNGs exist. I've designed a few. Others have designed many. But bad RNGs and naive kernel code outnumbers the good ones.

In the west, young academics today are typically far better than their older peers at their age, or at any age. Certainly, there aren't as many folks revolutionizing multiple fields since each field requires more background, but across the middle tier departments you typically find the strongest researchers and lecturers amongst the younger professors and postdocs.

We need to make several changes :

(1) Remove all NSF grant money for postdoctoral researchers. Postdocs exist so that young faculty headed for middle tier schools can spend a little time at upper tier schools. A middle tier school like say Perdue should not have any postdocs. If an elite institution (MIT) or flagship state institution (Berkeley) wants postdocs for this purpose, then it already has the endowment to fund them itself.

(2) Add NSF funding for staff scientists analogous to French CNRS positions. They'd negotiate with the NSF about which university and lab they're posted at, but their position is basically permanent. As in France, these would basically replace postdoc positions in that the best young scientists would take them and the best of those would move on to university professorships.

(3) Ban exploitive part-time lecturer positions by removing NSF and DoE funding for universities that hire them. Of course, universities could hire full-time lecturer positions, i.e. teachers who teach a lot more than professors.

(4) Augment salaries for NSF funded PhD students who work on research topics with believable applications in industry. These kids are going to spend their lives seeing their peers in applied fields get more for less, start that early so that they can change early if they want. We're not talking really immediate applications here, obviously very abstract fields like group representation theory or rough paths have applications, but very popular subjects like hyperbolic aka geometric group theory or set theory has no realistic industrial applications at this time. There is nothing wrong with people working on subjects without applications, but maybe paying them $300+ less per month will help them realize that they're making their lives more difficult than necessary. In practice, this dividing line might appear different in different fields, maybe CS PhD in programming language theory would earn less for example.

Totally agree with you about LaTeX. TeX (on which LaTeX is based) was done back in the days of punch cards -- it was the only game in town for typesetting mathematics papers on a computer.

Point of fact. No, it wasn't. You should read Knuth's "Mathematical Typography" (I think you might enjoy it).

As to your second point, LaTeX markup is a great way to communicate formulas in graphics poor enviroments (email, chat, etc), which is why it's lasted so long. That too was intentional, if you read Knuth's paper above.

American football quarterback, PhD in mathematics (yes, they are the same Frank Ryan).

Wittgenstein versus Turing

<my guess>
Space has dimensionality; a plane has 2 dimensions, a cube exists in 3, hypercube 4... the numbers here are positive. Mandelbrot said he was using negative dimensions to measure "emptiness". He mentions that only one set is considered "empty" (I presume the null set). My guess (and I only minored in math so don't go betting on this) is that a negative dimension is to a positive dimension what a negative number is to a positive one. I'm thinking that if an object existed in -2 dimensions, it would be capable of having negative area. If you could add that object to an object with positive area, you'd reduce the second object's area.
</my guess>

Here's Mandelbrot's homepage at Yale.

Here's more links.