Monckton says he'll take over the shuttered Pattern Recognition in Physics Journal

When any of you wiseacres currently dissing the PRiP papers UNREAD show any indication of having understood what has been written by the authors here having their reputations trashed, and, moreover, can explain to me, a non-physicist, the gist of their arguments and the faults therein, I will be prepared to the note your remarks but until then they are much hot air and self-serving, old-womanish tittle tattle. Please desist. You bring the board into disrepute.

Talking of code, here is the most relevant code for generically simulating climatic oscillations, the Lorenz attractor:
Function mylorenz (varargin)
MYLORENZ Plots the orbit of the Lorenz attractor with sigma = 10,
r = 28, b = 8/3 along with a Lorenz attractor with user-defined
parameters. MYLORENZ(sigma,r,b) takes input values.
MYLORENZ(‘sigma’) adds uncertainty to sigma.
Also: MYLORENZ(‘r’), MYLORENZ(‘b’). numsteps = 2 ^ 11; x = zeros(numsteps,3); dt = .01; sigma = 10; r = 28; b = 8/3;
beta = [.01;.01;.01];
x(1,:) = [10;20;30]; x(1,:) = [10*normrnd(0,1);10*normrnd(0,1);10*normrnd(0,1)]; x0 = x(1,:);
n = normrnd(0,1,3,1); for i=2:numsteps x(i,:) = x0(:);
+ sqrt(dt)*beta.*n; x(i,1) = x(i,1) + dt*sigma*(x0(2) – x0(1)); x(i,2) = x(i,2) + dt*(-x0(2) + x0(1)*(r – x0(3))); x(i,3) = x(i,3) + dt*(x0(1)*x0(2) – b*x0(3)); x0 = x(i,:);
n = normrnd(0,1,3,1); end y = zeros(numsteps,3); y(1,:) = x(1,:); y0 = y(1,:); if varargin{1}==’sigma’
newsigma = normrnd(sigma,7) newsigma = gamrnd(sigma,1) newr = r; newb = b; elseif varargin{1}==’r’ newsigma = sigma;
newr = normrnd(r,9) newr = gamrnd(r,1) newb = b; elseif varargin{1}==’b’ newsigma = sigma; newr = r;
newb = normrnd(b,2) newb = gamrnd(b,1) else newsigma = varargin{1}; newr = varargin{2}; newb = varargin{3}; end for i=2:numsteps y(i,:) = y0(:); y(i,1) = y(i,1) + dt*newsigma*(y0(2) – y0(1)); y(i,2) = y(i,2) + dt*(-y0(2) + y0(1)*(newr – y0(3))); y(i,3) = y(i,3) + dt*(y0(1)*y0(2) – newb*y0(3)); y0 = y(i,:); end figure;
plot(x(:,1),x(:,2),’k.’); hold on plot3(x(:,1),x(:,2),x(:,3),’g’); plot3(y(:,1),y(:,2),y(:,3),’r’); plot3(sqrt(b*(r-1)),sqrt(b*(r-1)),r-1,’g.’); plot3(-sqrt(b*(r-1)),-sqrt(b*(r-1)),r-1,’g.’); plot3(sqrt(newb*(newr-1)),sqrt(newb*(newr-1)),newr-1,’r*’); plot3(-sqrt(newb*(newr-1)),-sqrt(newb*(newr-1)),newr-1,’r*’); title(‘Lorenz attractor’); xlabel(‘x’);ylabel(‘y’);
hold off

@Steven Mosher at 10:08 m\
Stephan Rasey: ” But not all papers are written from government grants and Mosher’s demand that all data be released is unrealistic, undeserved and unwarranted in the majority of scientific fields financed by private capital.”
SM: Think about what you are saying.
First, you address what I said. You must live in a world where you do not have rights to publish data much less demand that others do so.
AAPG is an extremely well respected journal in an industry where information CANNOT be included with the paper. Ownership of the data cannot be ignored. Ownership of the full wave equation software cannot be ignored. That is only one jounal as an example.
Take the JAMA which lives and dies on Patients’ case histories. You do not have the right to make patient’s health records public.
That is why your demands that all data and code must accompany all papers are unrealistic, undeserved, and unwarranted.
Mind you, we are not talking about just climate journals here. Monckton in the main post never implied that the journal he speaks of will be limited to climate. In fact the cover illustration belies a climate focus. “Pattern Recognition in Physics” Not “Pattern Recognition in Climate Science”. Come to think of it, “Pattern Recognition in Science” might be more apt since many of the illustrations were biological.
I think it can be argued that in Climate Science the data is mostly owned by governments, if not in public domain. And most of the work on climate data is done with large helpings of government money from taxpayers. So demands that taxpayer funded researchers make their raw and processed datasets public is right and just. It might not be the journal editors’ job to enforce that policy, its really the job of the granting agency, but it seems a reasonable voluntary editorial policy.
Gail Combs at 1/23 7:49 am was real close with:

The real test is does the paper contain ALL the data, ALL the methods, ALL the computer code and everything else needed to make the information reproducible.

I would argue that the real test the peer reviewers and editors should subject the paper to is does the paper contain the methods and procedures and detailed references to replicate the conclusions of the paper. That’s all. Data and code are laudable. Required for publishing? Not necessarily.
Think about the profession of Cartography. By demanding all supporting data be published with a map, you would preclude the publication of all maps that were worth the paper they were printed on. Early world maps came from compilations of ship’s logs which were highly secret. Information about the world would be so restricted we might still think the world was flat.
The first line in the book “How to Lie with Maps” is:
“Not only is it easy to lie with map, it is essential.“
You must distort the data from necessity. Projecting 3D into 2D is a lie. Making highways wider than reality and red or green is a lie. But it makes maps useful by accentuating what is crucial and hiding what is unimportant to the purpose of the reader.
What protects us from bad maps? The map itself is provides the means to replicate it. People can take the map a verify it themselves. The reputation and fortune of the cartographer will suffer if the map cannot model reality usefully. Just as it should reflect upon researchers who’s conclusions cannot be verified either by omission or commission.
The final argument is that including the “data and code used” is a red herring. The real crimes in science are in the data that is not used and filtered out. It it better to debug some author’s code or to write your own from first principles described in the paper or references. Duplicating results from buggy code and error riddled datasets doesn’t progress science.
So the test of the paper is whether it provides the information within itself (or appendix supporting data) to enable other research to verify or refute the conclusions of the work of science.

That’s it…

😉

“So … once again I get abused by another in the endless list of anonymous internet popups”
It was this thread where you have exposed your true personality, and it is much different that what I had thought it was over the many, many times I cheered your contributions. You do some projection there with “too nasty” and the rest. Your two comments to me in the just above are all I need to see. You are a real piece of work and don’t even seem to know it.
I will never again read anything here posted by you without knowing what a real hater and foul little thing you are. Too bad. I just don’t know why you had to lash out at those folks at the Physics journal (and Monckton by extension for wanting to take it on) but I can see that you are real good at dishing it out but sorely poor at the reverse.
You are such a disappointment; but a good reminder not to think folks are decent people just because I happen to agree with their positions most of the time. For that timely reminder, thanks. Other than that, you are lower than your horsefly.

@Willis Eschenbach at 4:13 pm
Perhaps if you would link to a peer-reviewed map from a scientific cartographical journal it might be clearer
I said nothing about a cartographical journal. I spoke of the 3000 year old profession of cartography, the art and science of spatially representing in one document many types of information for many sources. The peer review comes from other cartographers, surveyors, navigators, captains and generals using the maps. When the British Admiralty or the US Defense Mapping Agency, create their maps, they use huge amounts of information. But they are under no obligation to release their source data to the public.
Even if you can replicate the code from the description in the Methods section, it still doesn’t allow us to determine if there is an error in the work.
No, what I wrote was:
[Is] it better to debug some author’s code or to write your own from first principles described in the paper or references[?].
The object is not to replicate the code, but to replicate the published procedures with completely different code and libraries. Thereby to attempt to replicate the results with an independent process. This is a far stronger test of the science than inspecting published code and rerunning it.
Successful replication should get reported. Faillure to replicate must get reported for then the differences can be further investigated, including enlisting the first author to evaluate his own code.
I truly, truly don’t understand people’s reluctance to endorse scientific transparency.
You are big on saying, “quote my words“. Where did I express “reluctance to endorse scientific transparency?” What I said to Mosher was, “You must live in a world where you do not have rights to publish data much less demand that others do so.”
When Blankenship, et al. applied for permission to present a paper on the Will K well, they had to face the dilemma of hiding some information. “Yes, you can present the paper. No, you cannot hand out copies of the logs, seismic, and thin sections. In fact, we don’t want you telling people how much this 425 day well cost. ”
You want transparency, Willis? Which alternative would be better?
A. to present the paper under these restrictions of less that full data disclosure,
B. not present the paper at all.
I endorse A. Be thankful for that much transparency.
And that’s all we’re asking for. As Mosh said, either show us exactly how you did it, nothing hidden, nothing kept secret, .
Asking?

Steven Mosher 1/23 8:37 am
1. will you require that all papers have a proper SI?
2. will you require that all papers supply their data AS USED in the paper.
That is, they should supply an actual copy of the data, rather than pointing
to a pile somewhere as Phil Jones did.
3. Will you require that all authors supply their code used to generate their results
4. Will you retract any paper where the author fails to supply this material?
I bet you’ll try to weasel out of these requirements and be worse than Mann or Jones ever were.

Sounds like demands to me. Point #4 is an impossible standard to enforce. It can be used maliciously to demand the retraction of papers for leaving out data of dubious quality, timeliness, control and an honest difference of opinion of relevance.
The rudeness and hostility of the statement that follows point 4 speaks for itself and why I am concerned that such standards could be used maliciously against honest researchers.
I repeat my opinion that completeness of data and code attached to the paper is laudable, but is a red herring. The greater danger is the over filtering of data and the omission of data that should be in the analysis.
or we are justified in ignoring you entirely because if you don’t reveal the data and code as used, you are not presenting scientific results, you are just advertising
That’s your choice. You don’t want to read papers that don’t hand out the author’s family jewels, that’s you choice. Most scientists in competitive industries don’t make that demand. They’ll read. And let’s face it… all papers are advertising in one for or another.
Ah, but if as you advocate that all journals and editors adopt that standard full transparency there will be far fewer papers published and far less overall scientific transparency as a result.

It looks my last post went into the filter.

“Please tell how you can “see” that. Onlookers observe the reverse.”
You speak for all onlookers? Hmmm.
I referred him to this thread and the continual assault on that journal and those who wrote there and then to the nasty, silly comments to me and another poster. It is fine by me if you see it differently as I only wanted Willis to know what I thought and I think he does. I don’t really care if he ever sees what he did or even understands what I said to him. Nor do I much care if you do; as there was enough trashing of that journal by several people here and you can see that or not — your choice.
I do agree with P Gosselin at http://notrickszone.com/2014/01/24/monckton-blasts-prp-journal-shutdown-21st-century-equivalent-of-nazi-era-book-burning-by-a-vicious-campaign/ when he wrote:

It’s becoming increasingly clear that a certain WUWT contributor is having perhaps too much influence at the number one skeptic blog, a blog to which we owe so much to. As talented as that person may be, it seems odd he would take it upon himself, given his relatively scant scientific credentials, to tell the rest of us which science is to be believed. One paper or two doesn’t make a person the authority. There were other self-anointed climate science quality czars who came out of the woodwork and over-extended. I was particularly disappointed with Poptech, from whom we found plenty of childish comments, like here.

… and with the rest of that post about this unfortunate instance of piling on these scientists without us knowing all the facts and with one hell of a double standard. And now Willis took yet another shot at them with his latest post. He could have refrained from mentioning them or that journal again but just could not work up the strength of character to do so.
Believe want you choose to believe my friend; but there was some serious and ugly piling on here at WUWT that left me gobsmacked. I was amazed.