TCO, you might try and consider what get's you banned on all these blogs. Perhaps the problem is you.
I'm not unsympathetic as I generally found your posts on tAV worthwhile. Indeed I frequently operate in a similar way, as most people tend to hate my comments on WUWT, and occasionally ClimateAudit.
Amac: I have some comments to you that are still held up in moderation (I guess will be forever) at CA. They are long and probably don't say anything important, but "just in case", will copy them here so you can read.
-------------------------- Response to your 093707AUG10 (me 122307AUG10)
Your comment is awaiting moderation.
1. I have a comment in moderation addressing the PMS issue and Mann reluctance to engage on substance.
2. I like your little math analyis on the variability changes over time. It puts what bothers about the big ramp into words. That said, should that be a formal screening mechanism? Can it be defended as such? Even if so, it’s not 2 plus 2 simple. We just need to be real careful that we are not harsh on BCPs and
3. I’m not so much trying to get you to be nice to Mike, although if you intend to indict him for not admitting something simple, you have to really show that it’s simple. I think the whole hammering on Mike to admit is a dry effort anyway. Better to argue the contamination issue, itself. Think about it this way, if you had that LANL director, me, Huybers, Curry, Huybers, Zorita, etc. and you wanted them to sign off that this thing was a slam dunk. they might want to really see that it was a slam dunk.
Little story: I remember Peilke and Boris(?) having some argument about some IPCC definition and if it was a mistake. And they were both all hot and bothered with each other and EACH thought he was not just right, but OBVIOUSLY right. And I read it real quick and certainly could not tell just from the blog debate who was right. I knew that I would have to read the source documents that they spoke of, maybe even have some context from being in the field to rule on who was right. So even though both thought it was crushingly obvious and not really even debatable, I couldn’t judge so quick. Well, maybe I’m dumb or wishywashy or something. But then an interesting thing happened. Zorita was appealed to (by Boris, I think, but no matter) to rule on the dispute. And I asked him to rule as well. Not because I think he needs to adjuticate all disputes. But because if it really were simple, he could do so quickly. Now this is a guy who has decades of climate science working exerience. And guess what he said…basically, what a bother, I’ll have to go read all the documents…it’s not obvious just from you two antagonists arguing in a blog.
Anyway…I think disconnecting the “what is correct science” from the “Mike fails to admit” (and not wasting toooo much time on the latter) is both better tactically for dealing with the Po-Mo scientist as an opponent AND better for the disinterested observer who cares about science.
—————- I was gonna reply to the reviewer comment, but I didn’t have anything additive to say over previous. ---------------------------------
I think I had a post erased as well. Basically agreeing that Mike etc. won't agree to anything other than obvious errors (like a coding mistake) and even then the "I don't know button" will be pushed). It had a bit more substance to it than that. And a recommendation that they start lifting heavy to generate natural testosterone to counteract PMS.
> 2. I like your little math analysis on the variability changes over time... should that be a formal screening mechanism?
Yeah, maybe. At a more basic level, perhaps there is a temptation to view proxy candidates as mathematical objects (strings, matricies, whatever), rather than as representations of some actual physical property. As soon as you look at Lightsum in that light, you can hear it shouting, "I'm unsuitable!"
> 3. be nice to Mike...
As I've said elsewhere, Prof. Mann's conduct probably isn't all that many standard deviations away from the norm. Yes it would be better if he put on his BBPs. But part of the function of a scientific community is to ensure that scientific findings aren't distorted by the human shortcomings of its human members. There's (1) the actual issue of the proxies' uses, (2) what that means for the results of the paper, and (3) what all this indicates about the state of climate science. What the authors choose to do is #4, and what I think about their choice is somewhere between #5 and #(1/0).
You are responsible for the methods you use in your paper. Saying that the reviewer made you use an incorrect method is bunk. It's your byline. You are responsible.
Your error bars for "the models" are unbeleiveably tight. You should have had enough physical intuition to know that did not make sense. That error is a real boner.
If correctly calculated, the error bars would still be different from observations, then you haven't SHOWN it in your paper.
Why not show the test both ways, if there was a debate about how to make the error bars?
What if the reviewer who told you to put the error bars down the wrong way was Douglas? What if a reviewer told you to put something down that said 2 plus 2 is 5? Would you?
Why is the correct method not shown in your SI?
You messed up, Steve. And didn't have the balls to let me criticize your paper last night. Crossposted at AMAC to display your censoring.
Last night you censored me and I said nothing about "balls". I made an apt point about what "the models" predict and how just having some physical/mathematical intution should make you realize that those tight error bars in your figures are not meaningful. You said it had nothing to do with the topic! Heck even if I'm WRONG, it was way more on topic (getting into your figures) than all the peer review whining.
You just don't want your ideas technically criticized, McIntyre.
Oh...and if you group RSS and UAH together, the error bars of "the sattelites" are pretty big too and overlap "the models", too...
Previous post was made before realizing that Steve has two MORE posts up at CA. The strange thing is he's introducting NEW analyses, not shown in his paper. As if it doesn't bother him, not to cover things in the paper and just wants to have an ongoing dialogue. He's also not really defending the key issue of Figure 2 and 3 too-small whiskers, but instead trying to redirect. Really shows how bad MMH was as a technical paper. (cross-posted for censorship check.)
I'd respect you if you left the post open for comments and the picture up there. Putting a comment up that you will correct it eventually is fine.
Closing it is passive aggressive cowardice and abuse of your moderating powers to influence the debate. No attaboys from me, for your graceless admission.
There is no question of moderation. First the site has other moderators. Second, there is at least one busy thread currently (the whole site is not locked down).
The issue is McI does not have time for REBUTTALs. And also, that he doesn’t want to take his medicine for an error. This is abused of the moderator/publisher power to influence the debate.
Also, note that at the same time he admits a mistake, he takes a shot at his opponents for not doing so, and pats himself on the back.
Interesting that Steve has time for new 300+ comment threads that attack his opponents, but closed comments on the one thread, where he made a mistake. I don't have an issue with him putting an update saying he was wrong, or that he will do a reanalysis. But stopping discussion is abuse of his moderator power.
Or you could just link to semi-violent hockey analogies. Yuck, yuck! Go hoi polloi! Let's all grunt and hoot like the audience on the Mort Downey Jr. show.
I guess you buckled down and got all her papers. I skimmed the paper, but I really need to print it to parse it. It's unfortunate that her thesis does not really tie all this stuff together. What I saw did not seem stickyouintheeye simple, so need to read the details to see if her inferences are based on strong info. Would also be good if she could somehow validate her earlier inferneces on LS, DS, XRD, etc. based on what the isotopes tell her.
I thin it is irrelevant whether Tiljander is correct or incorrect. It is possible to create hockey sticks without Tiljander. What matters is the behavior of the scientists involved. They have not argued for an alternative interpretation that Tiljander is incorrect. So what matters is whether they have properly implemented her interpretation, and when it was pointed out that they did not, how did they respond? Given their stonewalling and denial of the obvious, it lends credence to the idea that they cannot be trusted with other hockey stick studies as well. Take a look at the Mcshane and Wyner response at RealClimate. Do you think it is reasonable for them to test its sensitivity to Tiljander, have that be one of their primary graphs?
I added pointers to PDFs of two more studies of varved lakebed sediments in the "Primary Links" post. One focused on Baffin Island glaciation, the other on the Younger Dryas transition, 12,700 years ago. Different strengths and weaknesses cf. Tiljander-type papers.
CA cross-post (my comments are being pre-moderated, now)
1. I think it's bizarre that you spend time on this gotcha game investigation fol-de-rol, when you have major outstanding science/math issues. That said it's your blog and life blablabla. [However, if we have the ability to make free comments, this is MY, relevant, comment.]
2. What's really galling is that you ERASED evidence of a mistake of yours (after "publishing" in your blog) and then locked discussion of the mistake. And said that discussion could resume only when you published new analysis.
3. I'm not saying you have to finish that analysis. I'm not even saying you have to be a part of the discussion. But you should put that previously published mistake BACK UP for the record. And you should unlock the thread an allow discussion. Otherwise this place is justg being run as a PR organ, and you're dishonest, and "publishing in the blog" means nothing since you can alter or edit anything, (it's not a real record like a journal article.)
4. People like Gerg Burger, Ed Zorita, Judy Curry, etc. should have nothing to do with you. I urge them strongly to stay away from you...have fun with the hoi polloi and the Watts-like comment discussions.
AMAC, the whole NEverending Audit is tedious, but I like you. The issue with the "red noise" is that McI essentially uses circular logic. He doesn't make "simple red noise series". What he uses are rather complex models that take the proxies themselves and try to abstract a noise structure from them (across all freuncies). The issue is that any signal will now be treated as noise. So a noise model based on that, is circular logic. Essentially it is ASSUMING LTP. Assuming the case to be proved. In addition McI was resistant to clarification of what he meant by "red noise". (Similar to the pattern of resisting description of the MMH whiskers, etc.)
Thanks for writing. I think you make some good points. I expect you kinda know what I'm going to say, but anyway --
I wish I'd tagged the comment a few weeks back where Gavin brought up the point you raise about red noise (I don't even remember where). It got me to thinking; it seems to me to be one of the keys to dendro.
* Treerings contain redness (autocorrelation of a time series).
* Instrumental temperature records contain redness.
* Treering series are affected by a lot of things. The biggie seems to be precipitation for most trees. In some settings, precip isn't usually limiting, so temperatures become the biggie. Then there are a lot of other macro-environmental things, like days of sunlight in the growing season, and whether the winter freeze was damaging. And micro effects--insect infestations, nearby treefalls, whatever. Some of this, we can call "noise." If what we care about is the temperature signal, I guess we can call everything else noise.
* Some treering time series but not others include temperature signals for part or all of their length.
So the question would be, where does the redness come from? Temp? Precip? Stuff we can't interpret, ie "noise"? If it's a mix (these things usually are), how is it apportioned? Is the mix steady for a given stand? Can a mix be a priori estimated for a naive stand?
Since dendros aren't dummies, I presume these things have been investigated. I don't know that literature. Maybe somebody will toss in a link to a fantastic review that makes it all clear.
But for now, I no longer presume that paleos treat these sorts of confounding factors with proper diligence. It seems to me that too many people are prone to direct their thinking down particular channels.
It's hard to implement the proper caution into scientific work. That's why I transcribed that talk by Norman Wolmark on breast cancer clinical trials and made it an early post. If you went back in time to the 1970s and stood up at a meeting to discuss the limitations of the governing philosophy of clinical trials, clinician-researchers would have treated you with condescension, then with scorn. Because they knew that they were on the cutting edge of science (which was true).
So on the one hand we have the neverending audit (catchy phrase), and on the other we have that same sort of condescension and scorn, underlain with sclerotic thinking.
The science is very interesting and I agree that the Mike/Gavin crowd is a bit political (both in science clique sense and in Hansen advocacy sense) and I hate the Herrdoktorprofessor attitude.
That said, McI's way is not doing anything. He's been jerking around for literally the amount of time that someone could have written a Ph.D. and several fundamental papers. People like Burger, Huybers, VonStorch, Zorita, Christensen etc. need to pick away at things. And also, they need to publish when/if they find things that help Mann or his "cause". The McIntyre crap with the hoi polloi and the 4th tier market TV weatherman is just not....doing it. If you want an example of an amateur NAILING something in terms of methodlogy and clear communication (via a blog mostly), look to the British guy who found the small Met Office formula error. however, note how clear he is about explaining the level of impact and of not overselling things or running a stable with a bunch of forum commenters circle-jerking each other.
Man I am bored with the climate drama. Not angry at the sophistry. Not even an angry bored. Just bored, bored. Can't bring myself to read past the fold on Judy's blogs. Can't bring myself to read comment threads. I'm megabored.
I think I caught you saying some good things about Jeff Id. I think he's got some pretty big gaps (a bit of the Dunning Kruger kind where he does not see them). In terms of his data analyses, he has messed around for years, with lots of jumping to conclusions and trumpeting issues, and little final insight. He's also biased in terms of analyses.
Basically, the guy is a typical blustery climate blogger skeptic. You want a smart, honest guy who has found some corrections? Check out John Graham-cumming.
I really don't think Moshpit and Lucia are that good either. They are much to inclined on their initial hypotheses and take years to re-evaluate (meanwhile spewing like crazy). Just not real curious and objective analysts. Not Feynmanian. Not Rickoverian.
Yeah...they like to "play" with code and equations and crap. good for them. But the way they sound off before analyses are done is pretty flawed.
McIntyre is bad news. Look how he avoids discussion of the noise model decisions in his GRL article or of the "pick top 1%" mistake in the Wegman report. Those things are square in his area of interest, experience, and previous discussion. But nothing. The simple answer is he avoids discussing "his side's" errors. It's the attitude of a person who is more an amateur legalist (sealawyer) than a curious scientist/mathematician.
Instead we have McI concentrating on rehashing Climategate and playing to his Wattsupwiththat buddies and hoi polloi.
Stick to your petri dishes and business projects.
If you want a cool, smart, honest, curious guy; check out http://blog.jgc.org/ (do a search for his climate posts). This is a guy who I would trust to report findings that went either way in terms of inspecting climate work.
Posted at RC. Not sure if it will all make it through. I annoyingly have some mildly bad things to say about the team, mixed in with my comments on Id and ilk. Not out of sneakiness, or even false symmettry, that's just how I honestly see it, but as such this earns me lots of hate from the two tribes:
PolyisTCOandbanned says: Your comment is awaiting moderation.
10 December 2010 at 2:06 PM 1. Still waiting to read the paper. Looking at it, side by side, with Stieg will be interesting. Comments below are “gut” reactions on the whole kerfuffle, from a blog dispute follower perspective, fwiw:
2. From when I first understood what Steig had done, I thought it was a neat idea. Something definitely at least worth trying, and possibly something that would give us a better picture. The basic idea (I think) being to use sattelite period when we have HUGE information about how different weather patterns appear on the continent, to allow us to infer past temps at “blank spots” in the continent, back in the day when there was just surface stations.
3. It’s actually my LIKING of that cool concept that makes/made me worry about the approach of doing PCA and throwing away the stuff that does not go into the first three PCs. You lose the possibility of a lot of complicated patterns if you only have 3 PCs. Similar to mapping a complicated surface, to looking at a chemical compound, etc. 3 hybrid orbitals are better than 1, but may not be sufficient to understand some behaviors. For instance, let’s say there was some weather pattern that a spot on the continet would be warm when one other three other spots were warm, and two were cold (assume they are all far apart and somewhat independently variable). You could not follow that with the 3 PC approach. Of course, I don’t know that such complicated patterns exist. But if anything it would be nice if they did as it would give us more of a key to unlock a door…more of a “fingerprint”…to give us pattern detail of the past from this deft approach of Steig.
4. I’m interested in the PCs and their geographic appearance. I’m not a math jock, but from the beggining the PCs have always sort of reminded me of orbitals for electrons (the pictures), or perhaps those drum patterns you get when you take Engineering Mathematics and (try to) learn about Bessel functions and the like. I’m actually interested to learn how much Eric or others have thought about this issue and if there is any point, to pushing thinking about the analogies.
5. I’ve always been a little worried about some of the implicit, partial, hidden (pick a caveat, this is my impression!) assumptions about PCA by “the team”. I guess mostly Mike, since he’s the one who is most mathematically sophisticated given the math-phys background. It seems like there is often an approach of assuming that PCs represent something physical (a factor), but of course sometimes they may and sometimes they may just be a mathematical abstraction. Also, that one can take the important first few, throw out the rest of them and actualy sort of “clean” stuff up, almost like filtering noise out. It just makes me worry of the analogy of throwing away outliers versus doing a simple average or trend, with all the data. I think of PCA as being more suited to factor searching for, or perhaps data compression, or facial recognition, or operation of a Nate Lewis chemical nose. But not as a “filter”.
6. From the beginning, I’ve been annoyed by the approach on The Air Vent. First the incredible huge amount of noise to signal. All the different blind alleys and trial approaches. The mixing in of all kinds of triumphalism along the way, before the argument had even been refined. Not only is it annoying to read because of all the riffraff and forum games, but it’s just LONG.
6.5. A few examples. The krieging approach (did any of it make prime time?) Also the “negative thermometer” complaining. A quick think would show that mathematically an opposite signed predictor can happen in a data set. And even some examples of weather patterns, where one area being cold correlates to another being warm (hemispheres of the Earth, El Nino WESTPAC versus US SE, even the pictures of the jet stream reaching down the middle of the US and snaking around that we see on the evening news since at least the 70s; and note the last two could have somewhat of a time dynamic, that is more frequent than just the winter-summer simple example) make one think it might be possible in practice. And despite even several skeptics showing Jeff Id mathematically how he was wrong, he could not (or would not out of stubborness) get it. That was a time when RyanO failed me too. Here, he never directly corrected himself “I was wrong, I take it back” but instead said after interminable internet style repetitive debate and even writing to Jollife and getting a genuine PCA expert to grace their thread, Ryan came back with a “oh my view evolved during he thread, but I never really wanted to say so, you should figure it out”. A simple tiny point, but the real issue was the failure to understand a clarification and to admit it–not that it blew up their other concerns.
7. I’m curious why the pre-print was not circulated as a white paper. Is this JOC policy? Reading the pre-print, vice the blogorhea would be a way more efficient way to evaluate the counter-team’s thoughts. And those guys have a ready “bandwidth” and site and all to post it on. For that matter, given that they post the ongoing analysis (in blog posts), why not post the more refined and better communication?
I worry that this is a “tactical approach”. In other words, make a lot of blather-on forum posts, and then just submarine up papers without exposing them to criticism. For instance the recent MMH paper had some real (obvious) problems in defining standard deviation (even!) and in clearly discussing the hypothesis test. And in that case, it looks like a bunch of (unimportant to the key dispute) math was larded on to make it look fancy or new or distracting, rather than squarely focusing on the definitional and hypothesis test debate.
6. From the beginning, I’ve been annoyed by the approach on The Air Vent. First the incredible huge amount of noise to signal. All the different blind alleys and trial approaches. The mixing in of all kinds of triumphalism along the way, before the argument had even been refined. Not only is it annoying to read because of all the riffraff and forum games, but it’s just LONG.
6.5. A few examples. The krieging approach (did any of it make prime time?) Also the “negative thermometer” complaining. A quick think would show that mathematically an opposite signed predictor can happen in a data set. And even some examples of weather patterns, where one area being cold correlates to another being warm (hemispheres of the Earth, El Nino WESTPAC versus US SE, even the pictures of the jet stream reaching down the middle of the US and snaking around that we see on the evening news since at least the 70s; and note the last two could have somewhat of a time dynamic, that is more frequent than just the winter-summer simple example) make one think it might be possible in practice. And despite even several skeptics showing Jeff Id mathematically how he was wrong, he could not (or would not out of stubborness) get it. That was a time when RyanO failed me too. Here, he never directly corrected himself “I was wrong, I take it back” but instead said after interminable internet style repetitive debate and even writing to Jollife and getting a genuine PCA expert to grace their thread, Ryan came back with a “oh my view evolved during he thread, but I never really wanted to say so, you should figure it out”. A simple tiny point, but the real issue was the failure to understand a clarification and to admit it–not that it blew up their other concerns.
7. I’m curious why the pre-print was not circulated as a white paper. Is this JOC policy? Reading the pre-print, vice the blogorhea would be a way more efficient way to evaluate the counter-team’s thoughts. And those guys have a ready “bandwidth” and site and all to post it on. For that matter, given that they post the ongoing analysis (in blog posts), why not post the more refined and better communication?
I worry that this is a “tactical approach”. In other words, make a lot of blather-on forum posts, and then just submarine up papers without exposing them to criticism. For instance the recent MMH paper had some real (obvious) problems in defining standard deviation (even!) and in clearly discussing the hypothesis test. And in that case, it looks like a bunch of (unimportant to the key dispute) math was larded on to make it look fancy or new or distracting, rather than squarely focusing on the definitional and hypothesis test debate.
8. There is also a LOT of whining about reviewer comments. Certainly, it might be possible for a reviewer to be unfair, but my general impression has been that authors whine more and are more often in the wrong in these kinds of disputes. I’ve always found being brutally clear (even to admitting dropping a sample on the floor) to assuage reviewers. It’s only when one tries to spin too hard, that they dig in. Also, have found that very clear writing and following the POSTED guidelines of a journal does a lot to get a paper through the wickets. Many authors could save themselves trouble if they just put a higher level of care (basically make the paper as perfect as you know how, imagine that the reviewers will fall dead and you need to get that paper perfect on your own) before submission, rather than dealing with all the back and forth letter writing. If you really do it right, you should be able to get a lot of papers accepted without revision.
I have to take the reviewer whining with a grain of salt, on the most recent counter-team paper. Of course it COULD be justified. But how can we tell unless we see the inital paper, final paper, and perhaps the correspondance? Based on the blog posts I’ve seen, some previous McIntyre submissions, writing conference pressentatations the night before a meeting, etc., I can’t be confident that the initial draft was even smooth. Of course, NicL and Ryan may have done a better job and they seem to have had more of a hand in this one. But…I’m explaining my Bayesian betting basis…
This is tedious. Please get rid of the 4000 character limit. I have to keep chopping it into peices and then part of it got through and now got taken away. I'll just frigging post over at Jeff Id's. It'll give him satisfaction, damnit, but I bet he lets it through. Grr.
Sorry about the character limit, but it's nothing I have set. This blog is a freebie from Blogspot/Google, so I get what I get. I find the inability to do much formatting in comments rather more annoying. You could sign up with Blogspot or WordPress as long as you link, you know, just a thought...
For whatever reason, some of your RC comment copy got tagged as spam. It's fished out and in the proper order now.
On reviewing, my experience (cell/molec biology) is very different from yours. Reviewers usually had some critical things to say and demands to make. When they didn't, it often looked as though they hadn't bothered to read the MS with care. As a reviewer, I never had the problem of "no suggestions for improvements for this submission." I always separated "major" from "minor", and when I thought all my criticisms were in the latter category, made a point of praising the authors' work in that regard.
Anyway.
RC's comments are sorta hopeless with their "like!" (in-it-goes) and "don't like!" (aggressive moderation/delay/snip) policy. They undercut their own claims of being a science-focused site; productive discussion is very unlikely under their ground rules. I rarely bother, but good luck.
Jeff Id's somewhat erratic in his wide-ranging choice of topics (and words), but I think he genuinely respects what the data are, and doesn't spin the math (wrt the posts I've been interested in). RyanO and Nic_L's commentaries, nice IMO.
My experience is probably pretty untypical. I had three papers in a row accepted without revision and my professor had 200 before that all needed revisions, and 400 as an editor that needed revisions. He actually did not understand the concept of just "accepted". That said, honestly, the time taken to really be super candid and super attention to detail ends up paying off. You spend less time on reviews later, because of the work done upfront. And it makes it easier for a reviwer to add value if the thing is as well done as possible. It drives me to distraction to have to review a paper with major issues on the science as well as being written in Chinglish.
Id was all over the map on the Steig thing. Painful to read as it developed.
And it sounds like their paper went through huge changes (like they had fundamental stuff buried in the SI that didn't belong, sections that should have been in another paper, etc.) I bet that thing was no too smooth to start with, but more of some blog posts stuck together. Those guys get too used to self publishing and don't really understand how to raise their game.
I've said it before and shouldn't have to repeat it, but something about the interminable style of chat on the net, makes repitition needed: The skeptics don't moderate completely honestly either, but at least you can get some stuff in and having no pre-moderation is a big improvement. RC is bad news with their moderation. At the end of the day, you almost need to get your own blog to participate. But at the end of the day, it's some jerkoff on the net anyhow. Lot of sturm and drang for not that much payoff.
There's a lot of whining about the reviewers making them cut stuff on that paper, but from reading between the lines it sounds like they had a lot of disparate ideas of differing levels of development and that the inital paper was a bit of a "splat". Probably, they caused a lot of the delays on their own. I know how these guys operate and am not surprised. They just get used to very low standards from what they get away with on their blogging and then think that is quality work. Then get surprised in review.
I have absolutely NOTHING wrong with a post, saying look, the adjustment changed. Nothing wrong with the thoughtful impulse to fasten in on red flags and examine them. So I mega-agree with your comment that it is reasonable to flag the concern.
That said, it is a far, far cry from the wandering (even as a slam, it's just bizarre) snark-filled accusatory. And interesting that when the guy buckles down and looks at the Hansen paper, publicly available, that he finds no significant change in Hansen's methodology. Sure the input changed. But that is a far cry from "Hansen staying up working on adjustments". And again, it's at least POSSIBLE that the changed input is a more appropriate input.
I just can't grok this way of dealing with things. I'm ALL for digging into the target's 10K, etc. etc. Let's go...let's DIG. But let's keep an open mind. The current way of agitating hoi polloi and then never going back and upgrading the snark-filled half-accusatories (always with a plausible deniable, "I never said that" is just bizarre.) This kind of crap would never pass muster in a (good) company or the military or science.
Plus it's just wandering and disorganized and poorly stated. To the extent that comprehension is very seriously limited. Watts is definitely MORE of an idiot (much more, I have always praised Steve for code and math knowledge), but at least Watts is CLEAR!
Ryan is a punk and a whiner for getting into the peer review stuff.
He's also disorganized in his presentation of material and wastes too much time with emotional venting. If his submitted paper was anything like his blog posts (and McI's usually are) then no wonder it took him so long to get something shaped into publishable form.
I've followed this kit and caboodle for 6 years now, but still find it a pain to look for what point you are making. Think how bad this looks to the intelligent outsider reading your stuff. If you want an example of people who do things right, look at Ed Zorita or that British blogger who's found some code glitches. They actually break things down and explain them. Not meander and vent and draw crappy graphs without axes and figure captions.
As usual, there is an issue of McI changing two things at once and ascribing the resultant change to one. (Or glossing over his muddling.) Way better to have done the full factorial and presented it.
And NONE of this exonerates ES. It's just that I can't even make out your point of technical disaccusation. I'm all for improving the science. All for taking down some liberal latte drinkers. But you guys screw it up so bad. You don't drive any understanding of material. You just have little social internet games.
Perhaps Jeff "contrail" Id will flex his chest in happiness over the dramah...or the Cold Fusion lovers at your "ally" Anthony Watts will just give you attaboys without trying to understand what the heck is going on. Probably Amac and Hugh will sigh at your disorganized rant.
Hey TCO, thanks for tossing in your two cents, above. I'm guessing you submitted it to RyanO's Steig's Trick post. (The link provides context, hint, hint.)
You might be interested in my take on what L'Affaire Steig/O'Donnell says about the peer review process, as Comment #68919 at Lucia's.
Those guys step on their dicks so hard, man. It's been 6 years of McI silliness I've followed and he was at it a couple years before that. Long enough for someone to get almost 2 Ph.D.s. This shit is really a turnoff. Keep it real. (None of this is to say that Mann et all should not be scrutinized but the whole McI Watts set of them is full of shit.) I think Hugh has a clue too. The rest of them are pretty fucked up.
Stick to Zorita and such if you want someone smart and honest. McI is bad news.
I just read down. What a clusterfuck. Ryan never did the basic full factorial of method choices. He did not disaggregate issues to test their seperate effects.
It's the same McI crap of changing 5 things at once (well at least 2) and then making a complaint. He's done it repeatedly and been slammed for it. But continues. What trash. To think Ryan is some sort of manufacturing engineer.
And the excuses about money and page limits are total horseshit. They had 2 years working on this. They could have had all the extra analsyses done, they could have had them on their own website, etc. But when even people on his "side" push for the basic sensitivity analsysis, he has to go run it NOW. The whole thing is bad news, bad engineering, bad quality. They ran a gazillion half-baked analyses that they trumpeted on a website for their yukyuks to egg them on for, but couldn't do the basic taking apart of the problem.
Really underwhelmed with Ryan's "apology" (the titling of it, the placement with a bunch of excuses and other complaints, the having McI put another whine right on top of it). I remember a long time ago spending a bunch of time to pin Ryan down on a very small mistake. After a bunch of work to show the mistake and prove that he had made it, he said his view had evolved in the blog discussion, and that he had already taken back his point. But he never called it out as a correction.
I find this behavior dishonest and cowardly. Very sea lawyer. Very unmanly. Would not want to serve with such.
Move on, Amac. You are better than this crew. People like Zorita and such can correct Mann and their over-reaches. But you don't want to hang with the McI, Mosh, Lucia crew. They're second raters.
Look at the [snip] [snip] "skeptics" lately. they can't really scrunch down and descend a testicle to actially call out Wegman for his plagiarism or Watts for his years of false snark. But they don't want to be caught with really supporting it either. What [snip]. they're not even Americans.
Re: Wegman, I've commented on his plagiarism and related problems at Lucia's and elsewhere. Seems to me that Deep Climate and John Mashey have made a very substantial case. Sauce, goose, gander.
Re: Watts, I don't read WUWT very often and have little to contribute.
I wouldn't post here if I didn't think you were a pleasant surprise. But Watts, McI, (and Lucian and Moshpit) have a long history of evasion and of looking the other way. The whole scene is bad news. Bunch of losers.
Typical pattern. The guy asks for help on his submission with hours to go (for something he had 60 days to work on). I've seen him do this several times before and as in this case, his submissions were spotty. He doesn't help sympathy that he is being "put down by the man" when he is so repeatedly sloppy.
Is there anything new in the latest McI post? I skimmed it and it seemed like a total repeat? Could he maybe put any new parts in red or something? It's boring and confusing and makes me not even read what he has to say. Even the video snark was a rerun.
P.s. And just re-running stuff is not intellectual engagement, it's drum-banging.
It's been a year since McI closed the comments and erased a mistake of his in a thread, promising to redo the calcs. But he has not done so. He has also failed to engage with the other substantive criticism of the MMH paper itself.
Posted in the Hu thread at CA (post preservation, in moderation):
Hu:
I’m generally about 80% sympathetic to you on all things and about 90% on Mann doing you a dirty, here. Certainly you play more fair than McI and at least put things into a form (papers) that they can be engaged with. and I can follow you (not a meandering snark full of mistake-laden graphs, written in mystery story style, to the extent I can’t even tell what the assertion is). In other words, you are much more straightforward and academic.
Few issues though:
1. I would avoid “plagiarism” as a loaded word and as not precisely correct. They failed to credit you for an idea. This is different than using prose from soemone else. It was a failure to acknowledge/cite. McI repeats the same error here again wrt “Harry”. I have pointed this out before…
2. It is over the top to call Odonnel et al a “refutation” of Steig. I would be more measured (for example, the paper is more measured). It is probably halfway between a total refutation and a minor improvement in the same vein (as alleged by some of the Steigers). Maybe significant upgrade or some such word (not sure). But definitely not “refutation”. Talking that way is kind of stooping into the Michael Mann (the error is all theirs) type of rhetoric, which is a little gamey debatey stuff. not real science/math logic/disaggregation.
3. I advise avoiding EnE for publications. Do Climate of the Past. Even a very crappy paper is basically exposed and published as a submission. Even if you don’t get accepted, it is still displayed. You will not get a good peer review at EnE (look at the fellow you helped who had to do a total rewrite of a paper…how did that thing sail out, when I could see all kinds of issues just from being a blog-reader).
"1. I would avoid “plagiarism” as a loaded word and as not precisely correct. They failed to credit you for an idea. This is different than using prose from soemone else. It was a failure to acknowledge/cite. McI repeats the same error here again wrt “Harry”. I have pointed this out before…"
wrong. Plagiarism is not restricted to copying the words of another. One reason we require students to supply a complete bibliography of everything they read when writing a paper is so that we can insure that they credit other for the IDEAS that they got from reading others. Here is source:
scientist, I will take it that you have never detected a plagiarism or served on a committee defining what constitutes plagiarism or brought a plagiarism case against a student. You do not know what you are talking about. I suggest silence.
cut and paste from Climate Audit (my comments are not allowed on there).
Amusingly enough I was commenting to a thread where McI complains about moderation.
He is such a weak tit.
Still has not supplied the data and code for the Watts-McI paper.
Even Moshpit pushed him (kid took a few years, but he got to where I did in a few months).
"bad news bear
Posted Sep 10, 2012 at 10:10 PM | Permalink | Reply Your comment is awaiting moderation. Could you make a new post complaining about the proprieter of Climate Audit shutting down discussion (locking comments) when he had effed up an MMH analysis?
And also how he promised to redo the analysis…then didn’t for over a year now?"
TCO, you might try and consider what get's you banned on all these blogs. Perhaps the problem is you.
ReplyDeleteI'm not unsympathetic as I generally found your posts on tAV worthwhile. Indeed I frequently operate in a similar way, as most people tend to hate my comments on WUWT, and occasionally ClimateAudit.
I'm going to move TCO/scientist's comments from "MikeN weighs in..." to this thread. (Anonymous's remark supra will then make more sense.)
ReplyDelete[Originally posted by "scientist" to the MikeN weighs in on Mann08's use of the Tiljander proxies" thread on August 9, 2010 1:38:00 PM PDT -- AMac]
ReplyDeletescientist said...
Amac: I have some comments to you that are still held up in moderation (I guess will be forever) at CA. They are long and probably don't say anything important, but "just in case", will copy them here so you can read.
--------------------------
Response to your 093707AUG10 (me 122307AUG10)
Your comment is awaiting moderation.
1. I have a comment in moderation addressing the PMS issue and Mann reluctance to engage on substance.
2. I like your little math analyis on the variability changes over time. It puts what bothers about the big ramp into words. That said, should that be a formal screening mechanism? Can it be defended as such? Even if so, it’s not 2 plus 2 simple. We just need to be real careful that we are not harsh on BCPs and
3. I’m not so much trying to get you to be nice to Mike, although if you intend to indict him for not admitting something simple, you have to really show that it’s simple. I think the whole hammering on Mike to admit is a dry effort anyway. Better to argue the contamination issue, itself. Think about it this way, if you had that LANL director, me, Huybers, Curry, Huybers, Zorita, etc. and you wanted them to sign off that this thing was a slam dunk. they might want to really see that it was a slam dunk.
Little story: I remember Peilke and Boris(?) having some argument about some IPCC definition and if it was a mistake. And they were both all hot and bothered with each other and EACH thought he was not just right, but OBVIOUSLY right. And I read it real quick and certainly could not tell just from the blog debate who was right. I knew that I would have to read the source documents that they spoke of, maybe even have some context from being in the field to rule on who was right. So even though both thought it was crushingly obvious and not really even debatable, I couldn’t judge so quick. Well, maybe I’m dumb or wishywashy or something. But then an interesting thing happened. Zorita was appealed to (by Boris, I think, but no matter) to rule on the dispute. And I asked him to rule as well. Not because I think he needs to adjuticate all disputes. But because if it really were simple, he could do so quickly. Now this is a guy who has decades of climate science working exerience. And guess what he said…basically, what a bother, I’ll have to go read all the documents…it’s not obvious just from you two antagonists arguing in a blog.
Anyway…I think disconnecting the “what is correct science” from the “Mike fails to admit” (and not wasting toooo much time on the latter) is both better tactically for dealing with the Po-Mo scientist as an opponent AND better for the disinterested observer who cares about science.
—————-
I was gonna reply to the reviewer comment, but I didn’t have anything additive to say over previous.
---------------------------------
I think I had a post erased as well. Basically agreeing that Mike etc. won't agree to anything other than obvious errors (like a coding mistake) and even then the "I don't know button" will be pushed). It had a bit more substance to it than that. And a recommendation that they start lifting heavy to generate natural testosterone to counteract PMS.
scientist,
ReplyDelete> 2. I like your little math analysis on the variability changes over time... should that be a formal screening mechanism?
Yeah, maybe. At a more basic level, perhaps there is a temptation to view proxy candidates as mathematical objects (strings, matricies, whatever), rather than as representations of some actual physical property. As soon as you look at Lightsum in that light, you can hear it shouting, "I'm unsuitable!"
> 3. be nice to Mike...
As I've said elsewhere, Prof. Mann's conduct probably isn't all that many standard deviations away from the norm. Yes it would be better if he put on his BBPs. But part of the function of a scientific community is to ensure that scientific findings aren't distorted by the human shortcomings of its human members. There's (1) the actual issue of the proxies' uses, (2) what that means for the results of the paper, and (3) what all this indicates about the state of climate science. What the authors choose to do is #4, and what I think about their choice is somewhere between #5 and #(1/0).
[Originally posted by "scientist" to the "MikeN weighs in on Mann08's use of the Tiljander proxies" thread on August 10, 2010 9:57:00 AM PDT -- AMac]
ReplyDeletescientist said...
You are responsible for the methods you use in your paper. Saying that the reviewer made you use an incorrect method is bunk. It's your byline. You are responsible.
Your error bars for "the models" are unbeleiveably tight. You should have had enough physical intuition to know that did not make sense. That error is a real boner.
If correctly calculated, the error bars would still be different from observations, then you haven't SHOWN it in your paper.
Why not show the test both ways, if there was a debate about how to make the error bars?
What if the reviewer who told you to put the error bars down the wrong way was Douglas? What if a reviewer told you to put something down that said 2 plus 2 is 5? Would you?
Why is the correct method not shown in your SI?
You messed up, Steve. And didn't have the balls to let me criticize your paper last night. Crossposted at AMAC to display your censoring.
[Originally posted by "scientist" to the "MikeN weighs in on Mann08's use of the Tiljander proxies" thread on August 10, 2010 12:20:00 PM PDT -- AMac]
ReplyDeletescientist said...
Last night you censored me and I said nothing about "balls". I made an apt point about what "the models" predict and how just having some physical/mathematical intution should make you realize that those tight error bars in your figures are not meaningful. You said it had nothing to do with the topic! Heck even if I'm WRONG, it was way more on topic (getting into your figures) than all the peer review whining.
You just don't want your ideas technically criticized, McIntyre.
Oh...and if you group RSS and UAH together, the error bars of "the sattelites" are pretty big too and overlap "the models", too...
(cross posted for censorship prevention)
Lucia is moderating me now. What a coward.
ReplyDeletePrevious post was made before realizing that Steve has two MORE posts up at CA. The strange thing is he's introducting NEW analyses, not shown in his paper. As if it doesn't bother him, not to cover things in the paper and just wants to have an ongoing dialogue. He's also not really defending the key issue of Figure 2 and 3 too-small whiskers, but instead trying to redirect. Really shows how bad MMH was as a technical paper. (cross-posted for censorship check.)
ReplyDeleteCA crosspost
ReplyDeleteI'd respect you if you left the post open for comments and the picture up there. Putting a comment up that you will correct it eventually is fine.
Closing it is passive aggressive cowardice and abuse of your moderating powers to influence the debate. No attaboys from me, for your graceless admission.
CA crosspost
ReplyDeleteThere is no question of moderation. First the site has other moderators. Second, there is at least one busy thread currently (the whole site is not locked down).
The issue is McI does not have time for REBUTTALs. And also, that he doesn’t want to take his medicine for an error. This is abused of the moderator/publisher power to influence the debate.
Also, note that at the same time he admits a mistake, he takes a shot at his opponents for not doing so, and pats himself on the back.
You should just rename this the TCO whine page.
ReplyDeleteInteresting that Steve has time for new 300+ comment threads that attack his opponents, but closed comments on the one thread, where he made a mistake. I don't have an issue with him putting an update saying he was wrong, or that he will do a reanalysis. But stopping discussion is abuse of his moderator power.
Completed paint experiment
ReplyDeleteCompleted Basil solar silliness
Completed surface stations project
Opened Steve-mistaken-variance thread
---------------------
Or you could just link to semi-violent hockey analogies. Yuck, yuck! Go hoi polloi! Let's all grunt and hoot like the audience on the Mort Downey Jr. show.
http://en.wikipedia.org/wiki/Morton_Downey,_Jr.
scientist,
ReplyDeleteFigure 3 in this file at BitBucket, Holocene-ratios, shows that isotope ratio records in varves may contain a paleotemperature signal.
I guess you buckled down and got all her papers. I skimmed the paper, but I really need to print it to parse it. It's unfortunate that her thesis does not really tie all this stuff together. What I saw did not seem stickyouintheeye simple, so need to read the details to see if her inferences are based on strong info. Would also be good if she could somehow validate her earlier inferneces on LS, DS, XRD, etc. based on what the isotopes tell her.
ReplyDeleteI thin it is irrelevant whether Tiljander is correct or incorrect. It is possible to create hockey sticks without Tiljander. What matters is the behavior of the scientists involved. They have not argued for an alternative interpretation that Tiljander is incorrect. So what matters is whether they have properly implemented her interpretation, and when it was pointed out that they did not, how did they respond?
ReplyDeleteGiven their stonewalling and denial of the obvious, it lends credence to the idea that they cannot be trusted with other hockey stick studies as well.
Take a look at the Mcshane and Wyner response at RealClimate. Do you think it is reasonable for them to test its sensitivity to Tiljander, have that be one of their primary graphs?
I added pointers to PDFs of two more studies of varved lakebed sediments in the "Primary Links" post. One focused on Baffin Island glaciation, the other on the Younger Dryas transition, 12,700 years ago. Different strengths and weaknesses cf. Tiljander-type papers.
ReplyDeleteNot feeling any loving reason to read them.
ReplyDeleteCA cross-post (my comments are being pre-moderated, now)
ReplyDelete1. I think it's bizarre that you spend time on this gotcha game investigation fol-de-rol, when you have major outstanding science/math issues. That said it's your blog and life blablabla. [However, if we have the ability to make free comments, this is MY, relevant, comment.]
2. What's really galling is that you ERASED evidence of a mistake of yours (after "publishing" in your blog) and then locked discussion of the mistake. And said that discussion could resume only when you published new analysis.
3. I'm not saying you have to finish that analysis. I'm not even saying you have to be a part of the discussion. But you should put that previously published mistake BACK UP for the record. And you should unlock the thread an allow discussion. Otherwise this place is justg being run as a PR organ, and you're dishonest, and
"publishing in the blog" means nothing since you can alter or edit anything, (it's not a real record like a journal article.)
4. People like Gerg Burger, Ed Zorita, Judy Curry, etc. should have nothing to do with you. I urge them strongly to stay away from you...have fun with the hoi polloi and the Watts-like comment discussions.
[Crossposted in AMAC and Deep Climate]
AMac, you might be interested in this:
ReplyDeletehttp://rhinohide.wordpress.com/2010/10/13/smith-2010-elementary-reconstruction-of-the-hockey-stick-curve/
Ron links to his very informed commentary on an August 2010 paleotemperature reconstruction by RL Smith of UNC.
ReplyDeleteAMAC, the whole NEverending Audit is tedious, but I like you. The issue with the "red noise" is that McI essentially uses circular logic. He doesn't make "simple red noise series". What he uses are rather complex models that take the proxies themselves and try to abstract a noise structure from them (across all freuncies). The issue is that any signal will now be treated as noise. So a noise model based on that, is circular logic. Essentially it is ASSUMING LTP. Assuming the case to be proved. In addition McI was resistant to clarification of what he meant by "red noise". (Similar to the pattern of resisting description of the MMH whiskers, etc.)
ReplyDeletehttp://climateaudit.org/2006/09/28/climate-models-the-next-generation/
Scientist,
ReplyDeleteThanks for writing. I think you make some good points. I expect you kinda know what I'm going to say, but anyway --
I wish I'd tagged the comment a few weeks back where Gavin brought up the point you raise about red noise (I don't even remember where). It got me to thinking; it seems to me to be one of the keys to dendro.
* Treerings contain redness (autocorrelation of a time series).
* Instrumental temperature records contain redness.
* Treering series are affected by a lot of things. The biggie seems to be precipitation for most trees. In some settings, precip isn't usually limiting, so temperatures become the biggie. Then there are a lot of other macro-environmental things, like days of sunlight in the growing season, and whether the winter freeze was damaging. And micro effects--insect infestations, nearby treefalls, whatever. Some of this, we can call "noise." If what we care about is the temperature signal, I guess we can call everything else noise.
* Some treering time series but not others include temperature signals for part or all of their length.
So the question would be, where does the redness come from? Temp? Precip? Stuff we can't interpret, ie "noise"? If it's a mix (these things usually are), how is it apportioned? Is the mix steady for a given stand? Can a mix be a priori estimated for a naive stand?
Since dendros aren't dummies, I presume these things have been investigated. I don't know that literature. Maybe somebody will toss in a link to a fantastic review that makes it all clear.
But for now, I no longer presume that paleos treat these sorts of confounding factors with proper diligence. It seems to me that too many people are prone to direct their thinking down particular channels.
It's hard to implement the proper caution into scientific work. That's why I transcribed that talk by Norman Wolmark on breast cancer clinical trials and made it an early post. If you went back in time to the 1970s and stood up at a meeting to discuss the limitations of the governing philosophy of clinical trials, clinician-researchers would have treated you with condescension, then with scorn. Because they knew that they were on the cutting edge of science (which was true).
So on the one hand we have the neverending audit (catchy phrase), and on the other we have that same sort of condescension and scorn, underlain with sclerotic thinking.
The science is very interesting and I agree that the Mike/Gavin crowd is a bit political (both in science clique sense and in Hansen advocacy sense) and I hate the Herrdoktorprofessor attitude.
ReplyDeleteThat said, McI's way is not doing anything. He's been jerking around for literally the amount of time that someone could have written a Ph.D. and several fundamental papers. People like Burger, Huybers, VonStorch, Zorita, Christensen etc. need to pick away at things. And also, they need to publish when/if they find things that help Mann or his "cause". The McIntyre crap with the hoi polloi and the 4th tier market TV weatherman is just not....doing it. If you want an example of an amateur NAILING something in terms of methodlogy and clear communication (via a blog mostly), look to the British guy who found the small Met Office formula error. however, note how clear he is about explaining the level of impact and of not overselling things or running a stable with a bunch of forum commenters circle-jerking each other.
Man I am bored with the climate drama. Not angry at the sophistry. Not even an angry bored. Just bored, bored. Can't bring myself to read past the fold on Judy's blogs. Can't bring myself to read comment threads. I'm megabored.
ReplyDeleteAlibaba,
ReplyDeleteHeh.
This comment at Curry's was good, though.
I think I caught you saying some good things about Jeff Id. I think he's got some pretty big gaps (a bit of the Dunning Kruger kind where he does not see them). In terms of his data analyses, he has messed around for years, with lots of jumping to conclusions and trumpeting issues, and little final insight. He's also biased in terms of analyses.
ReplyDeleteSmall example:
http://noconsensus.wordpress.com/2009/06/09/tired-and-wrong-again/
(threads before and after are relevant as well.)
Basically, the guy is a typical blustery climate blogger skeptic. You want a smart, honest guy who has found some corrections? Check out John Graham-cumming.
I really don't think Moshpit and Lucia are that good either. They are much to inclined on their initial hypotheses and take years to re-evaluate (meanwhile spewing like crazy). Just not real curious and objective analysts. Not Feynmanian. Not Rickoverian.
Yeah...they like to "play" with code and equations and crap. good for them. But the way they sound off before analyses are done is pretty flawed.
The contrail thing was classic. Just pretty much epitomized the behavior pattern...
ReplyDeleteContrail thing? Link?
ReplyDeletehttp://noconsensus.wordpress.com/2010/11/09/hide-the-incline/
ReplyDeleteGood job. (Jonestown.) Keep it real, keep it fair.
ReplyDeleteHeh, at least two people read that, then...
ReplyDeleteMcIntyre is bad news. Look how he avoids discussion of the noise model decisions in his GRL article or of the "pick top 1%" mistake in the Wegman report. Those things are square in his area of interest, experience, and previous discussion. But nothing. The simple answer is he avoids discussing "his side's" errors. It's the attitude of a person who is more an amateur legalist (sealawyer) than a curious scientist/mathematician.
ReplyDeleteInstead we have McI concentrating on rehashing Climategate and playing to his Wattsupwiththat buddies and hoi polloi.
Stick to your petri dishes and business projects.
If you want a cool, smart, honest, curious guy; check out http://blog.jgc.org/ (do a search for his climate posts). This is a guy who I would trust to report findings that went either way in terms of inspecting climate work.
Posted at RC. Not sure if it will all make it through. I annoyingly have some mildly bad things to say about the team, mixed in with my comments on Id and ilk. Not out of sneakiness, or even false symmettry, that's just how I honestly see it, but as such this earns me lots of hate from the two tribes:
ReplyDeletePolyisTCOandbanned says:
Your comment is awaiting moderation.
10 December 2010 at 2:06 PM
1. Still waiting to read the paper. Looking at it, side by side, with Stieg will be interesting. Comments below are “gut” reactions on the whole kerfuffle, from a blog dispute follower perspective, fwiw:
2. From when I first understood what Steig had done, I thought it was a neat idea. Something definitely at least worth trying, and possibly something that would give us a better picture. The basic idea (I think) being to use sattelite period when we have HUGE information about how different weather patterns appear on the continent, to allow us to infer past temps at “blank spots” in the continent, back in the day when there was just surface stations.
3. It’s actually my LIKING of that cool concept that makes/made me worry about the approach of doing PCA and throwing away the stuff that does not go into the first three PCs. You lose the possibility of a lot of complicated patterns if you only have 3 PCs. Similar to mapping a complicated surface, to looking at a chemical compound, etc. 3 hybrid orbitals are better than 1, but may not be sufficient to understand some behaviors. For instance, let’s say there was some weather pattern that a spot on the continet would be warm when one other three other spots were warm, and two were cold (assume they are all far apart and somewhat independently variable). You could not follow that with the 3 PC approach. Of course, I don’t know that such complicated patterns exist. But if anything it would be nice if they did as it would give us more of a key to unlock a door…more of a “fingerprint”…to give us pattern detail of the past from this deft approach of Steig.
4. I’m interested in the PCs and their geographic appearance. I’m not a math jock, but from the beggining the PCs have always sort of reminded me of orbitals for electrons (the pictures), or perhaps those drum patterns you get when you take Engineering Mathematics and (try to) learn about Bessel functions and the like. I’m actually interested to learn how much Eric or others have thought about this issue and if there is any point, to pushing thinking about the analogies.
5. I’ve always been a little worried about some of the implicit, partial, hidden (pick a caveat, this is my impression!) assumptions about PCA by “the team”. I guess mostly Mike, since he’s the one who is most mathematically sophisticated given the math-phys background. It seems like there is often an approach of assuming that PCs represent something physical (a factor), but of course sometimes they may and sometimes they may just be a mathematical abstraction. Also, that one can take the important first few, throw out the rest of them and actualy sort of “clean” stuff up, almost like filtering noise out. It just makes me worry of the analogy of throwing away outliers versus doing a simple average or trend, with all the data. I think of PCA as being more suited to factor searching for, or perhaps data compression, or facial recognition, or operation of a Nate Lewis chemical nose. But not as a “filter”.
(length limit)
(rest of post)
ReplyDelete6. From the beginning, I’ve been annoyed by the approach on The Air Vent. First the incredible huge amount of noise to signal. All the different blind alleys and trial approaches. The mixing in of all kinds of triumphalism along the way, before the argument had even been refined. Not only is it annoying to read because of all the riffraff and forum games, but it’s just LONG.
6.5. A few examples. The krieging approach (did any of it make prime time?) Also the “negative thermometer” complaining. A quick think would show that mathematically an opposite signed predictor can happen in a data set. And even some examples of weather patterns, where one area being cold correlates to another being warm (hemispheres of the Earth, El Nino WESTPAC versus US SE, even the pictures of the jet stream reaching down the middle of the US and snaking around that we see on the evening news since at least the 70s; and note the last two could have somewhat of a time dynamic, that is more frequent than just the winter-summer simple example) make one think it might be possible in practice. And despite even several skeptics showing Jeff Id mathematically how he was wrong, he could not (or would not out of stubborness) get it. That was a time when RyanO failed me too. Here, he never directly corrected himself “I was wrong, I take it back” but instead said after interminable internet style repetitive debate and even writing to Jollife and getting a genuine PCA expert to grace their thread, Ryan came back with a “oh my view evolved during he thread, but I never really wanted to say so, you should figure it out”. A simple tiny point, but the real issue was the failure to understand a clarification and to admit it–not that it blew up their other concerns.
7. I’m curious why the pre-print was not circulated as a white paper. Is this JOC policy? Reading the pre-print, vice the blogorhea would be a way more efficient way to evaluate the counter-team’s thoughts. And those guys have a ready “bandwidth” and site and all to post it on. For that matter, given that they post the ongoing analysis (in blog posts), why not post the more refined and better communication?
I worry that this is a “tactical approach”. In other words, make a lot of blather-on forum posts, and then just submarine up papers without exposing them to criticism. For instance the recent MMH paper had some real (obvious) problems in defining standard deviation (even!) and in clearly discussing the hypothesis test. And in that case, it looks like a bunch of (unimportant to the key dispute) math was larded on to make it look fancy or new or distracting, rather than squarely focusing on the definitional and hypothesis test debate.
(post limit)
6. From the beginning, I’ve been annoyed by the approach on The Air Vent. First the incredible huge amount of noise to signal. All the different blind alleys and trial approaches. The mixing in of all kinds of triumphalism along the way, before the argument had even been refined. Not only is it annoying to read because of all the riffraff and forum games, but it’s just LONG.
ReplyDelete6.5. A few examples. The krieging approach (did any of it make prime time?) Also the “negative thermometer” complaining. A quick think would show that mathematically an opposite signed predictor can happen in a data set. And even some examples of weather patterns, where one area being cold correlates to another being warm (hemispheres of the Earth, El Nino WESTPAC versus US SE, even the pictures of the jet stream reaching down the middle of the US and snaking around that we see on the evening news since at least the 70s; and note the last two could have somewhat of a time dynamic, that is more frequent than just the winter-summer simple example) make one think it might be possible in practice. And despite even several skeptics showing Jeff Id mathematically how he was wrong, he could not (or would not out of stubborness) get it. That was a time when RyanO failed me too. Here, he never directly corrected himself “I was wrong, I take it back” but instead said after interminable internet style repetitive debate and even writing to Jollife and getting a genuine PCA expert to grace their thread, Ryan came back with a “oh my view evolved during he thread, but I never really wanted to say so, you should figure it out”. A simple tiny point, but the real issue was the failure to understand a clarification and to admit it–not that it blew up their other concerns.
7. I’m curious why the pre-print was not circulated as a white paper. Is this JOC policy? Reading the pre-print, vice the blogorhea would be a way more efficient way to evaluate the counter-team’s thoughts. And those guys have a ready “bandwidth” and site and all to post it on. For that matter, given that they post the ongoing analysis (in blog posts), why not post the more refined and better communication?
I worry that this is a “tactical approach”. In other words, make a lot of blather-on forum posts, and then just submarine up papers without exposing them to criticism. For instance the recent MMH paper had some real (obvious) problems in defining standard deviation (even!) and in clearly discussing the hypothesis test. And in that case, it looks like a bunch of (unimportant to the key dispute) math was larded on to make it look fancy or new or distracting, rather than squarely focusing on the definitional and hypothesis test debate.
(post limit)
8. There is also a LOT of whining about reviewer comments. Certainly, it might be possible for a reviewer to be unfair, but my general impression has been that authors whine more and are more often in the wrong in these kinds of disputes. I’ve always found being brutally clear (even to admitting dropping a sample on the floor) to assuage reviewers. It’s only when one tries to spin too hard, that they dig in. Also, have found that very clear writing and following the POSTED guidelines of a journal does a lot to get a paper through the wickets. Many authors could save themselves trouble if they just put a higher level of care (basically make the paper as perfect as you know how, imagine that the reviewers will fall dead and you need to get that paper perfect on your own) before submission, rather than dealing with all the back and forth letter writing. If you really do it right, you should be able to get a lot of papers accepted without revision.
ReplyDeleteI have to take the reviewer whining with a grain of salt, on the most recent counter-team paper. Of course it COULD be justified. But how can we tell unless we see the inital paper, final paper, and perhaps the correspondance? Based on the blog posts I’ve seen, some previous McIntyre submissions, writing conference pressentatations the night before a meeting, etc., I can’t be confident that the initial draft was even smooth. Of course, NicL and Ryan may have done a better job and they seem to have had more of a hand in this one. But…I’m explaining my Bayesian betting basis…
(post limit)
This is tedious. Please get rid of the 4000 character limit. I have to keep chopping it into peices and then part of it got through and now got taken away. I'll just frigging post over at Jeff Id's. It'll give him satisfaction, damnit, but I bet he lets it through. Grr.
ReplyDeleteScientist,
ReplyDeleteSorry about the character limit, but it's nothing I have set. This blog is a freebie from Blogspot/Google, so I get what I get. I find the inability to do much formatting in comments rather more annoying. You could sign up with Blogspot or WordPress as long as you link, you know, just a thought...
For whatever reason, some of your RC comment copy got tagged as spam. It's fished out and in the proper order now.
On reviewing, my experience (cell/molec biology) is very different from yours. Reviewers usually had some critical things to say and demands to make. When they didn't, it often looked as though they hadn't bothered to read the MS with care. As a reviewer, I never had the problem of "no suggestions for improvements for this submission." I always separated "major" from "minor", and when I thought all my criticisms were in the latter category, made a point of praising the authors' work in that regard.
Anyway.
RC's comments are sorta hopeless with their "like!" (in-it-goes) and "don't like!" (aggressive moderation/delay/snip) policy. They undercut their own claims of being a science-focused site; productive discussion is very unlikely under their ground rules. I rarely bother, but good luck.
Jeff Id's somewhat erratic in his wide-ranging choice of topics (and words), but I think he genuinely respects what the data are, and doesn't spin the math (wrt the posts I've been interested in). RyanO and Nic_L's commentaries, nice IMO.
My experience is probably pretty untypical. I had three papers in a row accepted without revision and my professor had 200 before that all needed revisions, and 400 as an editor that needed revisions. He actually did not understand the concept of just "accepted". That said, honestly, the time taken to really be super candid and super attention to detail ends up paying off. You spend less time on reviews later, because of the work done upfront. And it makes it easier for a reviwer to add value if the thing is as well done as possible. It drives me to distraction to have to review a paper with major issues on the science as well as being written in Chinglish.
ReplyDeleteId was all over the map on the Steig thing. Painful to read as it developed.
And it sounds like their paper went through huge changes (like they had fundamental stuff buried in the SI that didn't belong, sections that should have been in another paper, etc.) I bet that thing was no too smooth to start with, but more of some blog posts stuck together. Those guys get too used to self publishing and don't really understand how to raise their game.
No sweat on the character limit.
I've said it before and shouldn't have to repeat it, but something about the interminable style of chat on the net, makes repitition needed: The skeptics don't moderate completely honestly either, but at least you can get some stuff in and having no pre-moderation is a big improvement. RC is bad news with their moderation. At the end of the day, you almost need to get your own blog to participate. But at the end of the day, it's some jerkoff on the net anyhow. Lot of sturm and drang for not that much payoff.
ReplyDeleteThere's a lot of whining about the reviewers making them cut stuff on that paper, but from reading between the lines it sounds like they had a lot of disparate ideas of differing levels of development and that the inital paper was a bit of a "splat". Probably, they caused a lot of the delays on their own. I know how these guys operate and am not surprised. They just get used to very low standards from what they get away with on their blogging and then think that is quality work. Then get surprised in review.
ReplyDeleteBack on moderation. Guess those two posts you saw were too tough for Steve.
ReplyDeleteI have absolutely NOTHING wrong with a post, saying look, the adjustment changed. Nothing wrong with the thoughtful impulse to fasten in on red flags and examine them. So I mega-agree with your comment that it is reasonable to flag the concern.
ReplyDeleteThat said, it is a far, far cry from the wandering (even as a slam, it's just bizarre) snark-filled accusatory. And interesting that when the guy buckles down and looks at the Hansen paper, publicly available, that he finds no significant change in Hansen's methodology. Sure the input changed. But that is a far cry from "Hansen staying up working on adjustments". And again, it's at least POSSIBLE that the changed input is a more appropriate input.
I just can't grok this way of dealing with things. I'm ALL for digging into the target's 10K, etc. etc. Let's go...let's DIG. But let's keep an open mind. The current way of agitating hoi polloi and then never going back and upgrading the snark-filled half-accusatories (always with a plausible deniable, "I never said that" is just bizarre.) This kind of crap would never pass muster in a (good) company or the military or science.
Plus it's just wandering and disorganized and poorly stated. To the extent that comprehension is very seriously limited. Watts is definitely MORE of an idiot (much more, I have always praised Steve for code and math knowledge), but at least Watts is CLEAR!
Ryan is a punk and a whiner for getting into the peer review stuff.
ReplyDeleteHe's also disorganized in his presentation of material and wastes too much time with emotional venting. If his submitted paper was anything like his blog posts (and McI's usually are) then no wonder it took him so long to get something shaped into publishable form.
I've followed this kit and caboodle for 6 years now, but still find it a pain to look for what point you are making. Think how bad this looks to the intelligent outsider reading your stuff. If you want an example of people who do things right, look at Ed Zorita or that British blogger who's found some code glitches. They actually break things down and explain them. Not meander and vent and draw crappy graphs without axes and figure captions.
As usual, there is an issue of McI changing two things at once and ascribing the resultant change to one. (Or glossing over his muddling.) Way better to have done the full factorial and presented it.
And NONE of this exonerates ES. It's just that I can't even make out your point of technical disaccusation. I'm all for improving the science. All for taking down some liberal latte drinkers. But you guys screw it up so bad. You don't drive any understanding of material. You just have little social internet games.
Perhaps Jeff "contrail" Id will flex his chest in happiness over the dramah...or the Cold Fusion lovers at your "ally" Anthony Watts will just give you attaboys without trying to understand what the heck is going on. Probably Amac and Hugh will sigh at your disorganized rant.
(cross-posted at Amac for post preservation)
Hey TCO, thanks for tossing in your two cents, above. I'm guessing you submitted it to RyanO's Steig's Trick post. (The link provides context, hint, hint.)
ReplyDeleteYou might be interested in my take on what L'Affaire Steig/O'Donnell says about the peer review process, as Comment #68919 at Lucia's.
Good on ya, biologist.
ReplyDeleteThose guys step on their dicks so hard, man. It's been 6 years of McI silliness I've followed and he was at it a couple years before that. Long enough for someone to get almost 2 Ph.D.s. This shit is really a turnoff. Keep it real. (None of this is to say that Mann et all should not be scrutinized but the whole McI Watts set of them is full of shit.) I think Hugh has a clue too. The rest of them are pretty fucked up.
Stick to Zorita and such if you want someone smart and honest. McI is bad news.
I just read down. What a clusterfuck. Ryan never did the basic full factorial of method choices. He did not disaggregate issues to test their seperate effects.
ReplyDeleteIt's the same McI crap of changing 5 things at once (well at least 2) and then making a complaint. He's done it repeatedly and been slammed for it. But continues. What trash. To think Ryan is some sort of manufacturing engineer.
And the excuses about money and page limits are total horseshit. They had 2 years working on this. They could have had all the extra analsyses done, they could have had them on their own website, etc. But when even people on his "side" push for the basic sensitivity analsysis, he has to go run it NOW. The whole thing is bad news, bad engineering, bad quality. They ran a gazillion half-baked analyses that they trumpeted on a website for their yukyuks to egg them on for, but couldn't do the basic taking apart of the problem.
Move on.
Really underwhelmed with Ryan's "apology" (the titling of it, the placement with a bunch of excuses and other complaints, the having McI put another whine right on top of it). I remember a long time ago spending a bunch of time to pin Ryan down on a very small mistake. After a bunch of work to show the mistake and prove that he had made it, he said his view had evolved in the blog discussion, and that he had already taken back his point. But he never called it out as a correction.
ReplyDeleteI find this behavior dishonest and cowardly. Very sea lawyer. Very unmanly. Would not want to serve with such.
Move on, Amac. You are better than this crew. People like Zorita and such can correct Mann and their over-reaches. But you don't want to hang with the McI, Mosh, Lucia crew. They're second raters.
This comment has been removed by a blog administrator.
ReplyDeletePolytroll said (MAy 22, 2011 5:56:00 PM PDT)
ReplyDeleteLook at the [snip] [snip] "skeptics" lately. they can't really scrunch down and descend a testicle to actially call out Wegman for his plagiarism or Watts for his years of false snark. But they don't want to be caught with really supporting it either. What [snip]. they're not even Americans.
Polytroll,
ReplyDeleteC'mon, "please be civil." E.g. no obscenities.
Re: Wegman, I've commented on his plagiarism and related problems at Lucia's and elsewhere. Seems to me that Deep Climate and John Mashey have made a very substantial case. Sauce, goose, gander.
Re: Watts, I don't read WUWT very often and have little to contribute.
I wouldn't post here if I didn't think you were a pleasant surprise. But Watts, McI, (and Lucian and Moshpit) have a long history of evasion and of looking the other way. The whole scene is bad news. Bunch of losers.
ReplyDeleteTypical pattern. The guy asks for help on his submission with hours to go (for something he had 60 days to work on). I've seen him do this several times before and as in this case, his submissions were spotty. He doesn't help sympathy that he is being "put down by the man" when he is so repeatedly sloppy.
ReplyDeleteIs there anything new in the latest McI post? I skimmed it and it seemed like a total repeat? Could he maybe put any new parts in red or something? It's boring and confusing and makes me not even read what he has to say. Even the video snark was a rerun.
ReplyDeleteP.s. And just re-running stuff is not intellectual engagement, it's drum-banging.
It's been a year since McI closed the comments and erased a mistake of his in a thread, promising to redo the calcs. But he has not done so. He has also failed to engage with the other substantive criticism of the MMH paper itself.
ReplyDeletePosted in the Hu thread at CA (post preservation, in moderation):
ReplyDeleteHu:
I’m generally about 80% sympathetic to you on all things and about 90% on Mann doing you a dirty, here. Certainly you play more fair than McI and at least put things into a form (papers) that they can be engaged with. and I can follow you (not a meandering snark full of mistake-laden graphs, written in mystery story style, to the extent I can’t even tell what the assertion is). In other words, you are much more straightforward and academic.
Few issues though:
1. I would avoid “plagiarism” as a loaded word and as not precisely correct. They failed to credit you for an idea. This is different than using prose from soemone else. It was a failure to acknowledge/cite. McI repeats the same error here again wrt “Harry”. I have pointed this out before…
2. It is over the top to call Odonnel et al a “refutation” of Steig. I would be more measured (for example, the paper is more measured). It is probably halfway between a total refutation and a minor improvement in the same vein (as alleged by some of the Steigers). Maybe significant upgrade or some such word (not sure). But definitely not “refutation”. Talking that way is kind of stooping into the Michael Mann (the error is all theirs) type of rhetoric, which is a little gamey debatey stuff. not real science/math logic/disaggregation.
3. I advise avoiding EnE for publications. Do Climate of the Past. Even a very crappy paper is basically exposed and published as a submission. Even if you don’t get accepted, it is still displayed. You will not get a good peer review at EnE (look at the fellow you helped who had to do a total rewrite of a paper…how did that thing sail out, when I could see all kinds of issues just from being a blog-reader).
"1. I would avoid “plagiarism” as a loaded word and as not precisely correct. They failed to credit you for an idea. This is different than using prose from soemone else. It was a failure to acknowledge/cite. McI repeats the same error here again wrt “Harry”. I have pointed this out before…"
Deletewrong. Plagiarism is not restricted to copying the words of another. One reason we require students to supply a complete bibliography of everything they read when writing a paper is so that we can insure that they credit
other for the IDEAS that they got from reading others. Here is source:
http://www.plagiarism.org/plag_article_what_is_plagiarism.html
scientist, I will take it that you have never detected a plagiarism or served on a committee defining what constitutes plagiarism or brought a plagiarism case against a student. You do not know what you are talking about. I suggest silence.
At Climate Audit --
ReplyDeleteAMac Posted Mar 5, 2012 at 1:54 PM
TCO/scientist is apparently on moderation here. He cross-posted some critical remarks at my blog that are civil and topical.
cut and paste from Climate Audit (my comments are not allowed on there).
ReplyDeleteAmusingly enough I was commenting to a thread where McI complains about moderation.
He is such a weak tit.
Still has not supplied the data and code for the Watts-McI paper.
Even Moshpit pushed him (kid took a few years, but he got to where I did in a few months).
"bad news bear
Posted Sep 10, 2012 at 10:10 PM | Permalink | Reply
Your comment is awaiting moderation.
Could you make a new post complaining about the proprieter of Climate Audit shutting down discussion (locking comments) when he had effed up an MMH analysis?
And also how he promised to redo the analysis…then didn’t for over a year now?"