Auditing the auditors
Per
article, which uses proxies to map earth's surface temperature for the
last ~500 years. Another paper, McIntyre and McKitrick, tried to
replicate the MBH results from their published paper, and claimed to
be unable to do so. On this board, a whole host of contributors
claimed to be able to replicate the original MBH article, and some
provided ad hominem abuse of M&M.
You may be interested to read that Mann, Bradley and Hughes have now
published an extensive correction in Nature (Vol. 430, p.105),
apparently in response to a complaint.
I post the following from Ross McKitrick's website:
UPDATE: July 1 2004:
The Corrigendum in Nature today (July 1, 2004) by Professors Mann,
Bradley and Hughes is a clear admission that the disclosure of data
and methods behind MBH98 was materially inaccurate. The text
acknowledges extensive errors in the description of the data set. Even
more important is the new online Supplementary Information (SI) site,
which concedes for the first time that key steps in the computations
behind MBH98 were left out of (and indeed conflict with) the
description of methods in the original paper.
These items were published on the instruction of the Editorial Board
of Nature in response to a Materials Complaint that we filed in
November 2003. That our
complaint was upheld and the Corrigendum was ordered represents a
vindication of our view that, prior to our analysis, there had been no
independent attempt to verify or replicate this influential but deeply
flawed study, something which was forestalled, at least in part, by
inadequate and inaccurate disclosure of data and methods.
This is only the first step in resolving the dispute we initiated last
fall. The Corrigendum and the SI contain the gratuitous claim that the
errors, omissions and misrepresentations in MBH98 do not affect their
results. If this were true, then a simple constructive proof could
have been provided, showing before and after calculations. This is
conspicuously missing from the Corrigendum and the new SI. We have
done the calculations and can assert categorically that the claim is
false.
We have made a journal submission to this effect and will explain the
matter fully when that paper is published.
Further, detailed comments on the Corrigendum and new SI will be
released shortly.
************************
per
Ian St. John
> we had an interesting discussion about the Mann, Bradley and Hughes
> article, which uses proxies to map earth's surface temperature for the
> last ~500 years. Another paper, McIntyre and McKitrick, tried to
> replicate the MBH results from their published paper, and claimed to
Why the focus on a single early result? Sure there were slips in the
preparation. No paper if perfect, and usually any 'replication' leads to
discussions of a cordial nature that resolves any confusion on the details.
It was the confrontational nature of M&M from the getgo ( and illustrating
both their incompetence and their agenda ) that led to their failure, not
the paper itself.
Of more importance is the fact that there are dozens of papers on
reconstructing the NH temperature from such proxies and all of them agree in
large part on the results. There seems to be some sort of attempt to attack
this one preliminary result in order to cast doubt on the entire field. It
will fail. Too many competent researchers have cross checked the data.
Note as well that the Mann,et al 1998 paper was not used in the IPCC results
which were *derived* from a later paper published in 1999 with most of the
early stumbles fixed.
David Ball
>we had an interesting discussion about the Mann, Bradley and Hughes
>article, which uses proxies to map earth's surface temperature for the
>last ~500 years. Another paper, McIntyre and McKitrick, tried to
>replicate the MBH results from their published paper, and claimed to
>be unable to do so. On this board, a whole host of contributors
>claimed to be able to replicate the original MBH article, and some
>provided ad hominem abuse of M&M.
>
>You may be interested to read that Mann, Bradley and Hughes have now
>published an extensive correction in Nature (Vol. 430, p.105),
>apparently in response to a complaint.
>
>I post the following from Ross McKitrick's website:
>
Tell me, does E&E have a similar process to their peer-review
process. It would be interesting to take M&M to task for their shoddy
"audit". Imagine claiming to audit a paper by patently not following
the author's methodology.
w...@bas.ac.uk
>These items were published on the instruction of the Editorial Board
>of Nature in response to a Materials Complaint that we filed in
>November 2003. That our
All of this appears to be a twisting of the facts to make M&M appear
more powerful than they are...
>The Corrigendum and the SI contain the gratuitous claim that the
>errors, omissions and misrepresentations in MBH98 do not affect their
>results.
According to quark soup http://davidappell.com/archives/00000149.htm
MBH are saying that the corrigendum makes no difference because all it
affects is the listing of the datasets, a point M&M don't seem to have
understood.
-W.
--
William M Connolley | w...@bas.ac.uk | http://www.antarctica.ac.uk/met/wmc/
Climate Modeller, British Antarctic Survey | Disclaimer: I speak for myself
I'm a .signature virus! copy me into your .signature file & help me spread!
Per
> Of more importance is the fact that there are dozens of papers on
> reconstructing the NH temperature from such proxies and all of them agree in
> large part on the results. There seems to be some sort of attempt to attack
> this one preliminary result in order to cast doubt on the entire field. It
> will fail. Too many competent researchers have cross checked the data.
there are dozens of papers on such proxies ?
MBH'98 is now a PRELIMINARY result ?
There have been "cross-checks" of the MBH data set ?
waiting eagerly for the references which validate these assertions :-)
> Note as well that the Mann,et al 1998 paper was not used in the IPCC results
> which were *derived* from a later paper published in 1999 with most of the
> early stumbles fixed.
So perhaps you can tell me where it says that early stumbles were
fixed in the '99 paper, or anything remotely similar ?
per
Per
> All of this appears to be a twisting of the facts to make M&M appear
> more powerful than they are...
you know, I don't have the data to be able to evaluate that claim.
Perhaps you have access to the correspondence between Nature, and
either M&M or MBH ?
> >The Corrigendum and the SI contain the gratuitous claim that the
> >errors, omissions and misrepresentations in MBH98 do not affect their
> >results.
>
> According to quark soup http://davidappell.com/archives/00000149.htm
> MBH are saying that the corrigendum makes no difference because all it
> affects is the listing of the datasets, a point M&M don't seem to have
> understood.
err, that would be the listing of the datasets, new datasets not
previously disclosed, datasets being excluded, and start and end dates
which were not previously disclosed ? And that's not even beginning to
cover the "enhanced" coverage of the methods they have now provided ?
I have a slightly different analysis to yours. I think that M&M
understand perfectly well the words that MBH have used, because they
cite them. Moreover, I think they are attempting to directly
contradict MBH.
I look forward to a resolution :-)
per
w...@bas.ac.uk
>w...@bas.ac.uk wrote in message news:<40e8...@news.nwl.ac.uk>...
>> All of this appears to be a twisting of the facts to make M&M appear
>> more powerful than they are...
>you know, I don't have the data to be able to evaluate that claim.
>Perhaps you have access to the correspondence between Nature, and
>either M&M or MBH ?
No, hence "appears". M&M are trying to say that Nature "ordered" MBH to
do something. There is nothing to support that claim.
>> >The Corrigendum and the SI contain the gratuitous claim that the
>> >errors, omissions and misrepresentations in MBH98 do not affect their
>> >results.
>>
>> According to quark soup http://davidappell.com/archives/00000149.htm
>> MBH are saying that the corrigendum makes no difference because all it
>> affects is the listing of the datasets, a point M&M don't seem to have
>> understood.
>err, that would be...
Sigh, lets try again: MBH are saying that the corrigendum that they have
just published doesn't affect the results as previously published. And QS
is saying that the reason for that is that the corrigendum is just about
correcting the list of datasets to be the same as the ones actually used.
Ian St. John
> "Ian St. John" <ist...@noemail.ca> wrote in message
> news:<GKSFc.73$JG5....@news20.bellglobal.com>...
>> Of more importance is the fact that there are dozens of papers on
>> reconstructing the NH temperature from such proxies and all of them
>> agree in large part on the results. There seems to be some sort of
>> attempt to attack this one preliminary result in order to cast doubt
>> on the entire field. It will fail. Too many competent researchers
>> have cross checked the data.
>
> there are dozens of papers on such proxies ?
> MBH'98 is now a PRELIMINARY result ?
> There have been "cross-checks" of the MBH data set ?
> waiting eagerly for the references which validate these assertions :-)
http://www.grida.no/climate/ipcc_tar/wg1/067.htm#232
Actually, they had one earlier paper in 1995. You seem to forget that all
science is 'preliminary' in terms that each new paper improves on the past.
As to the other papers, you can start at the page above which gives
reference to some of them..
>
>> Note as well that the Mann,et al 1998 paper was not used in the IPCC
>> results which were *derived* from a later paper published in 1999
>> with most of the early stumbles fixed.
>
> So perhaps you can tell me where it says that early stumbles were
> fixed in the '99 paper, or anything remotely similar ?
> per
Time to move on Mr. Per. You are obviously not interested in the progress of
science, but exist to spin and shit.
Psalm 110
> we had an interesting discussion about the Mann, Bradley and Hughes
> article, which uses proxies to map earth's surface temperature for the
> last ~500 years. Another paper, McIntyre and McKitrick, tried to
> replicate the MBH results from their published paper, and claimed to
> be unable to do so. On this board, a whole host of contributors
> claimed to be able to replicate the original MBH article, and some
> provided ad hominem abuse of M&M.
>
> You may be interested to read that Mann, Bradley and Hughes have now
> published an extensive correction in Nature (Vol. 430, p.105),
> apparently in response to a complaint.
Here are the Global Warming counterfacts (partial list):
* 1992 -- "Warning to Humanity" petition circulates, signed by
1,700 working CREDIBLE scientists.
* 1992 -- Oil, Coal, pollution-industries counter-attack at Rio
conference, fake Heidelberg Appeal circulates with their funding,
using their paid-science-traitors, like the documented corrupt Fred
Singer, documented corrupt Fred Seitz, documented corrupt Bruce N.
Ames.
* 1997-1998 -- hottest ocean temperatures ever recorded in human
history.
* 1998 -- hottest year in recorded human history.
* 1998 -- 86% of all corals bleached, came 2 degrees from known
heat-death of species, and 10 days away from known heat-associated
digestion failure starvation deaths. No fossil record shows such
massive simultaneous incident ever in global geologic history other
than 5 mass extinction incidents.
* 1999 -- strongest super-cyclone (310km/h winds) in human
recorded history hits Orissa, India. 1.9 million houses damaged or
destroyed, 10,000 people swept out to drown at sea.
* 2002 -- Severe droughts on five continents simultaneously -- at
one moment the crops to feed one sixth the human race are threatened
by brutal killing weather.
* 2002-2003 -- Worst drought in Australia recorded human history
leads to wildfires invading Austrialian capital city and burning down
hundreds of city blocks. At one moment not long afterwards a 250
kilometer firefront threatened to engulf the entire largest city in
that country, Sydney.
* 2003 -- More people displaced by flooding than any year in
recorded human history, over twice the geographical area flooded than
second-worst flood disaster year in recorded human history.
* 2003 -- 562 tornadoes sweep the USA in ten days -- largest swarm
of tornadoes ever recorded in US history, almost three times worse
than previous historical record.
* 2003 -- Largest hailstone in human recorded history falls in
Nebraska, 18 inches diameter, "cannonball size" which denialists say
is impossible according to THEIR understanding of physics.
Canteloupe-sized, softball-sized, baseball-sized, golf ball-sized all
fell in 2003 hell storms.
* 2003 -- Weeks of unrelenting pounding of constant storms finally
break northeast US power grid, 50,000,000 utility customers blacked
out in US and Canada. Indiana governor, Frank O'Bannon, dies of heart
attack from stress of one declared sequential "state of emergency"
after another. Indiana, Ohio, Pennsylvania, Illinois, Toronto
especially hit hard.
* 2003 -- Worst forest firestorms in Portugal's recorded human
history, blamed on excessive heat and changed air currents denying
rainfall.
* 2003 -- 20,000 forest fires in Russia, 16,000 of them before the
heat of summer, the largest number in human recorded history.
* 2003 -- 1,400 heat-stroke deaths in Andhra Pradesh, India from
sustained heat wave and delayed monsoon -- largest mass heat-killing
in local recorded human history.
* 2003 -- 35,000 excess heat-stroke deaths across Europe, largest
mass heat-killing in European recorded history.
http://www.ecosyn.us/adti/ADTI_Frauds_01.html
http://www.ecosyn.us/adti/AdTI_Villians.htm
http://www.ecosyn.us/adti/Pelosi.html
http://www.ecosyn.us/adti/Singer-1993-1994.html
http://www.ecosyn.us/adti/Singer-Seitz.html
http://www.ecosyn.us/adti/Singer-Nightline.html
http://www.ecosyn.us/adti/Stohrer-Singer.html
http://www.ecosyn.us/adti/Hazeltine-Singer.html
http://www.ecosyn.us/adti/Heidelberg-Appeal.html
http://www.ecosyn.us/adti/Confronting_AdTI.html
http://www.ecosyn.us/adti/Seitz_Tobacco_Crimes.html
http://www.ecosyn.us/adti/Becky_Norton_Dunlop_AdTI.html
James
"Ian St. John" <ist...@noemail.ca> wrote in message
news:GKSFc.73$JG5....@news20.bellglobal.com...
James
Per
> Per wrote:
> > "Ian St. John" <ist...@noemail.ca> wrote in message
> > news:<GKSFc.73$JG5....@news20.bellglobal.com>...
> >> Of more importance is the fact that there are dozens of papers on
> >> reconstructing the NH temperature from such proxies and all of them
> >> agree in large part on the results. There seems to be some sort of
> >> attempt to attack this one preliminary result in order to cast doubt
> >> on the entire field. It will fail. Too many competent researchers
> >> have cross checked the data.
> >
> > there are dozens of papers on such proxies ?
> > MBH'98 is now a PRELIMINARY result ?
> > There have been "cross-checks" of the MBH data set ?
> > waiting eagerly for the references which validate these assertions :-)
>
> http://www.grida.no/climate/ipcc_tar/wg1/067.htm#232
>
> Actually, they had one earlier paper in 1995. You seem to forget that all
> science is 'preliminary' in terms that each new paper improves on the past.
> As to the other papers, you can start at the page above which gives
> reference to some of them..
so far, you have managed to cite a paper with 8 references, 4 of which
belong to Mann. That's kind of falling short of your assurance that
there are dozens of papers, and that there are cross-checks on the MBH
data set.
You have a fascinating perspective on the "preliminary" nature of
science. My view is somewhat different; when authors publish a paper,
it should be robust and reproducible.
> >> Note as well that the Mann,et al 1998 paper was not used in the IPCC
> >> results which were *derived* from a later paper published in 1999
> >> with most of the early stumbles fixed.
> >
> > So perhaps you can tell me where it says that early stumbles were
> > fixed in the '99 paper, or anything remotely similar ?
> > per
>
> Time to move on Mr. Per. You are obviously not interested in the progress of
> science, but exist to spin and shit.
fascinating logic. I trust you are able to read from the exchange
above that I made no spurious assertions, nor did I spin or shit. I
merely asked if you could justify your claims, in the hope that I
might learn something. It was in fact yourself that made the assertion
that there were early stumbles, and that the later paper fixed those
early stumbles.
I note that you haven't answered the question.
per
Per
> Sigh, lets try again: MBH are saying that the corrigendum that they have
> just published doesn't affect the results as previously published. And QS
> is saying that the reason for that is that the corrigendum is just about
> correcting the list of datasets to be the same as the ones actually used.
you know, until M&M published, the MBH paper stood unimpeached and
no-one had any reservations. When M&M published their audit, the
specific charges they made included : that the dataset MBH used could
not be established from the paper, and that there were a whole variety
of data problems, and a whole variety of methods information that
needed to be provided.
It would appear that this corrigendum vindicates an important aspect
of the M&M paper; MBH have accepted the validity of the charges above,
and have brought their work up to the minimum standard required for
publication with this corrigendum.
I don't see there is anything trivial about this, and Nature wouldn't
have published the correction if it was trivial. For a data analysis
paper, actually listing the datasets used, as opposed to a half-baked
fudge of datasets with incorrect listings, missing data and incorrect
data, seems to me to strike to the very heart of the integrity of this
scientific publication.
It strikes me that M&M have actually provided an important public and
scientific contribution already.
per
NobodyYouKnow
> w...@bas.ac.uk wrote in message news:<40e9...@news.nwl.ac.uk>...
>> Sigh, lets try again: MBH are saying that the corrigendum that they
>> have just published doesn't affect the results as previously
>> published. And QS is saying that the reason for that is that the
>> corrigendum is just about correcting the list of datasets to be the
>> same as the ones actually used.
>
> you know, until M&M published, the MBH paper stood unimpeached and
> no-one had any reservations.
And still don't. It passed peer review and there has not been shown to any
substantive errors. The trouble with two clowns trying to reproduce the
paper and the methodology is mostly the problem of the two clowns.
Ian St. John
There are a lot more references than that, and you forget I said 'start'.
Guess even that is beyond you.
>
> You have a fascinating perspective on the "preliminary" nature of
> science. My view is somewhat different; when authors publish a paper,
> it should be robust and reproducible.
Such as the MBH'98 paper which is both robuts and reproducable. This has
nothing to do with glitches in the listing of files. As time passes, such
things become moved, updated, etc. Any real audit should have taken place
during the peer review. The problems are the M&M are not educated in the
field of multiproxy dendrochronology and haven't a clue how to use the
knowledge that is given. Nor are their motives beyond question which is why
they published such a half assed job in journal dedicated to promotion of a
warped viewpoint.
>
>
>>>> Note as well that the Mann,et al 1998 paper was not used in the
>>>> IPCC results which were *derived* from a later paper published in
>>>> 1999 with most of the early stumbles fixed.
>>>
>>> So perhaps you can tell me where it says that early stumbles were
>>> fixed in the '99 paper, or anything remotely similar ?
>>> per
>>
>> Time to move on Mr. Per. You are obviously not interested in the
>> progress of science, but exist to spin and shit.
>
> fascinating logic. I trust you are able to read from the exchange
> above that I made no spurious assertions, nor did I spin or shit.
On the contrary. All of your claims are basically shit disturbing with no
real purpose in terms of the science or understanding.
> I
> merely asked if you could justify your claims, in the hope that I
> might learn something. It was in fact yourself that made the assertion
> that there were early stumbles, and that the later paper fixed those
> early stumbles.
And manage not even to find the first paper in a long series of similar ones
that reach the same conclusion from the same data. The 'cross checking' is
crucial to knowing that the finding are solid.
>
> I note that you haven't answered the question.
I answered all questions. I notice that you have nothing to offer. That is
not surprising since your whole post is based on other peoples spin and
shit.
Thomas Palm
news:bd7c1148.04070...@posting.google.com:
> w...@bas.ac.uk wrote in message news:<40e9...@news.nwl.ac.uk>...
>> Sigh, lets try again: MBH are saying that the corrigendum that they
>> have just published doesn't affect the results as previously
>> published. And QS is saying that the reason for that is that the
>> corrigendum is just about correcting the list of datasets to be the
>> same as the ones actually used.
>
> you know, until M&M published, the MBH paper stood unimpeached and
> no-one had any reservations.
Contrarians had complained about it since it first appeared, but only in
general terms. Either all other contrarians were so incompetent they
weren't able to check the result even if they didn't believe it, or M&M
have done something wrong.
Even in serious science the MBH paper has been discussed and people have
argued about the difference between this and other reconstructions. These
differences are a lot smaller than claimed by M&M, though, and are more
of an academic interest.
> It would appear that this corrigendum vindicates an important aspect
> of the M&M paper; MBH have accepted the validity of the charges above,
> and have brought their work up to the minimum standard required for
> publication with this corrigendum.
What do you think is a "minimum standard"? That you get everything right?
(M&M certainly didn't) Sorry, almost no papers do that, even very famous
papers tend to contain lots of errors, some of them even significant. As
long as the main result stands the test of time the paper still is
considered relevant. By the time the science has been rehashed a couple
of times and starts appearing in textbooks the errors may have been
sorted out so you get the real truth.
> I don't see there is anything trivial about this, and Nature wouldn't
> have published the correction if it was trivial. For a data analysis
> paper, actually listing the datasets used, as opposed to a half-baked
> fudge of datasets with incorrect listings, missing data and incorrect
> data, seems to me to strike to the very heart of the integrity of this
> scientific publication.
I think anyone who has published any paper of his own will understand. I
even discovered an error in the erratum of one of my advisor's most cited
papers. (Not that either the erratum or the error in it changed the
result).
> It strikes me that M&M have actually provided an important public and
> scientific contribution already.
Had they done an independent analysis and come up with a different and
correct result they would have done an important service. As it is they
have been more interesterd in muddying the water than with improving the
science, and any benefits are purely incidental.
w...@bas.ac.uk
>...and Nature wouldn't have published the correction if it was trivial.
True.
>For a data analysis
>paper, actually listing the datasets used, as opposed to a half-baked
>fudge of datasets...
But here you start ranting.
Note that you have ignored my (QS's) point: the corrigendum *doesnt
change the results* just the dataset listing.
Per
> Per wrote:
<snip>
> > you know, until M&M published, the MBH paper stood unimpeached and
> > no-one had any reservations.
>
> And still don't. It passed peer review and there has not been shown to any
> substantive errors. The trouble with two clowns trying to reproduce the
> paper and the methodology is mostly the problem of the two clowns.
On the contrary; I know for a fact that the original methodology
section in the MBH paper was inadequate, and in some instances, plain
wrong. I know this for a fact because MBH have now published a
corrigendum, which corrects and extends what was in their original
paper.
You may sneer at "two clowns"; but one of the principal conclusions
from M&M was that the methodology of MBH as reported was inadequate.
MBH agree with that conclusion, because they have issued a
corrigendum.
Now the funny thing is that you want to insult two guys who have made
a valuable contribution to the scientific process. Perhaps this is
"shooting the messenger" ?
per
Per
> process. It would be interesting to take M&M to task for their shoddy
> "audit". Imagine claiming to audit a paper by patently not following
> the author's methodology.
Just out of interest, you do realise that this thread is a result of a
corrigendum by MBH ? And that MBH specifically clarify what their
methodology was in this corrigendum ? You do realise that there is a
whole load of methods which wasn't in the original paper ?
You do realise that MBH are effectively admitting that that their
original description of their methodology was not adequate ?
You do realise that you cannot follow an inadequate methodology ?
just checking
per
David Ball
>w...@bas.ac.uk wrote in message news:<40e9...@news.nwl.ac.uk>...
>> Sigh, lets try again: MBH are saying that the corrigendum that they have
>> just published doesn't affect the results as previously published. And QS
>> is saying that the reason for that is that the corrigendum is just about
>> correcting the list of datasets to be the same as the ones actually used.
>
>you know, until M&M published, the MBH paper stood unimpeached and
>no-one had any reservations. When M&M published their audit, the
>specific charges they made included : that the dataset MBH used could
>not be established from the paper, and that there were a whole variety
>of data problems, and a whole variety of methods information that
>needed to be provided.
It still is unimpeached and M&M did no audit of any kind. That
would have required them to follow the author's methodology which they
clearly didn't do.
>
>It would appear that this corrigendum vindicates an important aspect
>of the M&M paper; MBH have accepted the validity of the charges above,
>and have brought their work up to the minimum standard required for
>publication with this corrigendum.
>
>I don't see there is anything trivial about this, and Nature wouldn't
>have published the correction if it was trivial. For a data analysis
>paper, actually listing the datasets used, as opposed to a half-baked
>fudge of datasets with incorrect listings, missing data and incorrect
>data, seems to me to strike to the very heart of the integrity of this
>scientific publication.
Interesting. I wonder why that same standard isn't applied to
what E&E published?
>
>It strikes me that M&M have actually provided an important public and
>scientific contribution already.
>
How? Nothing that has happened has invalidated MBH and
certainly nothing that has happened has validated the efforts of M&M.
Per
> Per <perox...@ntlworld.com> wrote:
> >For a data analysis
> >paper, actually listing the datasets used, as opposed to a half-baked
> >fudge of datasets...
>
> But here you start ranting.
>
> Note that you have ignored my (QS's) point: the corrigendum *doesnt
> change the results* just the dataset listing.
Dear W
if you think you can replicate a data analysis paper WITHOUT knowing
what the datasets used were, you are entirely entitled to call the
above "a rant".
For the record, it is not *just* the dataset listing; it is also a
considerable amount of methods information which wasn't originally
included.
Also for the record, I have cited MBH's claim that the corrigendum
doesn't change the results. Didn't QuarkSoup also carry some extremely
scathing commentary before on this issue before, which was shown to be
wholly wrong ?
yours
per
Per
> What do you think is a "minimum standard"?
previously published was not at that minimum standard.
> > It strikes me that M&M have actually provided an important public and
> > scientific contribution already.
>
> Had they done an independent analysis and come up with a different and
> correct result they would have done an important service. As it is they
> have been more interesterd in muddying the water than with improving the
> science, and any benefits are purely incidental.
M&M did an independent analysis of the MBH paper
They came up with results which differ from MBH.
They also concluded that the dataset description and methodology of
mbh were inadequate.
This latter has been shown to be true with MBH's corrigendum.
I think you don't like the message.
per
David Ball
>David Ball <wra...@mb.sympatico.ca> wrote in message news:<6a0ge0df82j2dkfnv...@4ax.com>...
>> Tell me, does E&E have a similar process to their peer-review
>> process. It would be interesting to take M&M to task for their shoddy
>> "audit". Imagine claiming to audit a paper by patently not following
>> the author's methodology.
>
>Just out of interest, you do realise that this thread is a result of a
>corrigendum by MBH ? And that MBH specifically clarify what their
>methodology was in this corrigendum ? You do realise that there is a
>whole load of methods which wasn't in the original paper ?
Just out of curiosity, why is your approach to the science so
uneven? What's sauce for the goose. If you are willing to take MBH to
task for disclosure problems, why aren't you taking M&M to task as
well?
>
>You do realise that MBH are effectively admitting that that their
>original description of their methodology was not adequate ?
So what? That's always the danger when experts are writing for
other experts. There's a tendency to gloss over what you assume to be
trivial points. Sometimes those trivial points aren't. I notice you've
been ducking the issue of the validity of MBH. Why?
>
>You do realise that you cannot follow an inadequate methodology ?
>
No, YOU cannot follow an inadequate methodology. The
methodology of MBH has been used since then. Apparently, those authors
had no trouble. What you're really saying is that non-scientists
cannot follow the methodology. OK. Since the paper was not produced
with them in mind, I have little trouble with it. That they've
hopefully filled in some of the blanks will make it a little easier to
follow, but that doesn't for a second invalidate the RESULTS of MBH
nor does it validate those of M&M, something you patently refuse to
address.
Per
> Such as the MBH'98 paper which is both robuts and reproducable.
will you please cite the reference which has independently reproduced MBH '98 ?
Perhaps you are all abuse, and no citation ?
per
Thomas Palm
> Thomas Palm <Thoma...@chello.removethis.se> wrote in message
> news:<Xns951E86CDBEFAAT...@212.83.64.229>...
>> What do you think is a "minimum standard"?
> the issue is that MBH accept with their corrigendum that what they
> previously published was not at that minimum standard.
No, it only proves it wasn't perfect. As I already explained essentially
no paper is perfect. To what extent people worry about corrections depend
on whether they see there is an interest for one. Had Nature withdrawn
the paper it would have showed it was below the minimum standard.
>> > It strikes me that M&M have actually provided an important public
>> > and scientific contribution already.
>>
>> Had they done an independent analysis and come up with a different
>> and correct result they would have done an important service. As it
>> is they have been more interesterd in muddying the water than with
>> improving the science, and any benefits are purely incidental.
>
> M&M did an independent analysis of the MBH paper
> They came up with results which differ from MBH.
The analysis M&M did was a lot more flawed than the one by MBH so that
isn't much of an achievement, nor was it independent, they just took the
description by MBH as they understood it based on no knowledge in the
field and plugged in the numbers. To make it truly independent they would
have to choose their own datasets and how to analyze them. This has been
done, but not by M&M.
> They also concluded that the dataset description and methodology of
> mbh were inadequate.
Most of their complaints were just silly, because they didn't know the
field and were too lazy to read the references supplied by MBH.
> This latter has been shown to be true with MBH's corrigendum.
> I think you don't like the message.
What message? That MBH are humans and make mistakes, yes, I can accept
that. That their result is wrong as M&M claims? No that hasn't been
shown.
If you want a howler of a mistake, have a look at this blog, entries from
mid May about the book "Taken by Storm" by Essex and McKitrick:
http://cgi.cse.unsw.edu.au/~lambert/cgi-bin/blog/?_start=21
For a starter they deny there is a well defined concept of how to take an
average temperature:
"In the absence of physical guidance, any rule for averaging temperature
is as good as any other. The folks who do the averaging happen to use the
arithmetic mean over the field with specific sets of weights, rather
than, say, the geometric mean or any other. But this is mere convention."
Really? Shouldn't you learn high school science before writing a book
about science?
Bill Corden
being discussed here, but I would like to point out some general criticisms
of credentialed science as compared to a simple search for truth.
I am an expert in some fields of accounting and programming. When I am
talking to an intelligent person who asks intelligent questions of my area
of expertise, I don't tell them that their questions are not worth answering
because they are not qualified to understand the answer. I get very
suspicious when people who are part of the "in" group ridicule thoughtful
people just because they in the "out" group. Fools are everywhere and
credentials don't protect from being a fool.
So the question is, why are people who seek to discuss these issues attacked
for who they are, rather than the weight of their argument? Should not the
truth of an argument be the important thing? Fine, some of you are
scientists. Is it really asking too much for you to show a little patience
in explaining why what appears to be true is not true in your view? Must
you treat non-scientist like inconsequential children? I will tell you that
I believe there are a number (I am still trying to decide how many) of
scientist that get lost in the trees of details and miss the big picture.
And for goodness sake, environmental science has been major league wrong
before. When I see personal attacks in response to rational questions, I
have to wonder if truth is taking a second seat to posturing.
For the sake of the laymen lurking, please try to patiently explain your
reasoning. When you refer to common knowledge or consensus, post a link or
two that illustrates the point.
In case you have not noticed, politics is a pretty major component in
environmental science today. I am quite sure that the rules of politics
require that people be treated with respect if you are going to be
persuasive.
George Burt
"Per" <perox...@ntlworld.com> wrote in message
news:bd7c1148.04070...@posting.google.com...
Per
> Note as well that the Mann,et al 1998 paper was not used in the IPCC results
> which were *derived* from a later paper published in 1999 with most of the
> early stumbles fixed.
Actually, it seems that there is an important point here, and I am
glad it has been brought to my attention. The IPCC relies heavily on
MBH '99, and the promotional graphics for IPCC are in some cases
allegedly derived directly from MBH '99.
Yet the point of MBH '99 is that it does not change the MBH'98 data/
methodology for post-1400; MBH'99 extends the record earlier to cover
the period 1000-1399.
So it seems that Ian St. John's point above is directly wrong; that
the IPCC results 1400-2000 are derived in large part from MBH '98. If
MBH '98 goes down, it would appear that the credibility of IPCC may be
at stake.
is this analysis wrong ?
per
Per
> On 6 Jul 2004 03:12:22 -0700, perox...@ntlworld.com (Per) wrote:
> >
> >you know, until M&M published, the MBH paper stood unimpeached and
> >no-one had any reservations. When M&M published their audit, the
> >specific charges they made included : that the dataset MBH used could
> >not be established from the paper, and that there were a whole variety
> >of data problems, and a whole variety of methods information that
> >needed to be provided.
>
> It still is unimpeached
This is not true. We now KNOW that the MBH'98 paper had errors and
omissions, because MBH themselves have made this statement in their
corrigendum.
> and M&M did no audit of any kind.
Strangely enough, when I have to choose between your baseless
assertion, and the M&M audit paper, which was reviewed by an expert
peer reviewer and an editorial board, I know which viewpoint I would
choose.
> How? Nothing that has happened has invalidated MBH and
> certainly nothing that has happened has validated the efforts of M&M.
This is the start of the text of the corrigendum; I think wmc may also
be interested in this text:
"It has been drawn to our attention (S. McIntyre and R. McKitrick)
that the listing of the 叢roxy' data set in the Supplementary
Information published with this Article contained several errors.
In Table 1 we provide a list of the records that were either
mistakenly
included in the Supplementary Information, or mistakenly left out."
Just in case you missed it, MBH have published this corrigendum
BECAUSE errors were drawn to their attention by M&M; i.e. this
corrigendum is a direct result of M&M. I think that validates M&M
straight off the bat.
yours
per
Per
> >you know, until M&M published, the MBH paper stood unimpeached and
> >no-one had any reservations. When M&M published their audit, the
> >specific charges they made included : that the dataset MBH used could
> >not be established from the paper, and that there were a whole variety
> >of data problems, and a whole variety of methods information that
> >needed to be provided.
>
> It still is unimpeached
not at all. We now know that the original publication had errors and
omissions in its description of the methodology and dataset used. We
know this, because MBH have published this in a corrigendum.
> and M&M did no audit of any kind. That
> would have required them to follow the author's methodology which they
> clearly didn't do.
>
<snip>
> How? Nothing that has happened has invalidated MBH and
> certainly nothing that has happened has validated the efforts of M&M.
Well, whether you like it or not, M&M published in the peer-reviewed
literature an audit of the MBH '98 paper. That means expert
peer-review, and an expert editorial board- which ranks somewhat
higher in my opinion than your baseless assertion that they "did no
audit of any kind".
wmc seems to share your view that MBH's corrigendum had no relation to
M&M, and maybe he even believes that MBH's corrigendum would have
appeared if there was no M&M. In fact, the corrigendum explicitly
refers to McIntyre and McKitrick:
"It has been drawn to our attention (S. McIntyre and R. McKitrick)
that the listing of the 叢roxy' data set in the Supplementary
Information published with this Article contained several errors.
In Table 1 we provide a list of the records that were either
mistakenly
included in the Supplementary Information, or mistakenly left out."
This corrigendum is a direct result of the M&M paper. I think that
validates the M&M paper per se.
yours
per
David Ball
wrote:
>It is true that I am largely ignorant about the particulars of the science
>being discussed here, but I would like to point out some general criticisms
>of credentialed science as compared to a simple search for truth.
>
>I am an expert in some fields of accounting and programming. When I am
>talking to an intelligent person who asks intelligent questions of my area
>of expertise, I don't tell them that their questions are not worth answering
>because they are not qualified to understand the answer. I get very
>suspicious when people who are part of the "in" group ridicule thoughtful
>people just because they in the "out" group. Fools are everywhere and
>credentials don't protect from being a fool.
Here's part of the problem, Bill, at least how I see it. There
is a lot of native experience on this forum in a variety of
disciplines. I don't think I've ever seen a single one of those people
take a newby to task for asking a question. Ever. The problem we have
is that many laymen arrive here with a lot of preconceived notions,
notions that don't have any validity, and it is almost impossible to
change their minds.
Look at a couple of recent threads started by someone calling
himself Adrian Vance. This person, obviously literate, knows little or
nothing about the atmospheric sciences, and instead of coming here and
asking a few questions, he proceeded to tell everyone here that he
knew the way of things and everyone else was wrong. That is entirely
the wrong approach to take. It took a lot of time and effort on the
part of a lot of people trying to correct the errors he was posting
and what was the result? He went away.
There's vast quantities of useful information made available
here, but unfortunately, it is covered in layers of mis-information,
dubious posts, op-ed pieces, and outright stupidity. That makes it
difficult for anyone to get at the real information.
>
>So the question is, why are people who seek to discuss these issues attacked
>for who they are, rather than the weight of their argument? Should not the
>truth of an argument be the important thing? Fine, some of you are
>scientists. Is it really asking too much for you to show a little patience
>in explaining why what appears to be true is not true in your view?
Patience is shown, but patience only lasts for so long.
>Must
>you treat non-scientist like inconsequential children?
Only if they insist on behaving like children.
>I will tell you that
>I believe there are a number (I am still trying to decide how many) of
>scientist that get lost in the trees of details and miss the big picture.
I doubt it. This is a science forum. The science deals with
the details and how those details fit into the big picture.
>And for goodness sake, environmental science has been major league wrong
>before. When I see personal attacks in response to rational questions, I
>have to wonder if truth is taking a second seat to posturing.
Give me an example of this happening.
>
>For the sake of the laymen lurking, please try to patiently explain your
>reasoning. When you refer to common knowledge or consensus, post a link or
>two that illustrates the point.
It's done all the time.
>
>In case you have not noticed, politics is a pretty major component in
>environmental science today. I am quite sure that the rules of politics
>require that people be treated with respect if you are going to be
>persuasive.
>
Incorrect. There are policy aspects to the science and that is
where the politics comes in. The science itself (i.e. understanding
the physical systems and how they work) has nothing to do with
politics. That comes later.
Per
> Per <perox...@ntlworld.com> wrote:
> >w...@bas.ac.uk wrote in message news:<40e8...@news.nwl.ac.uk>...
> >> All of this appears to be a twisting of the facts to make M&M appear
> >> more powerful than they are...
>
> >you know, I don't have the data to be able to evaluate that claim.
> >Perhaps you have access to the correspondence between Nature, and
> >either M&M or MBH ?
>
> No, hence "appears". M&M are trying to say that Nature "ordered" MBH to
> do something. There is nothing to support that claim.
>
You know, Nature says it won't print trivial corrections; it will only
print corrections which are thoroughly serious (e.g. scientific
integrity of paper, etc). So this correction can only have occurred
two ways.
1) MBH approached Nature, and asked to make the correction. Nature
would then have to form a view yes or no.
2) Nature formed a view that the correction was required, and then
told MBH.
Procedure 2 seems quite possible to me, especially if M&M approached
Nature to catalyse their review of the issue in the first place.
The only public information I have seen is mcKitrick's text that
Nature ordered the corrigendum, and this appears to be a statement of
fact. When you say that the McKitrick text "appears to be a twisting
of the facts", you have no facts to contradict this, and yet; you are
contradicting McKitrick's statement of fact.
hmmm
per
David Ball
>David Ball <wra...@mb.sympatico.ca> wrote in message news:<surle0d4m7gvmnd02...@4ax.com>...
>> On 6 Jul 2004 03:12:22 -0700, perox...@ntlworld.com (Per) wrote:
>> >
>> >you know, until M&M published, the MBH paper stood unimpeached and
>> >no-one had any reservations. When M&M published their audit, the
>> >specific charges they made included : that the dataset MBH used could
>> >not be established from the paper, and that there were a whole variety
>> >of data problems, and a whole variety of methods information that
>> >needed to be provided.
>>
>> It still is unimpeached
>
>This is not true. We now KNOW that the MBH'98 paper had errors and
>omissions, because MBH themselves have made this statement in their
>corrigendum.
No. We know that they did not describe their methodology as
fully as some might wish. That is a far cry from invalidating their
results.
>
>> and M&M did no audit of any kind.
>
>Strangely enough, when I have to choose between your baseless
>assertion, and the M&M audit paper, which was reviewed by an expert
>peer reviewer and an editorial board, I know which viewpoint I would
>choose.
LOL. Per, get a grip. An audit would have required them to
precisely follow the author's methodology. They didn't. They made it
up as they went along. The M&M paper was no more reviewed by an expert
than I can fly to the moon. The very first thing any half-decent
reviewer should have noticed was the warming that occurred in the
middle of the bloody Little Ice Age. And E&E's editorial board? Don't
make me laugh. It's pretty hard to take you seriously when you posture
like this.
>
>> How? Nothing that has happened has invalidated MBH and
>> certainly nothing that has happened has validated the efforts of M&M.
>
>This is the start of the text of the corrigendum; I think wmc may also
>be interested in this text:
>"It has been drawn to our attention (S. McIntyre and R. McKitrick)
>that the listing of the 叢roxy' data set in the Supplementary
>Information published with this Article contained several errors.
>In Table 1 we provide a list of the records that were either
>mistakenly
>included in the Supplementary Information, or mistakenly left out."
>
>Just in case you missed it, MBH have published this corrigendum
>BECAUSE errors were drawn to their attention by M&M; i.e. this
>corrigendum is a direct result of M&M. I think that validates M&M
>straight off the bat.
>
Are you daft? All they said was that some omissions were
identified in the Supplementary information. That doesn't mean that
the information wasn't used in their study. It doesn't invalidate
their methodology. It doesn't affect their results. What's more, it
does absolutely nothing to show that M&M did their work properly and
it in no way validates M&M's results.
David Ball
>David Ball <wra...@mb.sympatico.ca> wrote in message news:<surle0d4m7gvmnd02...@4ax.com>...
>> On 6 Jul 2004 03:12:22 -0700, perox...@ntlworld.com (Per) wrote:
>> >you know, until M&M published, the MBH paper stood unimpeached and
>> >no-one had any reservations. When M&M published their audit, the
>> >specific charges they made included : that the dataset MBH used could
>> >not be established from the paper, and that there were a whole variety
>> >of data problems, and a whole variety of methods information that
>> >needed to be provided.
>>
>> It still is unimpeached
>
>not at all. We now know that the original publication had errors and
>omissions in its description of the methodology and dataset used. We
>know this, because MBH have published this in a corrigendum.
Incorrect. We now that the supplementary information was not
complete. That is not an error, but an omission.
>
>> and M&M did no audit of any kind. That
>> would have required them to follow the author's methodology which they
>> clearly didn't do.
>>
><snip>
>> How? Nothing that has happened has invalidated MBH and
>> certainly nothing that has happened has validated the efforts of M&M.
>
>Well, whether you like it or not, M&M published in the peer-reviewed
>literature an audit of the MBH '98 paper. That means expert
>peer-review, and an expert editorial board- which ranks somewhat
>higher in my opinion than your baseless assertion that they "did no
>audit of any kind".
Bullshit. They claim to have published an 'audit'. Further
inspection shows that clearly not to be the case. The peer-review was
shoddy. There's a reason they published in E&E. As for experts, a rank
amateur should have spotted the problematic nature of M&M's results.
>
>wmc seems to share your view that MBH's corrigendum had no relation to
>M&M, and maybe he even believes that MBH's corrigendum would have
>appeared if there was no M&M. In fact, the corrigendum explicitly
>refers to McIntyre and McKitrick:
>"It has been drawn to our attention (S. McIntyre and R. McKitrick)
>that the listing of the 叢roxy' data set in the Supplementary
>Information published with this Article contained several errors.
>In Table 1 we provide a list of the records that were either
>mistakenly
>included in the Supplementary Information, or mistakenly left out."
>
>This corrigendum is a direct result of the M&M paper. I think that
>validates the M&M paper per se.
>
LOL. I write a paper and have it published in Science. You
write a paper on a similar topic, but the work is questionable at
best. You then proceed to tell me of a typo that appears in my paper.
I thank you for correcting the error. Tell me:
a. How does this invalidate my results
b. validate yours.
Bill Corden
> No, it only proves it wasn't perfect. As I already explained essentially
> no paper is perfect. To what extent people worry about corrections depend
> on whether they see there is an interest for one. Had Nature withdrawn
> the paper it would have showed it was below the minimum standard.
Is it at least possible that M&M had some small part in correcting the
imperfections? Would acknowleding that contribution kill you? I understand
that you must think M&M is some kind of danger, but it hurts your
credibility when you don't grant them what they objectively deserve.
> The analysis M&M did was a lot more flawed than the one by MBH so that
> isn't much of an achievement, nor was it independent, they just took the
> description by MBH as they understood it based on no knowledge in the
> field and plugged in the numbers. To make it truly independent they would
> have to choose their own datasets and how to analyze them. This has been
> done, but not by M&M.
So, M&M are completely irrelevant? Doesn't ring true. In fairness to you,
you did characterize it as an "achievement", but is it necessarily as small
as you say? What does this make you so angry? (this is a scencere
question).
> > They also concluded that the dataset description and methodology of
> > mbh were inadequate.
>
> Most of their complaints were just silly, because they didn't know the
> field and were too lazy to read the references supplied by MBH.
Ok, you might be right, but how should I know? You don't support the
contention with a single specific. Show how it is silly.
> What message? That MBH are humans and make mistakes, yes, I can accept
> that. That their result is wrong as M&M claims? No that hasn't been
> shown.
Do we not inherit an extra burden when we make mistakes? Is it
unreasonable, given that mistakes were made, that these scientist release a
complete record of data and methods? Or is it your contention that the data
has been release. M&M have asked very specific questions of the authors.
Either they will provide the data to answer the questions or they won't.
From my perspective, they screwed up and must now come clean and answer
their critics. And if they are unwilling, I will not believe the results
because they are acting as if they have something to hide.
> If you want a howler of a mistake, have a look at this blog, entries from
> mid May about the book "Taken by Storm" by Essex and McKitrick:
> http://cgi.cse.unsw.edu.au/~lambert/cgi-bin/blog/?_start=21
> For a starter they deny there is a well defined concept of how to take an
> average temperature:
> "In the absence of physical guidance, any rule for averaging temperature
> is as good as any other. The folks who do the averaging happen to use the
> arithmetic mean over the field with specific sets of weights, rather
> than, say, the geometric mean or any other. But this is mere convention."
>
> Really? Shouldn't you learn high school science before writing a book
> about science?
Thank you for posting this link. It does appear that that McKitrick made a
mistake and the blog detailed that mistake. Your last statement seems
overly harsh to me considering what was said in the blog. Maybe the method
of averaging is valid and maybe it is not. I don't doubt that it is the
best that current science can do. But, that is a long way from metaphysical
certainty. You have not the formula for the devine and it is possible that
the methods used are not reliable.
McKitrick replied: "Thanks for pointing this out. It implies there are now 4
averages to choose from, depending on the formula used and how missing data
are treated, and there are no laws of nature to guide the choice. The
underlying point is that there are an infinite number of averages to choose
from, quite apart from the practical problem of missing data."
I think he is spinning when he says there are now 4 averages that can be
used. He should admit his mistake, mea culpa. But, seeing how you extend
this minor mistake into disqualification, I can understand his being a
little defensive. Go back and read the exact quote. There are an infinite
number of averages to chose from. Read the weather guy "Dano" and you
realize how simplistic these averages are compared to what is _really_ going
on. It might be the best we have, but it is hard to argue an
"understanding" from this data.
Finally, there are a lot of references to "right-wingers" and disparaging
comments about the Republicans and the current administration. Is this
politcal, or is this science? I know how to deal with politics. Science is
supposed to be about truth regardless of politics.
George Burt
David Ball
wrote:
>"Thomas Palm" <Thoma...@chello.removethis.se> wrote in message
>news:Xns951EEC1676A1T...@212.83.64.229...
>> No, it only proves it wasn't perfect. As I already explained essentially
>> no paper is perfect. To what extent people worry about corrections depend
>> on whether they see there is an interest for one. Had Nature withdrawn
>> the paper it would have showed it was below the minimum standard.
>
>Is it at least possible that M&M had some small part in correcting the
>imperfections? Would acknowleding that contribution kill you? I understand
>that you must think M&M is some kind of danger, but it hurts your
>credibility when you don't grant them what they objectively deserve.
Sir, are you familiar with the paper in question? It's
history? Until you are, perhaps you should refrain from taking sides.
>
>> The analysis M&M did was a lot more flawed than the one by MBH so that
>> isn't much of an achievement, nor was it independent, they just took the
>> description by MBH as they understood it based on no knowledge in the
>> field and plugged in the numbers. To make it truly independent they would
>> have to choose their own datasets and how to analyze them. This has been
>> done, but not by M&M.
>
>So, M&M are completely irrelevant?
Yes. The author's were extremely sloppy in their methodology
and analysis and their results are highly questionable. They show, for
example, strong warming having taken place during The Little Ice Age,
a period of known cooling. Had they followed MBH's methodology
properly they might have been able to make some statements, but their
paper is a mess.
>Doesn't ring true. In fairness to you,
>you did characterize it as an "achievement", but is it necessarily as small
>as you say? What does this make you so angry? (this is a scencere
>question).
>
>> > They also concluded that the dataset description and methodology of
>> > mbh were inadequate.
>>
>> Most of their complaints were just silly, because they didn't know the
>> field and were too lazy to read the references supplied by MBH.
>
>Ok, you might be right, but how should I know? You don't support the
>contention with a single specific. Show how it is silly.
Go back through the archives and look for "Auditing the
Auditors". Josh Halpern and others clearly documented M&M's inability
to even read the references provided by MBH.
Ian St. John
> "Ian St. John" <ist...@noemail.ca> wrote in message
> news:<eAvGc.18556$WM5.8...@news20.bellglobal.com>...
>> Such as the MBH'98 paper which is both robuts and reproducable.
>
> will you please cite the reference which has independently reproduced
> MBH '98 ?
It was reproduced by a number of other researchers using similar but not
identical methods and data selection. However, you have decided to remain
'permanently ignorant' so your failure to comprehend is excused.
> Perhaps you are all abuse, and no citation ?
I'm getting there are you illustrate that you are all hot air and no
substance.
> per
Ian St. John
> "Ian St. John" <ist...@noemail.ca> wrote in message
> news:<GKSFc.73$JG5....@news20.bellglobal.com>...
>> Note as well that the Mann,et al 1998 paper was not used in the IPCC
>> results which were *derived* from a later paper published in 1999
>> with most of the early stumbles fixed.
>
> Actually, it seems that there is an important point here, and I am
> glad it has been brought to my attention. The IPCC relies heavily on
> MBH '99, and the promotional graphics for IPCC are in some cases
> allegedly derived directly from MBH '99.
No. The IPCC relies on all of the research on the past 1,000 years as an
illustration of how 'out of character' recent warming is. They just selected
to use a variation of Mann, et al 1999 because it included all of the NH and
was by far the most competent analysis. Most researchers in the field find
Mann to be the 'top dog' in dendrochronology reconstructions and the
methodology. It was, in fact, way too much for amateurs like M&M to follow,
much less critique.
>
> Yet the point of MBH '99 is that it does not change the MBH'98 data/
> methodology for post-1400; MBH'99 extends the record earlier to cover
> the period 1000-1399.
And clearly had the methodology checked by the IPCC when they recreated the
results and then modified the procedures to more suit their own needs.
>
> So it seems that Ian St. John's point above is directly wrong; that
> the IPCC results 1400-2000 are derived in large part from MBH '98.
That is wrong. I said it was derived from MBH '99, a different paper. Are
you really that clueless or just playing dumb?
> If
> MBH '98 goes down, it would appear that the credibility of IPCC may be
> at stake.
Not likely. But this may be the reason for the bullshit and bafflegab over
MBH '98
>
> is this analysis wrong ?
Yes. The IPCC is not dependent on an illustration of the comparitive global
temperature stability from Mann, or even from dendrochronology. It has beeen
established by a number of other proxy methods as well. Nor would even
elimination of this 'visible comparison' affect the major findings.
> per
Ian St. John
> "Thomas Palm" <Thoma...@chello.removethis.se> wrote in message
> news:Xns951EEC1676A1T...@212.83.64.229...
>> No, it only proves it wasn't perfect. As I already explained
>> essentially no paper is perfect. To what extent people worry about
>> corrections depend on whether they see there is an interest for one.
>> Had Nature withdrawn the paper it would have showed it was below the
>> minimum standard.
>
> Is it at least possible that M&M had some small part in correcting the
> imperfections? Would acknowleding that contribution kill you? I
> understand that you must think M&M is some kind of danger, but it
> hurts your credibility when you don't grant them what they
> objectively deserve.
They failed to include some of the data from MBH. They failed even to
question the discrepancies between the input data in their attempted
analysis and the documentation from MBH. They went ahead anyway while not
understanding the basics of the methodology. They got a totally wrong result
which divered in the area from which they excluded data. What is surprising?
As to what they objectively deserve, don't get me started...
Thomas Palm
> literature an audit of the MBH '98 paper. That means expert
> peer-review, and an expert editorial board- which ranks somewhat
> higher in my opinion than your baseless assertion that they "did no
> audit of any kind".
All "peer reviewed" litererature is not created equal, and when a journal
isn't even listed in Science Citation Index it is obviously not among the
more respected. E&E is not your average scientific journal, it has an
outright policy of publishing anything critical to the warming theory and
an editor who is quite open that she will publish basically anything in
that area and then let the readers decide if it makes any sense or not. In
case you have access to the journal read "The Greenhouse Effect as a
Function of Atmospheric Mass" by Hans Jelbring, v14 number 2&3, 2003. Come
back once you stopped laughing.
> This corrigendum is a direct result of the M&M paper. I think that
> validates the M&M paper per se.
The corrigendum is probably a result of M&M sending a message to Nature. It
says nothing at all about whether or not M&M were right in their "audit",
they weren't, only that MBH weren't perfect in describing the method and
data they used.
Thomas Palm
news:s2IGc.105$F16...@fe39.usenetserver.com:
> "Thomas Palm" <Thoma...@chello.removethis.se> wrote in message
> news:Xns951EEC1676A1T...@212.83.64.229...
>> No, it only proves it wasn't perfect. As I already explained
>> essentially no paper is perfect. To what extent people worry about
>> corrections depend on whether they see there is an interest for one.
>> Had Nature withdrawn the paper it would have showed it was below the
>> minimum standard.
>
> Is it at least possible that M&M had some small part in correcting the
> imperfections? Would acknowleding that contribution kill you?
Yes, M&M caused this correction, but was it an important correction? Does
it really matter that much at this point that MBH was a bit sloppy in
listing the data series they had used? Had the result been wrong it might
have mattered, but this does not seem to be the case.
> I
> understand that you must think M&M is some kind of danger, but it
> hurts your credibility when you don't grant them what they objectively
> deserve.
They are a danger only because of the propaganda value being spun into
their erroneous result. When it comes to the contrarians it doesn't
really matter if the result is right as long as you can convince a
sufficient number of people who know very little about the issue that it
is, or even that the matter isn't settled. Scientifically M&M are totally
insignificant since they don't understand the science.
>> > They also concluded that the dataset description and methodology of
>> > mbh were inadequate.
>>
>> Most of their complaints were just silly, because they didn't know
>> the field and were too lazy to read the references supplied by MBH.
>
> Ok, you might be right, but how should I know? You don't support the
> contention with a single specific. Show how it is silly.
As David responded, the article by M&M has been discussed a lot in this
forum, for example in the original "Auditing the Auditors" thread. Some
of the posters here even at least partially reproduced the MBH result and
in in the process discovered a number of errors by MBH.
> From my perspective, they
> screwed up and must now come clean and answer their critics. And if
> they are unwilling, I will not believe the results because they are
> acting as if they have something to hide.
Well, you can't please everyone, I guess. As it happens many scientists
are too busy trying to do new research to answer all questions asked by
non-scientists, especially when said non-scientists seems to have a
strong political agenda and doesn't even try to understand the methology
before complaining that it has to be wrong. (Several of the mistakes done
by M&M had nothing to do with omissions by MBH but simply came from their
own inability to comprehend the methods used. Beeing an accountant
doesn't automatically make you an expert in climate science.)
>> If you want a howler of a mistake, have a look at this blog, entries
>> from mid May about the book "Taken by Storm" by Essex and McKitrick:
>> http://cgi.cse.unsw.edu.au/~lambert/cgi-bin/blog/?_start=21
>> For a starter they deny there is a well defined concept of how to
>> take an average temperature:
>> "In the absence of physical guidance, any rule for averaging
>> temperature is as good as any other. The folks who do the averaging
>> happen to use the arithmetic mean over the field with specific sets
>> of weights, rather than, say, the geometric mean or any other. But
>> this is mere convention."
>>
>> Really? Shouldn't you learn high school science before writing a book
>> about science?
>
> Thank you for posting this link. It does appear that that McKitrick
> made a mistake and the blog detailed that mistake. Your last
> statement seems overly harsh to me considering what was said in the
> blog. Maybe the method of averaging is valid and maybe it is not. I
> don't doubt that it is the best that current science can do. But,
> that is a long way from metaphysical certainty. You have not the
> formula for the devine and it is possible that the methods used are
> not reliable.
There are technical issues on how to calculate the average temperature in
the best way given the limited data, but there is no uncertainty
whatsoever as to what "average temperature" means. Taking the geometric
mean to calculate it is simply wrong, not just "another convention", and
a stupid mistake at that. Have you even read any thermodynamics?
> McKitrick replied: "Thanks for pointing this out. It implies there are
> now 4 averages to choose from, depending on the formula used and how
> missing data are treated, and there are no laws of nature to guide the
> choice. The underlying point is that there are an infinite number of
> averages to choose from, quite apart from the practical problem of
> missing data."
>
> I think he is spinning when he says there are now 4 averages that can
> be used. He should admit his mistake, mea culpa. But, seeing how you
> extend this minor mistake into disqualification, I can understand his
> being a little defensive.
It is not a minor mistake, especially when he repeats and even enhances
it with his response. This mistakes shows a fundamental ignorance about
physics, and even data handling. If he doesn't know the physics, how do
you think he manages the much more specialized climate science?
> Finally, there are a lot of references to "right-wingers" and
> disparaging comments about the Republicans and the current
> administration. Is this politcal, or is this science? I know how to
> deal with politics. Science is supposed to be about truth regardless
> of politics.
For M&M it is definitely political. That's the only reason they started
to investigate an article so far outside their own competence, and it is
the only reason the erroneous results of a couple of amateurs have gotten
the publicity it has. Had they tried to do the same to something a bit
less politcally sensitive, like some particle experiment over at CERN
they would simple have been ignored or told to come back once they had
learned the basics. There are no special journals to publish any article
regardless of quality as long as it contradicts what is published from
CERN.
Per
> perox...@ntlworld.com (Per) wrote in
> news:bd7c1148.04070...@posting.google.com:
>
> > Thomas Palm <Thoma...@chello.removethis.se> wrote in message
> > news:<Xns951E86CDBEFAAT...@212.83.64.229>...
> >> What do you think is a "minimum standard"?
> > the issue is that MBH accept with their corrigendum that what they
> > previously published was not at that minimum standard.
>
> No, it only proves it wasn't perfect. As I already explained essentially
> no paper is perfect. To what extent people worry about corrections depend
> on whether they see there is an interest for one. Had Nature withdrawn
> the paper it would have showed it was below the minimum standard.
Nature categorically states that it will not publish trivial
corrections, only corrections that are (and I paraphrase) substantial,
going to the scientific integrity of the paper, or the like.
You seem very confused. Retractions are when the whole paper has been
busted. Corrigenda are corrections. I have not suggested that the
whole paper has been busted. MBH have admitted that they made errors
and omissions in MBH'98, and these were not trivial errors. The
corrections bring the paper back to the minimum standard required for
publication.
> > M&M did an independent analysis of the MBH paper
> > They came up with results which differ from MBH.
>
> The analysis M&M did was a lot more flawed than the one by MBH so that
> isn't much of an achievement, nor was it independent, they just took the
> description by MBH as they understood it based on no knowledge in the
> field and plugged in the numbers. To make it truly independent they would
> have to choose their own datasets and how to analyze them. This has been
> done, but not by M&M.
This is bizarre. M&M set out to audit MBH'98 by attempting to
replicate the study; this was their (limited) goal. That would be an
independent replication (or analysis) of MBH'98- which is what I said.
If you want to claim M&M is a lot more flawed, perhaps it would help
if you could specify, rather than just throwing brickbats ?
per
Per
> On 6 Jul 2004 16:52:04 -0700, perox...@ntlworld.com (Per) wrote:
> >
> >This is not true. We now KNOW that the MBH'98 paper had errors and
> >omissions, because MBH themselves have made this statement in their
> >corrigendum.
>
> No. We know that they did not describe their methodology as
> fully as some might wish. That is a far cry from invalidating their
> results.
These are the words of MBH, and they directly contradict you:
"that the listing of the 'proxy' data set in the Supplementary
Information published with this Article contained several errors."
I have not charged that their corrigendum invalidates their results !
Though I can understand why you might think so :-)
> LOL. Per, get a grip. An audit would have required them to
> precisely follow the author's methodology. They didn't. They made it
> up as they went along. The M&M paper was no more reviewed by an expert
> than I can fly to the moon. The very first thing any half-decent
> reviewer should have noticed was the warming that occurred in the
> middle of the bloody Little Ice Age. And E&E's editorial board? Don't
> make me laugh. It's pretty hard to take you seriously when you posture
> like this.
David
you are reckless with the truth. You have no idea who peer-reviewed
M&M, yet you make statements such as the above; that it effectively
did not receive expert peer-review. That is a baseless assertion.
And yes; M&M set out to follow MBH's methodology, which is an audit.
> Are you daft? All they said was that some omissions were
> identified in the Supplementary information.
Dear David
I understand you find it difficult to understand the relevance of a
dataset to final results. Nonetheless, most scientists find it quite
helpful to have the correct dataset to work with. The supplementary
information described the dataset of MBH'98. If you think that is
trivial, you differ from the editors of nature, and MBH.
yours
per
Per
> On 6 Jul 2004 17:01:08 -0700, perox...@ntlworld.com (Per) wrote:
>
> >not at all. We now know that the original publication had errors and
> >omissions in its description of the methodology and dataset used. We
> >know this, because MBH have published this in a corrigendum.
>
> Incorrect. We now that the supplementary information was not
> complete. That is not an error, but an omission.
So let me get this right; an omission is not an error ?
>
> Bullshit. They claim to have published an 'audit'. Further
> inspection shows that clearly not to be the case. The peer-review was
> shoddy. There's a reason they published in E&E. As for experts, a rank
> amateur should have spotted the problematic nature of M&M's results.
You have no knowledge of what happened during M&M's peer review; your
claims to the contrary reveal you as a fantasist and liar.
> > In fact, the corrigendum explicitly
> >refers to McIntyre and McKitrick:
> >"It has been drawn to our attention (S. McIntyre and R. McKitrick)
> >that the listing of the ?proxy' data set in the Supplementary
> >Information published with this Article contained several errors."
> >
> LOL. I write a paper and have it published in Science.
Dear David
READ the text of the corrigendum. I have stripped it down so you only
have to parse one sentence for its content. See if you can understand
the text.
good luck.
per
Thomas Palm
> Thomas Palm <Thoma...@chello.removethis.se> wrote in message
> You seem very confused. Retractions are when the whole paper has been
> busted. Corrigenda are corrections. I have not suggested that the
> whole paper has been busted.
If you go back to your first message in this thread you will find that M&M
made the claim that the result was wrong and that you at least appear to
support them.
>> > M&M did an independent analysis of the MBH paper
>> > They came up with results which differ from MBH.
>>
>> The analysis M&M did was a lot more flawed than the one by MBH so
>> that isn't much of an achievement, nor was it independent, they just
>> took the description by MBH as they understood it based on no
>> knowledge in the field and plugged in the numbers. To make it truly
>> independent they would have to choose their own datasets and how to
>> analyze them. This has been done, but not by M&M.
>
> This is bizarre. M&M set out to audit MBH'98 by attempting to
> replicate the study; this was their (limited) goal. That would be an
> independent replication (or analysis) of MBH'98- which is what I said.
And M&M:s "audit" was deeply flawed due to their own ability to comprehend
what MBH had done.
> If you want to claim M&M is a lot more flawed, perhaps it would help
> if you could specify, rather than just throwing brickbats ?
> per
Given that you reused the name of an old thread here in sci.environment I
assumed you had read the earlier discussion where several posters a lot
more competent than I went through the statements by M&M and found them
wanting. You will also find a number of entries over at Quark Soup from
when the discussion was fresh:
http://www.davidappell.com/
w...@bas.ac.uk
>...the M&M audit paper, which was reviewed by an expert
>peer reviewer and an editorial board,...
We don't really know this. E&E's editorial/review policy isn't
clearly stated. Did you really mean a singular reviewer? Most journals
use at least 2.
-W.
--
William M Connolley | w...@bas.ac.uk | http://www.antarctica.ac.uk/met/wmc/
Climate Modeller, British Antarctic Survey | Disclaimer: I speak for myself
I'm a .signature virus! copy me into your .signature file & help me spread!
Bill Corden
Bill Corden
> Sir, are you familiar with the paper in question? It's
> history? Until you are, perhaps you should refrain from taking sides.
I was responding to the tone. I am here with an open mind and your
thoughtful response has helped.
Thanks
Per
> > Yet the point of MBH '99 is that it does not change the MBH'98 data/
> > methodology for post-1400; MBH'99 extends the record earlier to cover
> > the period 1000-1399.
> >
> > the IPCC results 1400-2000 are derived in large part from MBH '98.
>
> That is wrong. I said it was derived from MBH '99, a different paper. Are
> you really that clueless or just playing dumb?
Maybe you can't read. I charged that MBH adds no new data for the
period 1400-2000; that MBH'99 simply extended the data from MBH'98 to
cover the period 1000-1399. So the Figures in MBH'99 that cover the
timespan 1400-2000 are a direct lift of the results from MBH'98.
I am directly contradicting you as a matter of fact.
yours
per
Bill Corden
> their erroneous result. When it comes to the contrarians it doesn't
> really matter if the result is right as long as you can convince a
> sufficient number of people who know very little about the issue that it
> is, or even that the matter isn't settled. Scientifically M&M are totally
> insignificant since they don't understand the science.
I understand this and frankly, it is what brought me here. I come from the
political side and I fight the fights there. But truth matters more than
politics and politics should be informed truth. Often it is not.
> As David responded, the article by M&M has been discussed a lot in this
> forum, for example in the original "Auditing the Auditors" thread. Some
> of the posters here even at least partially reproduced the MBH result and
> in in the process discovered a number of errors by MBH.
I will read the background. Thanks for pointing it out.
> Well, you can't please everyone, I guess. As it happens many scientists
> are too busy trying to do new research to answer all questions asked by
> non-scientists, especially when said non-scientists seems to have a
> strong political agenda and doesn't even try to understand the methology
> before complaining that it has to be wrong. (Several of the mistakes done
> by M&M had nothing to do with omissions by MBH but simply came from their
> own inability to comprehend the methods used. Beeing an accountant
> doesn't automatically make you an expert in climate science.)
Since integrity is central this current situation. Would now say that it is
important enough for these busy scientist to gather together their data and
lay it out? M&M ask specific questions. I don't M&M from Adam, but this
has drawn people's attention. It seems to me that either their numbers and
methods work or they don't. Having the methods and the data and the sources
of the data seems reasonable. This is not the formula for coke. Notice
that we are not talking about the validity of the methods here, but the
numbers should work, the methods should be documented and the source of the
numbers should be rational. What am I missing here?
Thanks for response.
> There are technical issues on how to calculate the average temperature in
> the best way given the limited data, but there is no uncertainty
> whatsoever as to what "average temperature" means. Taking the geometric
> mean to calculate it is simply wrong, not just "another convention", and
> a stupid mistake at that. Have you even read any thermodynamics?
No, but I do program in dynamic HTML, is that the same thing?
> It is not a minor mistake, especially when he repeats and even enhances
> it with his response. This mistakes shows a fundamental ignorance about
> physics, and even data handling. If he doesn't know the physics, how do
> you think he manages the much more specialized climate science?
Point taken. I wish there was not this "gotcha" aspect to this though.
McKitrick did respond substantively to this guy. Apparently (I don't know
the whole story) Mann and company did not think it important enough to
provide M&M with all the data used. This seems arrogant or perhaps
petulant. Am I missing something here?
> For M&M it is definitely political. That's the only reason they started
> to investigate an article so far outside their own competence, and it is
> the only reason the erroneous results of a couple of amateurs have gotten
> the publicity it has.
I would hope you are as vigorous in your condemnation of people who
exagerate the science of global warming for political purposes. If these
guys are pure hacks, then a full analysis will bear that out and that will
be important. Would you agree that taking MBH's word at face value that the
corrections did not alter the results is unreasonable? I think all sides
should be able to agree that a COMPLETELY transparent situation must occur.
This "too busy" to provide data thing must not happen in this case.
Per
> They failed to include some of the data from MBH.
Not a surprise; MBH have now admitted that they did not list all the
data, and that there are numerous defects in their original
supplementary information.
> They failed even to
> question the discrepancies between the input data in their attempted
> analysis and the documentation from MBH.
As a matter of fact, this is untrue. There is documentation to show
that M&M asked both Mann, and his research assistant, about these
discrepancies.
> They went ahead anyway while not understanding the basics of the methodology.
I don't see how you can speak with any authority on what M&M do or do
not understand. M&M's analysis requires only the math to repeat what
MBH did- and I haven't seen much by way of substantive criticism of
their maths.
> They got a totally wrong result
> which divered in the area from which they excluded data.
wrong means they made an error, something demonstrably bad. I don't
recall seeing that illustrated. They did exclude data, but said they
had reason to; and with the state of the MBH dataset, there may well
have been good reason. That would then be a judgement call, and not a
question of error.
per
Per
> The corrigendum is probably a result of M&M sending a message to Nature. It
> says nothing at all about whether or not M&M were right in their "audit",
> they weren't, only that MBH weren't perfect in describing the method and
> data they used.
A principal finding of M&M, which was bitterly fought in this forum,
was that there were defects in the dataset and description of the
methodology of MBH'98. MBH themselves now accept that there were
errors and omissions in MBH'98. So in this respect, MBH are accepting
that this result of the M&M paper is right.
You are still fighting, it seems, even when MBH have accepted this
point.
fight the good fight, Don Quixote !
per
Per
> Yes, M&M caused this correction, but was it an important correction? Does
> it really matter that much at this point that MBH was a bit sloppy in
> listing the data series they had used? Had the result been wrong it might
> have mattered, but this does not seem to be the case.
Gee whiz, it is back to basics again. Nature says it is an important
correction, because they refuse to publish trivial corrections; read
the editorial policy on corrections.
But do you really believe that anyone can reproduce an analysis when
they are given the wrong dataset ? Do you really think this is a
trivial issue ?
And as far as I know, no-one has yet independently verified that the
results of MBH'98 can be reproduced- and this is a fundamental of all
scientific work. If these results cannot be reproduced, MBH will be
without value. So maybe it is worth waiting to see if anyone else can
reproduce these results.
> They are a danger only because of the propaganda value being spun into
> their erroneous result. When it comes to the contrarians it doesn't
> really matter if the result is right as long as you can convince a
> sufficient number of people who know very little about the issue that it
> is, or even that the matter isn't settled. Scientifically M&M are totally
> insignificant since they don't understand the science.
Actually, you are wrong as a matter of fact. M&M found that the MBH'98
paper had defects in its dataset, and published that finding. We now
know that MBH agree with this finding; and the MBH'98 dataset has been
updated- which is scientifically significant.
You are wrong in principle as well. If M&M show that the results of
MBH'98 cannot be reproduced with the published methodology and
dataset, then they have busted MBH. Lest you forget, all science must
pass the test of reproducibility. This has yet to be demonstrated with
MBH'98.
yours
per
David Ball
>David Ball <wra...@mb.sympatico.ca> wrote in message news:<bkime0hiuih73tg7t...@4ax.com>...
>> On 6 Jul 2004 17:01:08 -0700, perox...@ntlworld.com (Per) wrote:
>>
>> >not at all. We now know that the original publication had errors and
>> >omissions in its description of the methodology and dataset used. We
>> >know this, because MBH have published this in a corrigendum.
>>
>> Incorrect. We now that the supplementary information was not
>> complete. That is not an error, but an omission.
>
>So let me get this right; an omission is not an error ?
>
>>
>> Bullshit. They claim to have published an 'audit'. Further
>> inspection shows that clearly not to be the case. The peer-review was
>> shoddy. There's a reason they published in E&E. As for experts, a rank
>> amateur should have spotted the problematic nature of M&M's results.
>
>You have no knowledge of what happened during M&M's peer review; your
>claims to the contrary reveal you as a fantasist and liar.
LOL. I claim that any peer-reviewer worth his or her salt
should have noticed the discrepency between M&M's results and reality
and should have questioned their methodology. Indeed, that is
something any competent analyst should have been asking. When one does
and analysis and arrives at results that don't make sense, one should
question exactly what one has done. You don't rush a paper to print
with the bloody error intact.
>
>> > In fact, the corrigendum explicitly
>> >refers to McIntyre and McKitrick:
>> >"It has been drawn to our attention (S. McIntyre and R. McKitrick)
>> >that the listing of the ?proxy' data set in the Supplementary
>> >Information published with this Article contained several errors."
>> >
>> LOL. I write a paper and have it published in Science.
>
>Dear David
>READ the text of the corrigendum. I have stripped it down so you only
>have to parse one sentence for its content. See if you can understand
>the text.
>good luck.
Hope you get better soon, Per. It's quite clear you have no
interest in the science or the results of this paper. Your inability
to look at the facts clearly shows.
David Ball
>Thomas Palm <Thoma...@chello.removethis.se> wrote in message news:<Xns951F54B5CFCDCT...@212.83.64.229>...
>
>> The corrigendum is probably a result of M&M sending a message to Nature. It
>> says nothing at all about whether or not M&M were right in their "audit",
>> they weren't, only that MBH weren't perfect in describing the method and
>> data they used.
>
>A principal finding of M&M, which was bitterly fought in this forum,
>was that there were defects in the dataset and description of the
>methodology of MBH'98. MBH themselves now accept that there were
>errors and omissions in MBH'98. So in this respect, MBH are accepting
>that this result of the M&M paper is right.
Completely wrong. A principle finding of M&M was that they
couldn't follow the author's methodology. As I've pointed out to you
repeatedly, there is no requirement that peer-reviewed papers be
written for the masses. They are intended to be read by fellow experts
and M&M are most certainly not that. Having said that, there is always
a danger that seemingly trivial points will get glossed over. That is
omission, not error. I daresay there are experts here would could
conduct discussions on aspects of the science that you wouldn't have a
clue about. That doesn't mean that they are wrong or hiding something,
just that they are experts and don't need to spell out every little
thing for your benefit. To expect that to happen would reduce the
peer-review to the level of a Dick & Jane reader.
>
>You are still fighting, it seems, even when MBH have accepted this
>point.
LOL. They accepted the point that the supplementary
information as presented is incomplete. Nothing more. Talk about Don
Quixote! You're the one imagining all manner of results from M&M that
have no basis in fact.
Nigel Persaud
> Auditors". Josh Halpern and others clearly documented M&M's inability
> to even read the references provided by MBH.
I don't recall Josh proving this at all.
He proved that he couldn't figure out the 159 series said to have been
used by MBH. He ARGUED that MBH were justified in truncating the early
portions of the Central England and Central Europe records (although
he did not provide an adequate reason for the MBH failure to disclose
the truncation). However, the early portion of the Central England
record is used in Jones et al (1998) (and in Jones and Mann (2004))
and Josh never responded to this. On balance, I think that Josh
failed to show that M&M were incorrect in pointing out the undisclosed
truncation. In fact, the Corrigendum goes against Josh's argument. It
says that the start year of the Central Europe series, as used in
MBH98, was 1525, not 1550 as stated in MBH98. This would mean that the
earliest part of the Central Europe series was actually used
(notwithstanding Josh's learned arguments on this). Actually, it looks
like the corrigendum looks to be wrong and to have just muddied the
waters on this point - it was the original source that started in
1525, not MBH98; the corrigendum got it backwards.
The earlier discussion also followed up on M&M's report of wild
geographical errors in MBH. Some more geographical errors not even
identified in M&M were also noticed. In the Corrigendum, MBH skate
around this - they say that the source for their instrumental series
was not Bradley and Jones 1992, as stated in MBH98, but NOAA. They do
not acknowledge any geographical errors, but do not identify the
series that they used from NOAA. NOAA has hundreds of thousands of
series. I'm surprised that Nature allowed such an inadequate
reference.
Nigel.
David Ball
>
>> They are a danger only because of the propaganda value being spun into
>> their erroneous result. When it comes to the contrarians it doesn't
>> really matter if the result is right as long as you can convince a
>> sufficient number of people who know very little about the issue that it
>> is, or even that the matter isn't settled. Scientifically M&M are totally
>> insignificant since they don't understand the science.
>
>Actually, you are wrong as a matter of fact. M&M found that the MBH'98
>paper had defects in its dataset, and published that finding. We now
>know that MBH agree with this finding; and the MBH'98 dataset has been
>updated- which is scientifically significant.
Bull. They found that there were omissions from the
supplementary data which apparently have been corrected. You are so
desperate to prove M&M right that you've lost all semblence of
objectivity.
>
>You are wrong in principle as well. If M&M show that the results of
>MBH'98 cannot be reproduced with the published methodology and
>dataset, then they have busted MBH. Lest you forget, all science must
>pass the test of reproducibility. This has yet to be demonstrated with
>MBH'98.
>
No. They can make no statement about the MBH methodology
because they didn't follow it. I'm not sure how you got from omissions
in the supplementary data to MBH's methodology. The methodology (i.e.
HOW they analyzed the data) is different than the data. One can use
different datasets and arrive at different results and still say
nothing about the methodology. For example, I can take two datasets
and calculate the mean value of both. I get two different results, but
at the end I'm still calculating the mean. The way the mean is
calculated doesn't change because I use a different input dataset.
David Ball
wrote:
>> Go back through the archives and look for "Auditing the
>> Auditors". Josh Halpern and others clearly documented M&M's inability
>> to even read the references provided by MBH.
>
>I don't recall Josh proving this at all.
He proved quite effectively that M&M couldn't be bothered to
read the references that came with MBH since many of the questions
that appeared on climate2003 website - I notice it's back - were
answered in the references. It speaks volumes about the "expertise"
and rigour shown by M&M. When you don't bother to RTFR, you have a
serious problem.
Thomas Palm
> "Thomas Palm" <Thoma...@chello.removethis.se> wrote in message
> Since integrity is central this current situation. Would now say that
> it is important enough for these busy scientist to gather together
> their data and lay it out?
For political reasons it would be nice if one of the authors of the
original article took the time of explaining the algorithm in enough
detail for anyone to follow and added the complete method to calculate it
rather than the partial data that has been available (although M&M didn't
find much of that as shown in the earlier discussion here). I suspect
this is a bit of work, though, as their original program isn't exaclty
user friendly.
As far as I'm concerned the fact that other researchers has been able to
get essentially the same result using their own selection of datasets and
calculations is a lot more important than if someone can get exactly the
same result as MBH by plugging the same numbers into the same program.
>> There are technical issues on how to calculate the average
>> temperature in the best way given the limited data, but there is no
>> uncertainty whatsoever as to what "average temperature" means. Taking
>> the geometric mean to calculate it is simply wrong, not just "another
>> convention", and a stupid mistake at that. Have you even read any
>> thermodynamics?
>
> No, but I do program in dynamic HTML, is that the same thing?
Not quite :-)
To understand their second stupidity you don't even need to know any
physics, though:
"You see, some stations had missing values, months where no temperature
had been recorded. When calculating the root mean square they treated the
missing values as if they were measurements of 0 degrees. This is
incorrect, since the temperature was not actually zero degrees. Because
the overall average temperature was positive this meant that the root
mean square was biased downwards when there were missing observations.
And since there were more missing values in the second half of the time
series, this produced a spurious cooling trend. "
Setting missing data to 0 C to get an average is crazy, which I hope you
agree on. With enough missing datapoints the temperature is guaranteed to
be close to 0 C no matter what the real temperature is. (I guess we
should be happy they didn't use 0 K instead or we might have gotten
really wierd results).
>> It is not a minor mistake, especially when he repeats and even
>> enhances it with his response. This mistakes shows a fundamental
>> ignorance about physics, and even data handling. If he doesn't know
>> the physics, how do you think he manages the much more specialized
>> climate science?
>
> Point taken. I wish there was not this "gotcha" aspect to this
> though. McKitrick did respond substantively to this guy.
Since M&M from the start used a "gotcha" approach towards MBH you
shouldn't be surprised if they are treated by the same attitude
themselves, and as far as I can tell the only substance in McKitrick's
response was that it reinforced the impression that he hasn't got a clue
what he is talking about. It wasn't just a temporary slip that let those
mistakes through in the book.
> Apparently
> (I don't know the whole story) Mann and company did not think it
> important enough to provide M&M with all the data used. This seems
> arrogant or perhaps petulant. Am I missing something here?
How would you react if some years after you have published a widely
acclaimed article some nobodies without any credentials sent a message
stating that they intended to audit your result? The whole approach just
screams "crackpot" and I'm not susprised that they got a fairly brief
response.
>> For M&M it is definitely political. That's the only reason they
>> started to investigate an article so far outside their own
>> competence, and it is the only reason the erroneous results of a
>> couple of amateurs have gotten the publicity it has.
>
> I would hope you are as vigorous in your condemnation of people who
> exagerate the science of global warming for political purposes.
You can search back for my responses to some of Roger Coppock's claims
that temperature and CO2 levels are rising exponentially. Since I do
believe that global warming is a genuine problem I'm not too hapy about
people discrediting it by overstating it with bogus analysis.
> If
> these guys are pure hacks, then a full analysis will bear that out and
> that will be important.
We already know that M&M are hacks. The discussion after their original
publication of their paper made that perfectly clear. The remaining
issue is whether or not they by pure chance managed to attack a paper
that turns out to be wrong. Their own alternative reconstruction is
discredited.
Ian St. John
> "Ian St. John" <ist...@noemail.ca> wrote in message
> news:<zpKGc.25715$WM5.1...@news20.bellglobal.com>...
>>> Yet the point of MBH '99 is that it does not change the MBH'98 data/
>>> methodology for post-1400; MBH'99 extends the record earlier to
>>> cover the period 1000-1399.
>>>
>>> So it seems that Ian St. John's point above is directly wrong; that
>>> the IPCC results 1400-2000 are derived in large part from MBH '98.
>>
>> That is wrong. I said it was derived from MBH '99, a different
>> paper. Are you really that clueless or just playing dumb?
>
> Maybe you can't read. I charged that MBH adds no new data for the
> period 1400-2000;
So? What has adding new data got to do with improving and extending a paper?
The same data that MBH processed to give good science was used by M&M to
produce total crap. The fact that they used much of the same data is
irrelevant.
> that MBH'99 simply extended the data from MBH'98 to
> cover the period 1000-1399.
Not simply extended. It did extend the data but also incorporated more
understanding of the data they had. This is called 'learning'. You should
give it a try sometime. Permanent ignorance does not change reality. It
just makes you look stupid.
> So the Figures in MBH'99 that cover the
> timespan 1400-2000 are a direct lift of the results from MBH'98.
You have compared their data sets, analysis, and methodology to that extent?
No changes in the data ( hint, check again ) for the period of 1400 to 2000?
And having done so, you then checked the data sets and methodologies for the
IPCC modifications?
No? Nuff said.
>
> I am directly contradicting you as a matter of fact.
Actually it would be better to describe your actions as bringing feces fully
formed from your anus.
Ian St. John
> "Ian St. John" <ist...@noemail.ca> wrote in message
> news:<guKGc.25813$WM5.1...@news20.bellglobal.com>...
>
>> They failed to include some of the data from MBH.
>
> Not a surprise; MBH have now admitted that they did not list all the
> data, and that there are numerous defects in their original
> supplementary information.
Much of the errors in data that M&M complained about were clearly given in
the reference. It was M&M that failed to properly process the data. The
'clarification for the simple minded' is more of a demonstration of what it
takes to make things clear to a couple of non-scientists clowns trying to
read the science, than a criticism of the paper.
>
>> They failed even to
>> question the discrepancies between the input data in their attempted
>> analysis and the documentation from MBH.
>
> As a matter of fact, this is untrue. There is documentation to show
> that M&M asked both Mann, and his research assistant, about these
> discrepancies.
And yet went ahead without a clear understanding of their basic errors.
>
>> They went ahead anyway while not understanding the basics of the
>> methodology.
>
> I don't see how you can speak with any authority on what M&M do or do
> not understand. M&M's analysis requires only the math to repeat what
> MBH did- and I haven't seen much by way of substantive criticism of
> their maths.
see. David Ball.
"> I don't recall Josh proving this at all.
He proved quite effectively that M&M couldn't be bothered to
read the references that came with MBH since many of the questions
that appeared on climate2003 website - I notice it's back - were
answered in the references. It speaks volumes about the "expertise"
and rigour shown by M&M. When you don't bother to RTFR, you have a
serious problem."
>
>> They got a totally wrong result
>> which divered in the area from which they excluded data.
>
> wrong means they made an error, something demonstrably bad.
Well, there are a LOT of ways that you can go wrong and not all of them are
obvious.
> I don't recall seeing that illustrated.
Look at their chart of the early period of the 1400s and compare it to MBHs.
The effect of their errors is obvious even to the half blind and stupid. No
qualified dendrochronology research reproduces these silly errors.
> They did exclude data, but said they
> had reason to; and with the state of the MBH dataset, there may well
> have been good reason. That would then be a judgement call, and not a
> question of error.
So they were just bad judges instead of incredibly incompetent
non-scientists making simple mistakes with a scientific paper by 'whoops,
was that supposed to be there'? I guess that is about as good a condemnation
of their work as it gets.
McKittrick is an economist. McIntyre is a businessman. They are to be
commended on their attempts to understand the science but condemned on their
distortions which arise from their errors and inability to ask for help when
they are way over their head.
Per
> perox...@ntlworld.com (Per) wrote in
> > You seem very confused. Retractions are when the whole paper has been
> > busted. Corrigenda are corrections. I have not suggested that the
> > whole paper has been busted.
>
> If you go back to your first message in this thread you will find that M&M
> made the claim that the result was wrong and that you at least appear to
> support them.
I have not made the suggestion that MBH'98 is busted, nor do I support
M&M's claim that MBH's result is wrong. If I have written anything
which you think does support these claims, I would be grateful if you
could point it out.
> And M&M:s "audit" was deeply flawed due to their own ability to comprehend
> what MBH had done.
>
> > If you want to claim M&M is a lot more flawed, perhaps it would help
> > if you could specify, rather than just throwing brickbats ?
> > per
>
> Given that you reused the name of an old thread here in sci.environment I
> assumed you had read the earlier discussion where several posters a lot
> more competent than I went through the statements by M&M and found them
> wanting. You will also find a number of entries over at Quark Soup from
> when the discussion was fresh:
> http://www.davidappell.com/
So I asked if you could provide specific examples of where M&M were
wrong; so far, you cannot justify your claim, other than to say
someone may have found something "wanting". Instead we get more
brickbats, and ad hominem abuse of M&M which you can't substantiate.
per
Per
> Per <perox...@ntlworld.com> wrote:
> >...the M&M audit paper, which was reviewed by an expert
> >peer reviewer and an editorial board,...
>
> We don't really know this. E&E's editorial/review policy isn't
> clearly stated. Did you really mean a singular reviewer? Most journals
> use at least 2.
>
> -W.
Dear w
I don't know the process of E&E in any detail. I am assuming at least
one referee; but it may be two or more.
I will be grateful if you can shed any light.
yours
per
David Ball
Oh, God, Per, it's been done repeatedly. Over and over again
ad-nauseum. M&M clearly did not RTFR provided by MBH. On their
climate2003 page they claim oversights on the part of MBH that are not
at all. All they had to do was read the references. For God's sake,
they claimed their were these huge errors in the spreadsheet provided
to them and instead of using their brains and thinking perhaps that
the spreadsheet was corrupt they leapt to the unwarranted conclusion
that this was indeed the data used by MBH and given that their results
were incorrect. It never occurred to the two of them that data coming
from a Linux/Unix environment might be shifted on a WinDoze machine.
Their work was sloppy, superficial and showed absolutely nothing,
except perhaps that science shouldn't be done by amateurs.
Thomas Palm
@posting.google.com:
Some things to consider about E&E:
E&E is mainly directed at social scientists. The editor is not, and does
not claim to be, competent to judge scientific issues. She has the policy
to publish almost anything and let the readers sort out what is correct.
Thus the publication in itself should not be taken as proof of the validity
of the contents of an article.
The journal uses several (or at least more than one) reviewer, but the
editor may chose to publish papers against the recomendation of the
reviewers (somewhat odd given her stated lack of competence in the field).
The editor is deliberately biased in her selection of papers to publish
because in her opinion the "sceptics" don't get enough room in other media.
(From a e-mail conversion I had with Dr.Sonja Boehmer-Christiansen
regarding a reply I submitted to an article by Hans Jelbring in E&E. I
later withdrew it since a similar reply had already been published at that
point.)
Bill Corden
> original article took the time of explaining the algorithm in enough
> detail for anyone to follow and added the complete method to calculate it
> rather than the partial data that has been available (although M&M didn't
> find much of that as shown in the earlier discussion here). I suspect
> this is a bit of work, though, as their original program isn't exaclty
> user friendly.
>
> As far as I'm concerned the fact that other researchers has been able to
> get essentially the same result using their own selection of datasets and
> calculations is a lot more important than if someone can get exactly the
> same result as MBH by plugging the same numbers into the same program.
So we should just trust that these results are just fine. Again, I am
ignorant of the sciences involved, but that strikes me as a dubious way
establish science. If no one can actually recreate the exact numbers using
the exact numbers, how can we know they are valid. Has anyone checked the
other "results"? Would it be reasonable for Enron's auditors to say, we use
the exact same accounting techniques used by Worldcom, should that satisfy?
Please understand that I am not impuning these scientist by comparing them
to Enron and Worldcom. But in auditing (in my field, accounting) you assume
the person is stealing, then prove that they could not have. Must we rely
on their MBH's reputation, or can we examine the evidence and methods
directly?
> To understand their second stupidity you don't even need to know any
> physics, though:
> "You see, some stations had missing values, months where no temperature
> had been recorded. When calculating the root mean square they treated the
> missing values as if they were measurements of 0 degrees. This is
> incorrect, since the temperature was not actually zero degrees. Because
> the overall average temperature was positive this meant that the root
> mean square was biased downwards when there were missing observations.
> And since there were more missing values in the second half of the time
> series, this produced a spurious cooling trend. "
I thought this related to an article unrelated to the issue. Yes it goes to
credibility, but has it been shown that M&M did this in the "Audit"?
> Setting missing data to 0 C to get an average is crazy, which I hope you
> agree on. With enough missing datapoints the temperature is guaranteed to
> be close to 0 C no matter what the real temperature is. (I guess we
> should be happy they didn't use 0 K instead or we might have gotten
> really wierd results).
It is well that we are not all disqualified when we make a mistake. In the
case at hand, McKitrick made the data available to his critics.
> Since M&M from the start used a "gotcha" approach towards MBH you
> shouldn't be surprised if they are treated by the same attitude
> themselves, and as far as I can tell the only substance in McKitrick's
> response was that it reinforced the impression that he hasn't got a clue
> what he is talking about. It wasn't just a temporary slip that let those
> mistakes through in the book.
Perhaps, but again, he subjected his data to his critic's scrutiny. SHOULD
not MBH be required to do the same now that errors have been admitted? In
otherwords, would it not be wise that everyone on both sides of this issue
want to look at the methods and data as applied by MBH. Am I correct that
there is no proprietary aspect to this. I would never allow my competitors
to see my source code, but is it not normal in this case to allow others to
see everything?
> > Apparently
> > (I don't know the whole story) Mann and company did not think it
> > important enough to provide M&M with all the data used. This seems
> > arrogant or perhaps petulant. Am I missing something here?
>
> How would you react if some years after you have published a widely
> acclaimed article some nobodies without any credentials sent a message
> stating that they intended to audit your result? The whole approach just
> screams "crackpot" and I'm not susprised that they got a fairly brief
> response.
Fair enough. These guys are busy with current research. But, after the
errors and/or omissions have been admitted, it seems to me that it calls
into question the results. In politics we would call it an "appearance of
impropriety." Many times investigations are begun merely because of
appearances, not substance, but in order to maintain public confidence, you
let an independant look into it. Now, this is not a legal matter and ought
not be politics, but there is now an appearance problem. One side is saying
one thing, the other side is saying something else. This, it seems to me as
a layman, is not a matter of opinion, but one of fact. I accept that MBH
knew what they were doing in science, but what is the harm in verifying
their work. Perhaps they were sloppy. Perhaps it was a trivial problem.
Perhaps they are frauds who cooked the study. Perhaps they are heroes who
are pioneering new frontiers. What is the harm in checking their work?
Ian St. John
> "Thomas Palm" <Thoma...@chello.removethis.se> wrote in message
> news:Xns951F9DE1CC288T...@212.83.64.229...
>> "Bill Corden" <Cor...@somwhere.com> wrote in
>> For political reasons it would be nice if one of the authors of the
>> original article took the time of explaining the algorithm in enough
>> detail for anyone to follow and added the complete method to
>> calculate it rather than the partial data that has been available
>> (although M&M didn't find much of that as shown in the earlier
>> discussion here). I suspect this is a bit of work, though, as their
>> original program isn't exaclty user friendly.
>>
>> As far as I'm concerned the fact that other researchers has been
>> able to get essentially the same result using their own selection of
>> datasets and calculations is a lot more important than if someone
>> can get exactly the same result as MBH by plugging the same numbers
>> into the same program.
>
> So we should just trust that these results are just fine.
No. But you should assume that competent researchers all checking each
others work are almost certain to get the 'right' results in terms of the
state of the art today.
> Again, I am ignorant of the sciences involved, but that strikes me as a
dubious
> way establish science.
You are welcome to question the science but only if you have some basis for
that question. As you claim you know fuck all, the 'dubious' distinction
should be to your opinion, since it is, by definition, baseless and
ignorant.
> If no one can actually recreate the exact
> numbers using the exact numbers, how can we know they are valid.
How can you know that the sun will rise tomorrow? Probability says that
competent researchers, gettting the same values as other competent
researchers, probably did it right. There is no need to exactly replicate
the work. Now, if you could come up with an actual error in the data or
methodology, you might be worth listening to.
Thomas Palm
> "Thomas Palm" <Thoma...@chello.removethis.se> wrote in message
>> As far as I'm concerned the fact that other researchers has been able
>> to get essentially the same result using their own selection of
>> datasets and calculations is a lot more important than if someone can
>> get exactly the same result as MBH by plugging the same numbers into
>> the same program.
>
> So we should just trust that these results are just fine. Again, I am
> ignorant of the sciences involved, but that strikes me as a dubious
> way establish science. If no one can actually recreate the exact
> numbers using the exact numbers, how can we know they are valid. Has
> anyone checked the other "results"? Would it be reasonable for
> Enron's auditors to say, we use the exact same accounting techniques
> used by Worldcom, should that satisfy?
There are two ways to become famous in science. If you are creative you
discover something new, if you are less creative you try to repeat
someone elses experiment in the hope of finding him wrong. There has been
several groups using the same techiques for several years now, and I have
no doubt that if the true answer looked anything like what M&M gets
someone would have found out so by now. In fact, it would be downright
embarrasing to the contrarian society if it turned out that they have for
years trying to discredit the MBH paper without anyone finding anything
wrong with it if it actually is wrong.
> Please understand that I am not impuning these scientist by comparing
> them to Enron and Worldcom. But in auditing (in my field, accounting)
> you assume the person is stealing, then prove that they could not
> have. Must we rely on their MBH's reputation, or can we examine the
> evidence and methods directly?
To make the situation comparable, assume that all the numbers used by the
auditors at Enron had been publically available, how long would they have
gotten away with what they did? Even if MBH did some mistakes in
describing the details of what they had done, the datasets they started
from are generally available, and once you have the basic idea it isn't
that hard to reconstruct the method, in fact, I would imagine that there
are a number of grad students who have done it as an excercise to learn
it.
Until there is any evidence of any serious mistakes by MBH and the other
groups getting similar results I'm going to trust they knew what they
were doing.
>> To understand their second stupidity you don't even need to know any
>> physics, though:
>> "You see, some stations had missing values, months where no
>> temperature had been recorded. When calculating the root mean square
>> they treated the missing values as if they were measurements of 0
>> degrees. This is incorrect, since the temperature was not actually
>> zero degrees. Because the overall average temperature was positive
>> this meant that the root mean square was biased downwards when there
>> were missing observations. And since there were more missing values
>> in the second half of the time series, this produced a spurious
>> cooling trend. "
>
> I thought this related to an article unrelated to the issue. Yes it
> goes to credibility, but has it been shown that M&M did this in the
> "Audit"?
No, this has nothing to do with the audit, it is just that the mistakes
they did there are of a kind that are technical and somewhat hard to
understand, so instead it is easier to point to much simpler errors made
by one of the authors to show his competence, or lack thereof.
>> Setting missing data to 0 C to get an average is crazy, which I hope
>> you agree on. With enough missing datapoints the temperature is
>> guaranteed to be close to 0 C no matter what the real temperature is.
>> (I guess we should be happy they didn't use 0 K instead or we might
>> have gotten really wierd results).
>
> It is well that we are not all disqualified when we make a mistake.
> In the case at hand, McKitrick made the data available to his critics.
It was nice of him to make the data available, but that doesn't change
the fact that he made a serious error on a trivial issue, and didn't
understand it even when it was pointed out to him. Making the mistake can
be forgiven, even smart people sometimes makes stupid mistakes, but
insisting on it, that was stupid.
>> Since M&M from the start used a "gotcha" approach towards MBH you
>> shouldn't be surprised if they are treated by the same attitude
>> themselves, and as far as I can tell the only substance in
>> McKitrick's response was that it reinforced the impression that he
>> hasn't got a clue what he is talking about. It wasn't just a
>> temporary slip that let those mistakes through in the book.
>
> Perhaps, but again, he subjected his data to his critic's scrutiny.
> SHOULD not MBH be required to do the same now that errors have been
> admitted? In otherwords, would it not be wise that everyone on both
> sides of this issue want to look at the methods and data as applied by
> MBH. Am I correct that there is no proprietary aspect to this. I
> would never allow my competitors to see my source code, but is it not
> normal in this case to allow others to see everything?
I know that in the articles I've published most of the data and programs
haven't been publically available. Had someone asked me, I would have
denied it to them, not because I considered it secret but because it was
a mess and it would take far too long to transform it into something that
could be understood by others.
Many researchers are a lot more secretive, they know that their software
or some experimental method gives them an adantage over the competition
so they make sure to leave it out of any publication. This is somewhat
against the idea of research, but in a "publish or perish" environment it
is unavoidable. If the result is important someone will do a similar
analysis, as indeed happened soon after MBH published their first paper.
>> How would you react if some years after you have published a widely
>> acclaimed article some nobodies without any credentials sent a
>> message stating that they intended to audit your result? The whole
>> approach just screams "crackpot" and I'm not susprised that they got
>> a fairly brief response.
>
> Fair enough. These guys are busy with current research. But, after
> the errors and/or omissions have been admitted, it seems to me that it
> calls into question the results. In politics we would call it an
> "appearance of impropriety."
I guess MBH aren't politicians. In science you have a lot of people with
easily bruised egos. How you appear to the general public is simply
something that most researchers don't think that much about. Consider for
example Schneider's infamous quote about honesty in presenting research.
Taken in context it makes sense, but anyone trained in PR would have
realized you don't say anything like that in public since it is so easily
used by your enemies.
> One side is saying one thing, the other side is
> saying something else. This, it seems to me as a layman, is not a
> matter of opinion, but one of fact.
Had M&M published their result in a respectable scientific journal it
might have been like that, but now they picked E&E with a friendly editor
and lax review practices while MBH got published in Nature. I think I
trust Nature's reviewers more.
> I accept that MBH knew what they
> were doing in science, but what is the harm in verifying their work.
> Perhaps they were sloppy. Perhaps it was a trivial problem.
If I'm allowed to guess it wasn't the same person who put together the
text of the article that made the simulations. That way it is easy to use
an old file without realizing it.
> Perhaps
> they are frauds who cooked the study. Perhaps they are heroes who are
> pioneering new frontiers. What is the harm in checking their work?
None. Feel welcome to do it. If you search back in this forum you will
find the directories where their data can be found. Just be prepared to
spend some time learning about the method they used so you don't just
repeat the mistakes by M&M.
Robert Grumbine
Bill Corden <Cor...@somwhere.com> wrote:
>"Thomas Palm" <Thoma...@chello.removethis.se> wrote in message
>news:Xns951F9DE1CC288T...@212.83.64.229...
>> "Bill Corden" <Cor...@somwhere.com> wrote in
>> For political reasons it would be nice if one of the authors of the
>> original article took the time of explaining the algorithm in enough
>> detail for anyone to follow and added the complete method to calculate it
>> rather than the partial data that has been available (although M&M didn't
>> find much of that as shown in the earlier discussion here). I suspect
>> this is a bit of work, though, as their original program isn't exaclty
>> user friendly.
>>
>> As far as I'm concerned the fact that other researchers has been able to
>> get essentially the same result using their own selection of datasets and
>> calculations is a lot more important than if someone can get exactly the
>> same result as MBH by plugging the same numbers into the same program.
>
>So we should just trust that these results are just fine.
No, quite the contrary, and that was Thomas's point. A much
_more_ challenging test has been done, by independant investigators.
Namely, rather than believe that MBH got _any_ of their analysis
correct, or _any_ of the data correct, these other investigators
went out after their own, independant (often) data sources and used
their own independant analysis methods.
In other words, multiple assemblages of data analyzed in multiple
ways all are arriving at comparable results.
>Again, I am
>ignorant of the sciences involved, but that strikes me as a dubious way
>establish science. If no one can actually recreate the exact numbers using
>the exact numbers, how can we know they are valid.
The 'exact' numbers are, push come to shove, not especially important.
This applies as well to the Spencer and Christy analyses.
We're looking at science, specifically observing something about
nature. Your only source for what Enron did is Enron. They lied
high, wide, and handsome (and with the knowledge of their auditors).
Since we're doing science, or talking about doing science, we
don't have to rely on what any particular scientist(s) say the data
are. We're observing nature. There are many possible sources of
observations. Any given author or group of authors knows a subset
of them, and a particular set of tools for analysis. So they analyze
that subset, using the tools they know. And they do tell us
(people in the field who understand the sorts of data and methods
involved) what data and methods they used.
One thing which can be done is to take the 'exact same' data and
analyze it in the 'exact same' way. This is uninteresting as science,
and MM showed that it can't be done either by people who don't
understand that there really are reasons that you have to pick
particular ways of averaging.
The interesting thing is to use either or both different methods
or different data sources. This has been done for MBH, with people
finding much the same answers as MBH did, in spite of the differences
in data and analysis. That lends confidence (to scientists) that
MBH really were observing something from nature, and not just
artefacts of their methods or choice of data sets. To the extent
that the results differ, there is then something to explain -- about
nature (one hopes) or about the data sources and methods. Getting
the same result with the same data set tells us nothing about the
value or reliability of either the data sources or the methods. That's
why it's scientifically uninteresting.
The same sort of thing (primarily difference in analysis methods)
has been done with Spencer and Christy's work. Different people
taking different (_a_priori_ reasonable) analysis methods have
for years consistently arrived at results that disagree substantially
with Spencer and Christy, and agree more so with each other and
with other observing methods. After each round of this, Spencer
and Christy revise their work, and arrive at a higher warming rate
than their previous method. It does not instill confidence in
their science that they are _always_ getting the lowest trends,
in spite of the fact that they have gone through several iterations
of revising their methods.
To repeat, when other people attack the same problem as MBH, in
their respectively different ways, and using different data sources,
they _do_ get largely similar results. The differences are capable
of study. And at no time do we have to believe that any particular
group has it all perfectly correct.
[snip]
>Please understand that I am not impuning these scientist by comparing them
>to Enron and Worldcom.
Er, it's hardly an accolade.
>But in auditing (in my field, accounting) you assume
>the person is stealing, then prove that they could not have.
Too bad the auditors of Enron et al. didn't know that.
>> To understand their second stupidity you don't even need to know any
>> physics, though:
>> "You see, some stations had missing values, months where no temperature
>> had been recorded. When calculating the root mean square they treated the
>> missing values as if they were measurements of 0 degrees. This is
>> incorrect, since the temperature was not actually zero degrees. Because
>> the overall average temperature was positive this meant that the root
>> mean square was biased downwards when there were missing observations.
>> And since there were more missing values in the second half of the time
>> series, this produced a spurious cooling trend. "
>
>I thought this related to an article unrelated to the issue. Yes it goes to
>credibility, but has it been shown that M&M did this in the "Audit"?
Would you trust the books done by someone who thought you could just
fill in arbitrary values for checks which were missing? We don't trust
that sort of thing in science, and it's what McKitrick did in the previously
cited example.
>> Since M&M from the start used a "gotcha" approach towards MBH you
>> shouldn't be surprised if they are treated by the same attitude
>> themselves, and as far as I can tell the only substance in McKitrick's
>> response was that it reinforced the impression that he hasn't got a clue
>> what he is talking about. It wasn't just a temporary slip that let those
>> mistakes through in the book.
>
>Perhaps, but again, he subjected his data to his critic's scrutiny. SHOULD
>not MBH be required to do the same now that errors have been admitted? In
>otherwords, would it not be wise that everyone on both sides of this issue
>want to look at the methods and data as applied by MBH. Am I correct that
>there is no proprietary aspect to this. I would never allow my competitors
>to see my source code, but is it not normal in this case to allow others to
>see everything?
The sources and methods were documented sufficient for people in the
field. It turns out, and now corrected, that some peripheral sources
weren't as precisely documented.
In the prior incarnation of the thread, it was gone over at some
length that if M+M had the vaguest notion of the area they were auditing,
they wouldn't have had many of the problems they said they did. While
accurate description of the data sets is indeed something to expect,
going in to sufficient detail for people far outside the field to
reproduce your work is not expected. Nor is it really possible.
McKitrick would need to start with freshman chemistry and physics,
at the least, and it's absurd to expect working scientists to
re-teach elementary courses to anybody emailing them. (Working
scientists do (often) teach courses -- as a specific part of their
jobs. McKitrick really could sign up for freshman chem and physics
at his university. It's just silly, though, to expect Mann to do it
as private tutoring.)
[snip]
>Fair enough. These guys are busy with current research. But, after the
>errors and/or omissions have been admitted, it seems to me that it calls
>into question the results.
The fact that other people, using other methods on other data, get
substantially the same results, however, says that the details which
just got corrected did not substantially affect the original paper.
This is science. The first paper on a new idea is important (only)
for being the first paper on a new idea. We always revise as we learn
more and better data and methods. It isn't pronouncements from
on high that can never be questioned, challenged, or modified.
What, exactly, was done in the 1998 paper to exactly what data is,
at this point, pretty much only of historical interest. In the
intervening 6-8 years we've found more data sources, recalibrated
some of the old ones, learned even more about how to analyze the sources,
etc, etc. -- and still arrive at similar conclusions to the 1998 paper.
Had the paper been fraudulent, then papers (by other people) from 1999
through 2004 would now have shown it up.
The obsession some have with the 1998 paper is looking a lot like
the Young Earth Creationists who (try to) attack biology by saying
that Darwin got something wrong. They're generally wrong in that,
but, even so, so what -- he didn't lie, and did have a good idea.
Even if every page of the old work were shown to be wrong, by current
knowledge, biology would still be a perfectly decent science. We've
just learned more in the mean time. So it goes with MBH 1998, though a
more recent and less grand idea.
We now have, not just MBH 1998, but ME 2000, YOU 2001, HIM 2000,
OTHER 1999, Anon et al. 2003, Ibid 2004, ...
Once a good new idea is proposed, others start beating the heck out
of it. We just don't do it by the trivial challenge of trying to
get the same numbers in the same way.
>Now, this is not a legal matter and ought
>not be politics, but there is now an appearance problem. One side is saying
>one thing, the other side is saying something else. This, it seems to me as
>a layman, is not a matter of opinion, but one of fact. I accept that MBH
>knew what they were doing in science, but what is the harm in verifying
>their work. Perhaps they were sloppy. Perhaps it was a trivial problem.
>Perhaps they are frauds who cooked the study. Perhaps they are heroes who
>are pioneering new frontiers. What is the harm in checking their work?
To repeat, as you repeat the question: Their work _has been_ checked,
_is being_ checked, and _will be_ checked on in to the future. It was/is/
will be done in the much more challenging way of using different data sources
and different methods.
--
Robert Grumbine http://www.radix.net/~bobg/ Science faqs and amateur activities notes and links.
Sagredo (Galileo Galilei) "You present these recondite matters with too much
evidence and ease; this great facility makes them less appreciated than they
would be had they been presented in a more abstruse manner." Two New Sciences
Per
> On 7 Jul 2004 01:17:35 -0700, perox...@ntlworld.com (Per) wrote:
> >> Bullshit. They claim to have published an 'audit'. Further
> >> inspection shows that clearly not to be the case. The peer-review was
> >> shoddy. There's a reason they published in E&E. As for experts, a rank
> >> amateur should have spotted the problematic nature of M&M's results.
> >
> >You have no knowledge of what happened during M&M's peer review; your
> >claims to the contrary reveal you as a fantasist and liar.
>
> LOL. I claim that any peer-reviewer worth his or her salt
> should have noticed the discrepency between M&M's results and reality
You have no knowledge of what happened during M&M's peer review; your
claims to the contrary reveal you as a fantasist and liar.
>
> >
> >> > In fact, the corrigendum explicitly
> >> >refers to McIntyre and McKitrick:
> >> >"It has been drawn to our attention (S. McIntyre and R. McKitrick)
> >> >that the listing of the ?proxy' data set in the Supplementary
> >> >Information published with this Article contained several errors."
> >Dear David
> >READ the text of the corrigendum. I have stripped it down so you only
> >have to parse one sentence for its content. See if you can understand
> >the text.
No sooner do I post MBH's explicit acknowledgement that they have made
ERRORS, than david says:
"LOL. They accepted the point that the supplementary information as
presented is incomplete. Nothing more."
It was only one sentence, and yet, somehow, David Ball couldn't quite-
ah, well...
per
Per
> On 7 Jul 2004 05:31:37 -0700, perox...@ntlworld.com (Per) wrote:
> >Actually, you are wrong as a matter of fact. M&M found that the MBH'98
> >paper had defects in its dataset, and published that finding. We now
> >know that MBH agree with this finding; and the MBH'98 dataset has been
> >updated- which is scientifically significant.
>
> Bull. They found that there were omissions from the
> supplementary data which apparently have been corrected. You are so
> desperate to prove M&M right that you've lost all semblence of
> objectivity.
An omission of data is a defect in a dataset; which is what I said.
MBH themselves are stronger; they say that their original paper had
errors and omissions.
It doesn't matter how much it upsets you; the errors pointed out by
M&M have been acknowledged as errors by MBH themselves. No spin can
take away the fact that MBH have acknowledged M&M's identification of
errors as correct, and that they have done so in print.
It is thoroughly weird that you try and minimise a factual reality.
per
Per
It is one thing to read the references. It is another thing to read
the references, and then have to guess that maybe- just maybe- some of
the stuff in the references MAY have been truncated in the original
paper. This isn't science you are talking about- it is parlour
gimmickry. What is more, MBH now acknowledge that this standard is
inadequate, because they now identify the ranges of the data that they
used.
per
Per
> As far as I'm concerned the fact that other researchers has been able to
> get essentially the same result using their own selection of datasets and
> calculations is a lot more important than if someone can get exactly the
> same result as MBH by plugging the same numbers into the same program.
this is yet another set of weasel words to get around an absolute, and
guiding principle, of science; that science must be reproducible. If
MBH isn't reproducible, it ain't science. If the other datasets aren't
reproducible, they ain't science either. It wouldn't be the first time
that people have come up with numbers in agreement with one another,
and everyone has been wrong.
> How would you react if some years after you have published a widely
> acclaimed article some nobodies without any credentials sent a message
> stating that they intended to audit your result? The whole approach just
> screams "crackpot" and I'm not susprised that they got a fairly brief
> response.
This is an "ad hominem" attack. In fact, M&M have the data analysis
expertise to repeat the number crunching used by MBH. "nobodies",
"crackpot"; calling people names simply serves to underline the fact
that MBH have been forced to issue a corrigendum which explicitly
states that M&M brought MBH's attention to errors in MBH'98.
> We already know that M&M are hacks. The discussion after their original
> publication of their paper made that perfectly clear. The remaining
> issue is whether or not they by pure chance managed to attack a paper
> that turns out to be wrong. Their own alternative reconstruction is
> discredited.
Oh, now we have heard it all ! "M&M are hacks"- but they might have
discovered- by PURE CHANCE- a paper that turns out to be wrong. They
might have made specific allegations - presumably by pure chance
again- which MBH now acknowledge to be correct.
ooh, that is funny.
per
Steve Schulin
bo...@radix.net (Robert Grumbine) wrote:
> In article <b0YGc.2$WS...@fe39.usenetserver.com>,
> Bill Corden <Cor...@somwhere.com> wrote:
> >"Thomas Palm" <Thoma...@chello.removethis.se> wrote in message
> >news:Xns951F9DE1CC288T...@212.83.64.229...
> >> "Bill Corden" <Cor...@somwhere.com> wrote in
> >> For political reasons it would be nice if one of the authors of the
> >> original article took the time of explaining the algorithm in enough
> >> detail for anyone to follow and added the complete method to calculate it
> >> rather than the partial data that has been available (although M&M didn't
> >> find much of that as shown in the earlier discussion here). I suspect
> >> this is a bit of work, though, as their original program isn't exaclty
> >> user friendly.
> >>
> >> As far as I'm concerned the fact that other researchers has been able to
> >> get essentially the same result using their own selection of datasets and
> >> calculations is a lot more important than if someone can get exactly the
> >> same result as MBH by plugging the same numbers into the same program.
> >
> >So we should just trust that these results are just fine.
>
> No, quite the contrary, and that was Thomas's point. A much
> _more_ challenging test has been done, by independant investigators.
> Namely, rather than believe that MBH got _any_ of their analysis
> correct, or _any_ of the data correct, these other investigators
> went out after their own, independant (often) data sources and used
> their own independant analysis methods.
All "multiproxy" approaches share a common disadvantage -- the smearing
of individual series' dating error. Relatively flat results are to be
expected. A recent paper on this was the starting point for a thread
back in May titled: Loehle on "Using Historical Climate Data to Evaluate
Climate Trends: Issues of Statistical Inference"
the thread-starting post is archived at
http://makeashorterlink.com/?J1C821FB8
which points to
http://groups.google.com/groups?q=g:thl3293691096d&dq=&hl=en&lr=&ie=UTF-8
&selm=steve.schulin-38CAC3.11112903052004%40comcast.ash.giganews.com
But that's not to say that every combination of proxies will indeed
yield flat results. In MBH's Oct 31, 2003 "NOTE ON PAPER BY MCINTYRE AND
MCKITRICK IN 'ENERGY AND ENVIRONMENT'", for example, Mann et al. show
the difference that just three "critical datasets" have on their results
for 1400-1500.
There is absolutely no doubt that Mann et al. 1998 was unreplicable as
published. In the recent Corrigendum, MBH refer to over 500 series being
analyzed. Before M&M, the only number (as used by MBH98 and Zorita et
al.) was 112. In initial reply to M&M, the number 159 was offered. I
don't presume to know why we keep being told different numbers by Mann
et al, but I sure am curious. That the hockey stick got elevated to
consensus status so quickly reflects poorly upon the so-called consensus
group. To those of you who have seen Mann's ftp site get vacuumed as
clean as Rose Law Firm files, and have remained silent while these
sci.environment thugs like Grumbine, Ball, and others pour out their
dissembling abuse, I urge you to take a stand for the integrity of your
discipline.
Very truly,
Steve Schulin
http://www.nuclear.com
Bill Corden
> discover something new, if you are less creative you try to repeat
> someone elses experiment in the hope of finding him wrong. There has been
> several groups using the same techiques for several years now, and I have
> no doubt that if the true answer looked anything like what M&M gets
> someone would have found out so by now. In fact, it would be downright
> embarrasing to the contrarian society if it turned out that they have for
> years trying to discredit the MBH paper without anyone finding anything
> wrong with it if it actually is wrong.
Has anyone reviewed it and said: Yeah, I reviewed all the data and it is
correct? Have they said "I stake my reputation on this." ?
> > Please understand that I am not impuning these scientist by comparing
> > them to Enron and Worldcom. But in auditing (in my field, accounting)
> > you assume the person is stealing, then prove that they could not
> > have. Must we rely on their MBH's reputation, or can we examine the
> > evidence and methods directly?
>
> To make the situation comparable, assume that all the numbers used by the
> auditors at Enron had been publically available, how long would they have
> gotten away with what they did? Even if MBH did some mistakes in
> describing the details of what they had done, the datasets they started
> from are generally available, and once you have the basic idea it isn't
> that hard to reconstruct the method, in fact, I would imagine that there
> are a number of grad students who have done it as an excercise to learn
> it.
Well, M&M claim very specifically what they could not find, and what MBH did
not provide links to. You agree that this data should be made available and
if it is not, there is reason for suspicion? That is: Why not?
> Until there is any evidence of any serious mistakes by MBH and the other
> groups getting similar results I'm going to trust they knew what they
> were doing.
Really going out on the limb there, huh? I mean really, what is that
supposed to mean? You are not going to check it out yourself, but if you
are proven wrong, you will accept it, but you will defend these guys until
then?
> > It is well that we are not all disqualified when we make a mistake.
> > In the case at hand, McKitrick made the data available to his critics.
>
> It was nice of him to make the data available, but that doesn't change
> the fact that he made a serious error on a trivial issue, and didn't
> understand it even when it was pointed out to him. Making the mistake can
> be forgiven, even smart people sometimes makes stupid mistakes, but
> insisting on it, that was stupid.
I agree he was bull-headed when his mistake was pointed out.
> I know that in the articles I've published most of the data and programs
> haven't been publically available. Had someone asked me, I would have
> denied it to them, not because I considered it secret but because it was
> a mess and it would take far too long to transform it into something that
> could be understood by others.
So, a scientific paper is just an idea? We looked at this stuff and did
some figuring and this is what it means. Maybe you should take our idea and
see if you you can get the same result. I did not know that is how it is
done. I guess I assumed it was like a mathmaitical proof. Either your
could provide the "proof" or you could not. Credibility appears to play a
bigger role than I thought in all this.
> Many researchers are a lot more secretive, they know that their software
> or some experimental method gives them an adantage over the competition
> so they make sure to leave it out of any publication. This is somewhat
> against the idea of research, but in a "publish or perish" environment it
> is unavoidable. If the result is important someone will do a similar
> analysis, as indeed happened soon after MBH published their first paper.
So if questioned, does the author have no obligation to back up his work
with data? I just can't believe there is so much "trust" in the system.
Won't prove it, but trust me, I am right.
> > Fair enough. These guys are busy with current research. But, after
> > the errors and/or omissions have been admitted, it seems to me that it
> > calls into question the results. In politics we would call it an
> > "appearance of impropriety."
>
> I guess MBH aren't politicians. In science you have a lot of people with
> easily bruised egos. How you appear to the general public is simply
> something that most researchers don't think that much about. Consider for
> example Schneider's infamous quote about honesty in presenting research.
> Taken in context it makes sense, but anyone trained in PR would have
> realized you don't say anything like that in public since it is so easily
> used by your enemies.
The ends justify the means? I don't agree with his statement (which I found
thanks to Google). Truth has a strength all its own. It is interesting
that you site poor "PR" when ALL he was talking about was PR. The arrogance
of presuming to know what is "effective" set in contrast to what is "true."
If truth does not exist, perhaps he is right. If truth does exist, then he
should be ashamed. And as far as that man is concerned: I do not trust what
he says. Only a fool would trust someone who states up front that their job
is to deceive.
> > One side is saying one thing, the other side is
> > saying something else. This, it seems to me as a layman, is not a
> > matter of opinion, but one of fact.
>
> Had M&M published their result in a respectable scientific journal it
> might have been like that, but now they picked E&E with a friendly editor
> and lax review practices while MBH got published in Nature. I think I
> trust Nature's reviewers more.
Why should this be a matter of trust? Either the data and procedures are
sound or they were not. And if they are unwilling to subject themselves to
evaluation, why should I, not knowing their reputation myself, trust your
assesment of their reliability? In accounting, the auditor cannot just say
"Trust me, they are honest and all their friends think they are honest too."
Is the scientific standard different?
> > I accept that MBH knew what they
> > were doing in science, but what is the harm in verifying their work.
> > Perhaps they were sloppy. Perhaps it was a trivial problem.
>
> If I'm allowed to guess it wasn't the same person who put together the
> text of the article that made the simulations. That way it is easy to use
> an old file without realizing it.
> > Perhaps
> > they are frauds who cooked the study. Perhaps they are heroes who are
> > pioneering new frontiers. What is the harm in checking their work?
>
> None. Feel welcome to do it. If you search back in this forum you will
> find the directories where their data can be found. Just be prepared to
> spend some time learning about the method they used so you don't just
> repeat the mistakes by M&M.
So, if they (MSB) is unwilling to provide necessary data to corroborate
their findings, you would criticise them?
Thanks for you thoughtful response.
Bill Corden
news:10eooqv...@corp.supernews.com...
> The obsession some have with the 1998 paper is looking a lot like
> the Young Earth Creationists who (try to) attack biology by saying
> that Darwin got something wrong. They're generally wrong in that,
> but, even so, so what -- he didn't lie, and did have a good idea.
> Even if every page of the old work were shown to be wrong, by current
> knowledge, biology would still be a perfectly decent science. We've
> just learned more in the mean time. So it goes with MBH 1998, though a
> more recent and less grand idea.
>
> We now have, not just MBH 1998, but ME 2000, YOU 2001, HIM 2000,
> OTHER 1999, Anon et al. 2003, Ibid 2004, ...
Thank you for the effort you put into the post. I sense that you are indeed
earnist. I am a little annoyed at you because I tried to Google "ME 2000"
then "YOU 2001" and did not get very satisfactory results<g>. You did
mention Spencer and Christy, and I will read their work.
> Once a good new idea is proposed, others start beating the heck out
> of it. We just don't do it by the trivial challenge of trying to
> get the same numbers in the same way.
>
> >Now, this is not a legal matter and ought
> >not be politics, but there is now an appearance problem. One side is
saying
> >one thing, the other side is saying something else. This, it seems to me
as
> >a layman, is not a matter of opinion, but one of fact. I accept that MBH
> >knew what they were doing in science, but what is the harm in verifying
> >their work. Perhaps they were sloppy. Perhaps it was a trivial problem.
> >Perhaps they are frauds who cooked the study. Perhaps they are heroes
who
> >are pioneering new frontiers. What is the harm in checking their work?
>
> To repeat, as you repeat the question: Their work _has been_ checked,
> _is being_ checked, and _will be_ checked on in to the future. It was/is/
> will be done in the much more challenging way of using different data
sources
> and different methods.
I really hate to quibble, but their work proper cannot be checked unless
they provide sufficient information to check it. The "go do your own damn
study" seems to me a rather insecure response. If I am trying to prove
something to someone, I provide the data. Again, I understand my ignorance
and appreciate the time you have taken with me. I am really trying to
understand. I do not yet.
Thanks,
Bill
Bill Corden
> Bill Corden wrote:
> > "Thomas Palm" <Thoma...@chello.removethis.se> wrote in message
> > news:Xns951F9DE1CC288T...@212.83.64.229...
> >> "Bill Corden" <Cor...@somwhere.com> wrote in
> dubious
> > way establish science.
>
> You are welcome to question the science but only if you have some basis
for
> that question. As you claim you know fuck all, the 'dubious' distinction
> should be to your opinion, since it is, by definition, baseless and
> ignorant.
So just shut up? Got it. Thanks. Too bad I am not six years old and too
bad you are not my father. But, other than that, it is a real fine approach
to the dealing with people. Maybe you are having a bad day.
> > If no one can actually recreate the exact
> > numbers using the exact numbers, how can we know they are valid.
>
> How can you know that the sun will rise tomorrow? Probability says that
> competent researchers, gettting the same values as other competent
> researchers, probably did it right. There is no need to exactly replicate
> the work. Now, if you could come up with an actual error in the data or
> methodology, you might be worth listening to.
"Trust us. We are scientists." Perhaps we should continue after you have
had some sleep or a little less coffee.
Per
> Oh, God, Per, it's been done repeatedly. Over and over again
> ad-nauseum. M&M clearly did not RTFR provided by MBH. On their
> climate2003 page they claim oversights on the part of MBH that are not
> at all. All they had to do was read the references. For God's sake,
> they claimed their were these huge errors in the spreadsheet provided
> to them and instead of using their brains and thinking perhaps that
> the spreadsheet was corrupt they leapt to the unwarranted conclusion
> that this was indeed the data used by MBH and given that their results
> were incorrect. It never occurred to the two of them that data coming
> from a Linux/Unix environment might be shifted on a WinDoze machine.
> Their work was sloppy, superficial and showed absolutely nothing,
> except perhaps that science shouldn't be done by amateurs.
First of all, David, let me encourage you to remove irrelevant text.
Secondly, when you claim that M&M did not read the references in
MBH'98, this is a baseless fabrication, made in total ignorance of
what M&M did, or did not read.
Your claims about the spreadsheet have previously been shown to be
untrue, and you know this. M&M corresponded with MBH by email, and
received assurances that the spreadsheet was correct. This
correspondence is available on the M&M web site. This has been pointed
out to you on several occasions, and yet you persist in this untruth
which you know to be untrue.
Finally, MBH have now themselves admitted many of the inadequacies
pointed out by M&M, yet you persist in denying. The horse has bolted,
it is over the hill, yet you are frantically trying to keep the barn
door shut.
per
Per
> The 'exact' numbers are, push come to shove, not especially important.
> One thing which can be done is to take the 'exact same' data and
> analyze it in the 'exact same' way. This is uninteresting as science,
> Getting
> the same result with the same data set tells us nothing about the
> value or reliability of either the data sources or the methods. That's
> why it's scientifically uninteresting.
I thought it was worth taking these three quotes, because they reveal
so much. In fact, getting the same result with the same data set is a
PREREQUISITE for any science. The exact numbers you get are
fundamentally important. And a fundamental of all science is that it
is reproducible.
> To repeat, as you repeat the question: Their work _has been_ checked,
> _is being_ checked, and _will be_ checked on in to the future. It was/is/
> will be done in the much more challenging way of using different data sources
> and different methods.
Strangely, Bob won't be giving us any literature references for people
who have directly checked MBH'98, and got the same results. This is
because the only one of which I am aware is M&M. And the unfortunate
thing (for Bob) is that M&M have found errors in the MBH dataset,
which MBH have now accepted as being errors.
No-one else can have replicated the MBH data until they had the
correct data set and methods to work with, and MBH have admitted that
that was not available from looking at their original paper. It will
be interesting to see what a replication throws up.
per
Nigel Persaud
Bill,
most of the so-called "independent" studies are made by a few guys -
Mann and Jones write one paper, Briffa, Jones, Schweingruber another,
Bradley, Hughes and Diaz write one, Jones and Bradley write one, Mann,
Bradley and Hughes write one; Crowley and Lowery write one using data
from Jones; Hughes and Diaz write one. A very very small group are
being relied on, swapping partners from time to time. I'd add Cook and
Jacoby to the group. The various studies cited as independent often
use the same data: Briffa's Tornetrask and Polar Urals series are
nearly always used. There's not nearly as much independence as one
thinks.
You will notice that there is much sputtering on this list about how
awful M&M are, but not much actual evidence, beyond quoting someone
else who said they were awful. I disagree with the RTFR point - as I
mentioned before, this pertained to the Central England and Central
Europe series in which MBH used truncated versions without disclosure.
M&M pointed this out. Josh and others argued that the references could
be read in a way to justify the truncation. Perhaps so, but I think
not. They don't like to talk about the opposite usage in Jones et al
1998. But the issue is hardly embarrassing to M&M, whose position on
this is entirely reasonable IMHO.
Regards, Nigel Persaud
Per
> There are two ways to become famous in science. If you are creative you
> discover something new, if you are less creative you try to repeat
> someone elses experiment in the hope of finding him wrong. There has been
> several groups using the same techiques for several years now, and I have
> no doubt that if the true answer looked anything like what M&M gets
> someone would have found out so by now.
There are many examples of fraud in science. Many of these were
published in high impact journals, and many of them were not found out
till much later.
You are trying to escape the fundamental of science; that science must
be repeatable. If it isn't repeatable, it is not science. This is a
pillar of the scientific edifice.
As a point of fact, can you show us any paper which has directly
replicated the results of MBH '98 ? If you cannot show us such a
paper, how can you assure us it is correct ?
> I know that in the articles I've published most of the data and programs
> haven't been publically available. Had someone asked me, I would have
> denied it to them, not because I considered it secret but because it was
> a mess and it would take far too long to transform it into something that
> could be understood by others.
I don't know what field you publish in, but I do not see how that can
be an ethically acceptable standpoint. Nature, Science, and the other
major publications make it a condition of publication that you release
data, programs, and some specialised reagents. I know of no reputable
professional body that would find your viewpoint ethically acceptable.
per
Thomas Palm
> "Robert Grumbine" <bo...@radix.net> wrote in message
>> We now have, not just MBH 1998, but ME 2000, YOU 2001, HIM 2000,
>> OTHER 1999, Anon et al. 2003, Ibid 2004, ...
>
> Thank you for the effort you put into the post. I sense that you are
> indeed earnist. I am a little annoyed at you because I tried to
> Google "ME 2000" then "YOU 2001" and did not get very satisfactory
> results<g>.
Here are a couple of links you may find useful:
http://www.ngdc.noaa.gov/paleo/pubs/jones2001/jones2001.html
http://www.ngdc.noaa.gov/paleo/recons.html
http://www.ngdc.noaa.gov/paleo/pubs/briffa2001/briffa2001.html
http://www.ngdc.noaa.gov/paleo/pubs/mann2003/mann2003.html
(The last study use an entirely different method by taking temperature
measurments from boreholes. That's how you do a independent study)
> You did mention Spencer and Christy, and I will read
> their work.
S&C are analyzing data from meteorological satellites so that debate
concerns a much more recent temperature record. What is amusing is that the
same people who have critizised MBH for not beeing reproducible, despite
other getting the same results, have uncritically accepted the results by
S&C despite the fact that those haven't been reproduced either, and that in
this case others get considerably different results. S&C has even had to
admit to a couple of pretty serious errors over the years.
Ian St. John
> "Ian St. John" <ist...@noemail.ca> wrote in message
> news:_oYGc.30071$WM5.1...@news20.bellglobal.com...
>> Bill Corden wrote:
>>> "Thomas Palm" <Thoma...@chello.removethis.se> wrote in message
>>> news:Xns951F9DE1CC288T...@212.83.64.229...
>>>> "Bill Corden" <Cor...@somwhere.com> wrote in
>>> Again, I am ignorant of the sciences involved, but that strikes me
>>> as a dubious way establish science.
>>
>> You are welcome to question the science but only if you have some
>> basis for that question. As you claim you know fuck all, the
>> 'dubious' distinction should be to your opinion, since it is, by
>> definition, baseless and ignorant.
>
> So just shut up?
No. You should speak when you have something meaningful to say.
> Got it. Thanks.
I guess you have no expectation of having anything meaningful in the near
future.
> Too bad I am not six years old
> and too bad you are not my father. But, other than that, it is a
> real fine approach to the dealing with people. Maybe you are having
> a bad day.
Or maybe I am tired of ( admitted) ignorant opinions being expresssed as
facts.This is a science newsgroup. If you cannot contribute to the science,
then just listen and learn.
>
>>> If no one can actually recreate the exact
>>> numbers using the exact numbers, how can we know they are valid.
>>
>> How can you know that the sun will rise tomorrow? Probability says
>> that competent researchers, gettting the same values as other
>> competent researchers, probably did it right. There is no need to
>> exactly replicate the work. Now, if you could come up with an actual
>> error in the data or methodology, you might be worth listening to.
>
> "Trust us. We are scientists."
More correctly "Trust but verify,and always cut the cards" Robert A.
Heinlein.
Now, the point is that the verification has been done and you have nothing
to contribute.
> Perhaps we should continue after you
> have had some sleep or a little less coffee.
Or maybe you should continue on alt.dumbass.opiinion.whocares
Thomas Palm
>> There are two ways to become famous in science. If you are creative
>> you discover something new, if you are less creative you try to
>> repeat someone elses experiment in the hope of finding him wrong.
>> There has been several groups using the same techiques for several
>> years now, and I have no doubt that if the true answer looked
>> anything like what M&M gets someone would have found out so by now.
>> In fact, it would be downright embarrasing to the contrarian society
>> if it turned out that they have for years trying to discredit the MBH
>> paper without anyone finding anything wrong with it if it actually is
>> wrong.
>
> Has anyone reviewed it and said: Yeah, I reviewed all the data and it
> is correct? Have they said "I stake my reputation on this." ?
Not as far as I know, but then if someone did that it wouldn't be
considered worth publicizing. Scientific journals only publish if there
is something new included.
> Well, M&M claim very specifically what they could not find, and what
> MBH did not provide links to. You agree that this data should be made
> available and if it is not, there is reason for suspicion? That is:
> Why not?
Most of M&M:s claims just showed their own incompetence because when
others looked they did find the data although you had to look around a
bit in the subdirectories.
>> Until there is any evidence of any serious mistakes by MBH and the
>> other groups getting similar results I'm going to trust they knew
>> what they were doing.
>
> Really going out on the limb there, huh? I mean really, what is that
> supposed to mean? You are not going to check it out yourself, but if
> you are proven wrong, you will accept it, but you will defend these
> guys until then?
Until there is a serious attack on their credibility I will assume them
to be innocent of any wrongdoings. That to me is a fairly normal
approach. If some people who have proven that they know what they are
doing start questioning the multiple climate reconstructions I may take
notice, but until then I'm going to assume that the established
scientists know better what they are doing than the amateurs. So far M&M
seem more like anti-evolutionists trying to find minor flaws in some work
by a known biologist's work in the hope to use that flaw as a tool to
claim to prove that all of evolution is a fraud, and in my experience
that kind of people are good at sounding convincing, because they spend
more time on it than your average scientist who spends his time trying to
be correct instead of sounding correct.
>> I know that in the articles I've published most of the data and
>> programs haven't been publically available. Had someone asked me, I
>> would have denied it to them, not because I considered it secret but
>> because it was a mess and it would take far too long to transform it
>> into something that could be understood by others.
>
> So, a scientific paper is just an idea? We looked at this stuff and
> did some figuring and this is what it means. Maybe you should take
> our idea and see if you you can get the same result. I did not know
> that is how it is done. I guess I assumed it was like a mathmaitical
> proof. Either your could provide the "proof" or you could not.
> Credibility appears to play a bigger role than I thought in all this.
It depends on the kind of paper. If you write a paper about a physical
theory you had better make sure people can reproduce your mathemetics,
but experimental papers are a lot more based on faith that the
experimenter knows what he is doing.
>> Many researchers are a lot more secretive, they know that their
>> software or some experimental method gives them an adantage over the
>> competition so they make sure to leave it out of any publication.
>> This is somewhat against the idea of research, but in a "publish or
>> perish" environment it is unavoidable. If the result is important
>> someone will do a similar analysis, as indeed happened soon after MBH
>> published their first paper.
>
> So if questioned, does the author have no obligation to back up his
> work with data? I just can't believe there is so much "trust" in the
> system. Won't prove it, but trust me, I am right.
If there is an official inquiry by the university or some funding agency
you should be prepared to hand over what you got and convince them that
you have been correct, or at least honest. You do not have the obligation
to hand over everything to some random bloke walking up to you.
>> I guess MBH aren't politicians. In science you have a lot of people
>> with easily bruised egos. How you appear to the general public is
>> simply something that most researchers don't think that much about.
>> Consider for example Schneider's infamous quote about honesty in
>> presenting research. Taken in context it makes sense, but anyone
>> trained in PR would have realized you don't say anything like that in
>> public since it is so easily used by your enemies.
>
> The ends justify the means? I don't agree with his statement (which I
> found thanks to Google).
Did you find all of it, or just the part the contrarians love to use?
http://rpuchalsky.home.att.net/sci_env/sch_quote.html
It is not "the end justifies the means" but "How do we tell the truth so
that what reachers the readers of papers is what we intend".
> Truth has a strength all its own. It is
> interesting that you site poor "PR" when ALL he was talking about was
> PR.
Yes, he was talking about PR, but as an amateur who had no schooling in
it. If scientists had that schooling what he said would be redundant. Not
that people involved in PR worry much about the "double ethical bind".
They routinely drop all disclaimers about uncertainties without even
reflecting on whether or not this is justified.
> The arrogance of presuming to know what is "effective" set in
> contrast to what is "true." If truth does not exist, perhaps he is
> right. If truth does exist, then he should be ashamed. And as far as
> that man is concerned: I do not trust what he says. Only a fool would
> trust someone who states up front that their job is to deceive.
"Each of us has to decide what the right balance is between being
effective and being honest. I hope that means being both."
He does not state that he is going to be deceptive. He says that since
the media are going to abbreviate and simplify any story they publish you
can't tell the full truth and expect it to reach the readers. You need to
simplify, and it is then a matter of how you do it. If you think other
people are not doing the same you really are a fool, myself I'd rather
trust the guy who states openly that he is aware of the problem and
considers the compromises you have to make.
>> > One side is saying one thing, the other side is
>> > saying something else. This, it seems to me as a layman, is not a
>> > matter of opinion, but one of fact.
>>
>> Had M&M published their result in a respectable scientific journal it
>> might have been like that, but now they picked E&E with a friendly
>> editor and lax review practices while MBH got published in Nature. I
>> think I trust Nature's reviewers more.
>
> Why should this be a matter of trust?
Because I'm not willing to spend the time required to try to reproduce
the result, and apparently neither are you. In our specialized society
much is based on trust.
> Either the data and procedures
> are sound or they were not. And if they are unwilling to subject
> themselves to evaluation, why should I, not knowing their reputation
> myself, trust your assesment of their reliability? In accounting, the
> auditor cannot just say "Trust me, they are honest and all their
> friends think they are honest too." Is the scientific standard
> different?
In auditing I imagine not anyone can walk up to a company like Enron and
demand to see their books to make an audit, right? MBH has not shown
themselves to be unwilling to an audit by anyone who has a reasonable
claim to be doing one. You may not be aware of it, but Nature is
generally considered to be the number one prestigeous scientific journal.
If you know nothing about the authors of a certain article, it is a good
guess that the editors of such a journal has checked it.
>> > Perhaps
>> > they are frauds who cooked the study. Perhaps they are heroes who
>> > are pioneering new frontiers. What is the harm in checking their
>> > work?
>>
>> None. Feel welcome to do it. If you search back in this forum you
>> will find the directories where their data can be found. Just be
>> prepared to spend some time learning about the method they used so
>> you don't just repeat the mistakes by M&M.
>
> So, if they (MSB) is unwilling to provide necessary data to
> corroborate their findings, you would criticise them?
If Nature or their university or some other body of authority started to
demand to see their complete data and they refused to cooperate, then I
would start to get suspicious.
Steve Schulin
Thomas Palm <Thoma...@chello.removethis.se> wrote:
> "Bill Corden" <Cor...@somwhere.com> wrote in
> news:Sr1Hc.96$WS...@fe39.usenetserver.com:
>
> > "Robert Grumbine" <bo...@radix.net> wrote in message
> >> We now have, not just MBH 1998, but ME 2000, YOU 2001, HIM 2000,
> >> OTHER 1999, Anon et al. 2003, Ibid 2004, ...
> >
> > Thank you for the effort you put into the post. I sense that you are
> > indeed earnist. I am a little annoyed at you because I tried to
> > Google "ME 2000" then "YOU 2001" and did not get very satisfactory
> > results<g>.
>
> Here are a couple of links you may find useful:
> http://www.ngdc.noaa.gov/paleo/pubs/jones2001/jones2001.html
> http://www.ngdc.noaa.gov/paleo/recons.html
> http://www.ngdc.noaa.gov/paleo/pubs/briffa2001/briffa2001.html
> http://www.ngdc.noaa.gov/paleo/pubs/mann2003/mann2003.html
>
> (The last study use an entirely different method by taking temperature
> measurments from boreholes. That's how you do a independent study)
>
> > You did mention Spencer and Christy, and I will read
> > their work.
>
> S&C are analyzing data from meteorological satellites so that debate
> concerns a much more recent temperature record. What is amusing is that the
> same people who have critizised MBH for not beeing reproducible, despite
> other getting the same results, have uncritically accepted the results by
> S&C ...
That sounds like an exaggeration. I can think of at least one who has
"criticized" MBH for being unreplicable as published, yet has not done
as you say regarding S/C.
Very truly,
Steve Schulin
http://www.nuclear.com
> ... despite the fact that those haven't been reproduced either, and that in
Thomas Palm
news:steve.schulin-613...@comcast.dca.giganews.com:
> In article <Xns95204FF5E44B4T...@212.83.64.229>,
> Thomas Palm <Thoma...@chello.removethis.se> wrote:
>> S&C are analyzing data from meteorological satellites so that debate
>> concerns a much more recent temperature record. What is amusing is
>> that the same people who have critizised MBH for not beeing
>> reproducible, despite other getting the same results, have
>> uncritically accepted the results by S&C ...
>
> That sounds like an exaggeration. I can think of at least one who has
> "criticized" MBH for being unreplicable as published, yet has not done
> as you say regarding S/C.
Sorry, I didn't mean to imply that everyone complaining about MBH was that
kind of hypocrite, only that a large number of them were.
w...@bas.ac.uk
>You are trying to escape the fundamental of science; that science must
>be repeatable. If it isn't repeatable, it is not science. This is a
>pillar of the scientific edifice.
>As a point of fact, can you show us any paper which has directly
>replicated the results of MBH '98 ? If you cannot show us such a
>paper, how can you assure us it is correct ?
You have carefully skipped from repeat to replicate.
As was pointed out in the previous incarnation of this thread,
studies simply *replicating* other peoples work are rare (I can't
think of a single one published in climate science. Can you find any in
any other science?) for the obvious reason that exact replication is
not usually worthy of publication. If you want *repeatability* then
there is Jones, Briffa, Esper etc.
-W.
--
William M Connolley | w...@bas.ac.uk | http://www.antarctica.ac.uk/met/wmc/
Climate Modeller, British Antarctic Survey | Disclaimer: I speak for myself
I'm a .signature virus! copy me into your .signature file & help me spread!
w...@bas.ac.uk
>All "multiproxy" approaches share a common disadvantage -- the smearing
>of individual series' dating error. Relatively flat results are to be
>expected.
Which (if it were true) would be yet another reason for disbelieving
M&M's result.
w...@bas.ac.uk
>w...@bas.ac.uk wrote in message news:<40eb...@news.nwl.ac.uk>...
>> Per <perox...@ntlworld.com> wrote:
>> >...the M&M audit paper, which was reviewed by an expert
>> >peer reviewer and an editorial board,...
>>
>> We don't really know this. E&E's editorial/review policy isn't
>> clearly stated. Did you really mean a singular reviewer? Most journals
>> use at least 2.
>I don't know the process of E&E in any detail. I am assuming at least
>one referee; but it may be two or more.
>I will be grateful if you can shed any light.
As far as I know no-one outside of E&E knows anything about their
review process. Try reading the journal web page if you don't want
to know any more...
M&M say at http://www.uoguelph.ca/~rmckitri/research/trcqa.html
"Was your article peer-reviewed?
* Yes. Our article was read by numerous colleagues in Canada, the US,
Australia and Europe, including experts in mathematics and statistics,
geology, paleoclimatology, climatology and physics. It was refereed
for Environment and Energy by reviewers selected by the editor."
The first bit is weaselly - thats not peer review. The second bit does
appear to assert that reviewers in the plural ref'd the paper.
Per
> Most of M&M:s claims just showed their own incompetence because when
> others looked they did find the data although you had to look around a
> bit in the subdirectories.
This is the most bizarre conclusion ! MBH have now accepted that M&M
pointed out errors and omissions of a non-trivial nature in their
paper. They have put this in print.
There is a fundamental issue about scientific integrity here. When you
say "My dataset is available as a spreadsheet, or on this website at
position X", it is not acceptable to say, "oh well, the data isn't
where I said it is. I only used parts of the data I said I used. You
can find the data somewhere else, but I am not telling you where."
These are the errors of MBH, and MBH have accepted them for the errors
that they are.
MBH have accepted they were wrong. Why aren't you prepared to take
MBH's own word that they were wrong ?
per
Per
> No. You should speak when you have something meaningful to say.
> I guess you have no expectation of having anything meaningful in the near
> future.
> Or maybe I am tired of ( admitted) ignorant opinions being expresssed as
> facts.This is a science newsgroup. If you cannot contribute to the science,
> then just listen and learn.
> Now, the point is that the verification has been done and you have nothing
> to contribute.
> Or maybe you should continue on alt.dumbass.opiinion.whocares
I think your words speak all too eloquently about yourself.
per
Robert Grumbine
Bill Corden <Cor...@somwhere.com> wrote:
>"Robert Grumbine" <bo...@radix.net> wrote in message
>news:10eooqv...@corp.supernews.com...
>> The obsession some have with the 1998 paper is looking a lot like
>> the Young Earth Creationists who (try to) attack biology by saying
>> that Darwin got something wrong. They're generally wrong in that,
>> but, even so, so what -- he didn't lie, and did have a good idea.
>> Even if every page of the old work were shown to be wrong, by current
>> knowledge, biology would still be a perfectly decent science. We've
>> just learned more in the mean time. So it goes with MBH 1998, though a
>> more recent and less grand idea.
>>
>> We now have, not just MBH 1998, but ME 2000, YOU 2001, HIM 2000,
>> OTHER 1999, Anon et al. 2003, Ibid 2004, ...
>
>Thank you for the effort you put into the post. I sense that you are indeed
>earnist. I am a little annoyed at you because I tried to Google "ME 2000"
>then "YOU 2001" and did not get very satisfactory results<g>. You did
>mention Spencer and Christy, and I will read their work.
Read the other works which look at the MSU as well.
Others have mentioned, now, less whimsical and more accurate citations
to other of the climate reconstruction papers.
[much unmarked deletia by Corden, here and elsewhere]
Again, their work can be checked, as the term has meaning in science,
even if they didn't provide a single one of the data sets they used.
Conversely, someone could provide all the data sets they used, and
be highly incorrect about their conclusions as regards nature.
Ideally, I agree that all data, programs, codes, equipment, etc.
should be openly available to all researchers, at all times, for
unlimited duration. As a practical matter, nobody has ever found a
way to make that happen. I make my own attempts in that direction
for some things at my day job, and it is fairly difficult to go even
as far as I do. And I'm far from reaching that ideal. Nevertheless,
my work is useful.
Auditing is, simply, a terrible analogy or model for science.
It might even be worse than courtrooms, which are also a terrible
model.
The thing is, science is concerned with making testable statements
or descriptions of the natural world. In that case, unlike auditing,
there are no priviledged data sources. So it is here. The data that
were used in the 1998 paper had already existed and been used in other
scientific works. They had already been used in other papers, by
other researchers. The quality of those data, then, is a matter that
sits with the original researchers, not with the 1998 paper.
Further, the 1998 paper only used certain data sets. That's a given,
as there are always new data sets being constructed, and some of the
old ones aren't very widely known.
Years have passed now. The data sets used in 1998 have been improved
(most of them, most likely), and some (probably) have been found to not
be as good as thought in 1998 (calibration or chronology problems
discovered thanks to the fact that the data get used for a number of
_different_ problems by a number of _different_ researchers). With
the interest that the 1998 paper generated, a number of those
less-well-known data sources have been unearthed, relocated, readvertised,
etc.
Also sparked by the 1998 paper, a number of new data sets have been
constructed. Some, no doubt, were constructed by people who wanted to
prove the 1998 conclusions wrong (for instance, about the size and
scope of the 'little ice age' or 'medieval warm period'). If it
were an auditing matter, no such thing would be possible, nor of interest.
But it's science. If the interest is global climate, then anything
capable of being affected by climate in an understandable way, and which
can leave a record for an interesting period of time, is a potential
data source. So we _can_ go out and collect more and better data.
And folks do.
Something else we (at least try to) do in science is to use the
_best_ possible data, in the _best_ possible way.
Given those points, the business of 'auditing' a 6 year old paper
is just ludicrous, except for historical interest. We have more
and better data today than 6 years ago. Telling scientists that
they should _ignore_ the new data sets, the improved data sets,
and the data sets that weren't used 6 years ago is just silly.
You may choose to ignore all more recent science in the area
because an earlier paper in the field wasn't absolutely perfect.
If so, though, you should never use any electromagnetic wave
devices. Coulomb's law is fundamental to all radio transmission,
and Coulomb could not have derived his law from his experiment.
(Historical interest lead someone to try the experiment. It was
written up in Physics Today, iirc, a few years ago.)
The thing is, though, you don't have to trust the 1998 paper.
It has been superceded by several others. The reworking with more
and better data and different methods is how science attacks the
'trust' problem. In the Spencer and Christy case, the result of the
reworking has been consistently that they were neglecting important
effects, and not accurately including the ones they did. Now, if an
'audit' were done, one would (I'm confident) get their results if
you used their data and their methods. They've just been wrong --
as regards nature. Being wrong about what you're observing is far
more a concern to science.
Bill Corden
> More correctly "Trust but verify,and always cut the cards" Robert A.
> Heinlein.
>
> Now, the point is that the verification has been done and you have nothing
> to contribute.
>
> > Perhaps we should continue after you
> > have had some sleep or a little less coffee.
>
> Or maybe you should continue on alt.dumbass.opiinion.whocares
I am new to these groups, so I am unfamiliar with your bona fides. I
suspect you have an inflated image of your own knowledge and importance.
But much can be discerned about one's character from what they say and
write. I went back a month, to June 8th to see what you were saying to
people here before I arrived. This is the list of personal attacks or
anti-social comments I found. Perhaps they are not representative, but my
12 year old daughter would say "What is your DAMAGE?"
That is your delusion.
People die, dimshit.
Lie.
Nothing with blind ignorance such as your is ever settled rationally.
That is your delusion.
You are abysmally stupid or perhaps just mired in a fantasy world of
ideology.
An igorant and unsupported claim.
What a dimwit.
Try going to your doctor for an approved 'anti-psychotic' to help with your
delusional state.
dimwitted and unreasoned opposition
His 'background' is meaningless
snip of further appeals to igorance
You *are* a complete ignorant fool. No question.
Such stupidity is rather incestuous
what an ass you are by nature.
neighbors who 'sat on their asses'.
Anything to offer besides ignorance and logical fallacy?
--no personal attacks--
I would consider it 'evolution in action'.
--no personal attacks--
<more shit deleted>
You seem to be ignorant on SO many levels.
--no personal attacks--
he's a clueless dimshit, hoping to graduate to moron status.
you show a remarkable ingnorance
You are sooo clueless.
And such children need to be shown that their babblings are not appreciated
disturbing the grown up discussions
--no personal attacks--
Don't be any stupider than you have to be.
--no personal attacks--
--no personal attacks--
--no personal attacks--
I could probably get into Mensa on most days.
--no personal attacks--
I suspect that you have just earned your 'moron equivalency'.
--no personal attacks--
--no personal attacks--
Feeble. Not even a stinging rebuke.
--no personal attacks--
Do you just like to act the idiot?
Bullshit.
--no personal attacks--
I am not unaquainted with women.
You might actually have to listen to the party platforms!
44 responses, only 12 were civil. Some things I noticed: You tend to
become less anti-social as the day goes on. You must not be a morning
person. You are happiest when you are talking about trains, which you do
quite a lot.
That you for all your criticism of me. I will consider it and if necessary,
change my life. But I can't help but think that a small part of your anger
toward me has little to do with me and more to do with the fact that you
have enough time to post 44 messages and over 10 thousand words on a
Tuesday. Maybe getting a job would make you feel better about yourself.
Bill
Bill Corden
"Per" <perox...@ntlworld.com> wrote in message
news:bd7c1148.04070...@posting.google.com...
True enough and his speak more clearly than most.
Bill
Ian St. John
> "Ian St. John" <ist...@noemail.ca> wrote in message
> news:Gm5Hc.32342$WM5.1...@news20.bellglobal.com...
>> More correctly "Trust but verify,and always cut the cards" Robert A.
>> Heinlein.
>>
>> Now, the point is that the verification has been done and you have
>> nothing to contribute.
>>
>>> Perhaps we should continue after you
>>> have had some sleep or a little less coffee.
>>
>> Or maybe you should continue on alt.dumbass.opiinion.whocares
>
> I am new to these groups, so I am unfamiliar with your bona fides.
None required. All of the research is public knowledge. The only question is
whether you can read or understand it.
All I admit to is two university degrees from two separate universities and
a lifelong interest in science and technology but also philosophy, religion,
ethics, economics, and a number of other things. Science and technology
provide the 'horsepower' but the social dimension provides the direction.
> I
> suspect you have an inflated image of your own knowledge and
> importance.
Everyone does. However, I can honestly say that the facts show that I earned
some of that pride.
> But much can be discerned about one's character from what
> they say and write.
So I am intolerant of the postings of ignoramuses making blanket
declarations for which they have no evidence. By the way, you are making an
improper survey since you do not include the fact that I mostly post
rebuttals to fools. I leave the elucidation of science beyond the 'summary'
of a well read and educated person to those with the specific expertise.
Thus I tend to tell them they suck. Mostly because... well... they suck.
If you did an equal survey of my postings to the educated scientists here
like Josh Halpern, Eric Swanson, David Ball, Thomas Palm, etc you would find
no put downs at all. You get what you give.
Just wanted to give you a 'peer review' on your flawed methodology.
<snip>
> Some things I noticed: You tend to
> become less anti-social as the day goes on. You must not be a morning
> person.
Gee. One truth emerges... ;-)
> You are happiest when you are talking about trains, which
> you do quite a lot.
And one error.
>
> That you for all your criticism of me. I will consider it and if
> necessary, change my life.
Don't do it for me. Do it for you. You mignt audit what *you* have *stated*
and see how much is truly known or knowable and how much is you speaking ex
cathedra from your navel.
> But I can't help but think that a small
> part of your anger toward me
You are projecting. I have no anger to you. I just think that you should
think before speaking, not after. And speak of what you know, not what you
think. I.e. be ready to back it up with references to the source material.
> has little to do with me and more to do
> with the fact that you have enough time to post 44 messages and over
> 10 thousand words on a Tuesday. Maybe getting a job would make you
> feel better about yourself.
Self employed. My 'workday' spans 24 hours ( except when I am not posting,
reading, eating, or sleeping). I do need to 'get out more' but there are
some medical issues. Consider me to be 'making lemonade' with my life.
P.S. Again, you make many statements of fact for which you have NO clue.
This is the problem with ignorance. It does not recognise itself.
>
> Bill
Bill Corden
> And one error.
>
> > That you for all your criticism of me. I will consider it and if
> > necessary, change my life.
>
> Don't do it for me. Do it for you. You mignt audit what *you* have
*stated*
> and see how much is truly known or knowable and how much is you speaking
ex
> cathedra from your navel.
Audit of what I have said has now been done per your suggestion. I have
stated nothing regarding science. I have mostly asked questions. I have
wondered whether honesty and openness is an attribute of this kind of
science, but I have come with an open mind and a desire to learn. I have to
think you have confused me with someone else.
> > But I can't help but think that a small
> > part of your anger toward me
>
> You are projecting. I have no anger to you. I just think that you should
> think before speaking, not after. And speak of what you know, not what you
> think. I.e. be ready to back it up with references to the source material.
What have I said that sounds angry?
Questions and a search for knowledge require no references or source
material. I have suggested nothing. I don't know enough about it to have
an informed opinion. And my questions have been about process, not the
science. I can understand process.
> > has little to do with me and more to do
> > with the fact that you have enough time to post 44 messages and over
> > 10 thousand words on a Tuesday. Maybe getting a job would make you
> > feel better about yourself.
>
> Self employed. My 'workday' spans 24 hours ( except when I am not posting,
> reading, eating, or sleeping). I do need to 'get out more' but there are
> some medical issues. Consider me to be 'making lemonade' with my life.
I am self employed as well. I am sorry about your medical issues.
But let me gently say something my grandfather use to say: It doesn't cost
any more to be nice.
> P.S. Again, you make many statements of fact for which you have NO clue.
> This is the problem with ignorance. It does not recognise itself.
I don't see them. I don't don't know what facts you are referring to.
Could you provide a single (let alone many) statement of fact for which I
have no clue? (by the way, statements are the sentences that don't end in a
question mark.)
Nothing but love and fellowship to you my friend.
Bill
Ian St. John
> "Ian St. John" <ist...@noemail.ca> wrote in message
> news:GifHc.34346$WM5.1...@news20.bellglobal.com...
>> And one error.
>>
>>> That you for all your criticism of me. I will consider it and if
>>> necessary, change my life.
>>
>> Don't do it for me. Do it for you. You mignt audit what *you* have
>> *stated* and see how much is truly known or knowable and how much is
>> you speaking ex cathedra from your navel.
>
> Audit of what I have said has now been done per your suggestion. I
> have stated nothing regarding science.
Nothing with regard to the actual science but lots of opinions on what the
science should say.
> I have mostly asked questions.
Auditing your posts I find few questions and lots of statement or assertions
such as this.
> I have wondered whether honesty and openness is an
> attribute of this kind of science, but I have come with an open mind
> and a desire to learn. I have to think you have confused me with
> someone else.
No. I only get confused about other relatives birthdays.
>
>>> But I can't help but think that a small
>>> part of your anger toward me
>>
>> You are projecting. I have no anger to you. I just think that you
>> should think before speaking, not after. And speak of what you know,
>> not what you think. I.e. be ready to back it up with references to
>> the source material.
>
> What have I said that sounds angry?
non-sequitor.
>
> Questions and a search for knowledge require no references or source
> material.
Asking questions doesn't but then you do not ask questions so much as you
base questions on some assumption. This is the 'are you still beating your
wife' technique that asks a 'questions' but still makes a statement ( that
you used to beat your wife ).
> I have suggested nothing.
[Bill/George Burt?]
"I would like to point out some general criticisms"
"In case you have not noticed"
(Must) "You treat non-scientist like inconsequential children"
"Would acknowledging that (M&Ms: presumed by you in the guise of a
'question' which you then assume true.) contribution kill you?"
etc. etc. All troll techniques of long standing along with the non-sequitors
and 'you are angry' pop psychology dismissal.
> I don't know enough about it to
> have an informed opinion. And my questions have been about process,
> not the science. I can understand process.
Not so far. Unless you mean the 'process' of 'trolling'.
>
>>> has little to do with me and more to do
>>> with the fact that you have enough time to post 44 messages and over
>>> 10 thousand words on a Tuesday. Maybe getting a job would make you
>>> feel better about yourself.
>>
>> Self employed. My 'workday' spans 24 hours ( except when I am not
>> posting, reading, eating, or sleeping). I do need to 'get out more'
>> but there are some medical issues. Consider me to be 'making
>> lemonade' with my life.
>
> I am self employed as well. I am sorry about your medical issues.
>
> But let me gently say something my grandfather use to say: It
> doesn't cost any more to be nice.
And it doesn't costs any less. Moreover it tends to attract flies and con
men. Better to be as honest as you can while hiding the worst of it to let
them know that you aren't a fool..
>
>> P.S. Again, you make many statements of fact for which you have NO
>> clue. This is the problem with ignorance. It does not recognise
>> itself.
>
> I don't see them.
See! ROTFLOL!
> I don't don't know what facts you are referring to.
> Could you provide a single (let alone many) statement of fact for
> which I have no clue? (by the way, statements are the sentences that
> don't end in a question mark.)
>
> Nothing but love and fellowship to you my friend.
Try it on some rich lonely widow lady..
>
> Bill
Per
> You have carefully skipped from repeat to replicate.
well, it wan't careful, and replicate is defined as a synonym of
repeat in my dictionary.
> As was pointed out in the previous incarnation of this thread,
> studies simply *replicating* other peoples work are rare (I can't
> think of a single one published in climate science. Can you find any in
> any other science?) for the obvious reason that exact replication is
> not usually worthy of publication. If you want *repeatability* then
> there is Jones, Briffa, Esper etc.
Just one example, but there are many.
S. F. Arnold, D. M. Klotz, B. M. Collins, P. M. Vonier, L. J.
Guillette Jr., J. A. McLachlan, Science 272 1489 (1996).
this paper was incredibly influential, and spawned legislation on
environmental estrogens in the US. Groups in the UK and US published
papers in Nature and Science to say that the results could not be
repeated. Mclachlan published a retraction, and a report was published
characterising the paper as fraud.
I am not sure that you aren't being a little bit disingenuous when you
suggest that Jones, Briffa, Esper, etc have repeated the MBH result;
perhaps you can provide the references which *repeat* MBH, and we can
then form a view ?
yours
per
w...@bas.ac.uk
>w...@bas.ac.uk wrote in message news:<40ed...@news.nwl.ac.uk>...
>> You have carefully skipped from repeat to replicate.
>well, it wan't careful, and replicate is defined as a synonym of
>repeat in my dictionary.
Dictionaries describe natural language, you have to be careful about the
scientific meanings though. Replicate and repeat are distinct.
>> As was pointed out in the previous incarnation of this thread,
>> studies simply *replicating* other peoples work are rare (I can't
>> think of a single one published in climate science. Can you find any in
>> any other science?) for the obvious reason that exact replication is
>> not usually worthy of publication. If you want *repeatability* then
>> there is Jones, Briffa, Esper etc.
>Just one example, but there are many.
>S. F. Arnold, D. M. Klotz, B. M. Collins, P. M. Vonier, L. J.
>Guillette Jr., J. A. McLachlan, Science 272 1489 (1996).
>this paper was incredibly influential, and spawned legislation on
>environmental estrogens in the US. Groups in the UK and US published
>papers in Nature and Science to say that the results could not be
>repeated. Mclachlan published a retraction, and a report was published
>characterising the paper as fraud.
No no, they did *not* replicate the work: if they had, they would have got
the same result! Try again.
>I am not sure that you aren't being a little bit disingenuous when you
>suggest that Jones, Briffa, Esper, etc have repeated the MBH result;
>perhaps you can provide the references which *repeat* MBH, and we can
>then form a view ?
You need to ponder the differences between the two R words.
Bill Corden
[Y]ou make many statements of fact for which you have NO clue.
Ian flounders, tries misdirection, muddies the water a bit, but does he site
a single example? See for yourself. Will he admit he is wrong when he is
so obviously wrong! Not in a million years. His ego is too frail for it.
Ian, I dub thee wannabeScienceBoy2000.
> Nothing with regard to the actual science but lots of opinions on what the
> science should say.
>
> > I have mostly asked questions.
>
> Auditing your posts I find few questions and lots of statement or
assertions
> such as this.
"I have mostly asked questions" is an example of statements or assertions?
This makes no sense.
> > I have wondered whether honesty and openness is an
> > attribute of this kind of science, but I have come with an open mind
> > and a desire to learn. I have to think you have confused me with
> > someone else.
>
> No. I only get confused about other relatives birthdays.
I was giving you an out. You didn't take it. So now, if you were
intellectually honest, you would back up the claims you made about me,
namely:
wannabeScienceBoy2000 claim: [Y]ou make many statements of fact for which
you have NO clue.
> >>> But I can't help but think that a small
> >>> part of your anger toward me
> >>
> >> You are projecting. I have no anger to you. I just think that you
> >> should think before speaking, not after. And speak of what you know,
> >> not what you think. I.e. be ready to back it up with references to
> >> the source material.
> >
> > What have I said that sounds angry?
>
> non-sequitor.
You're original post to me was hostile in tone (that is the way I interpret
insults), which often comes from the emotion called anger. I then point out
34 mostly nasty insults you posted 1 month ago. This is further evidence of
anger or perhaps hate. (How should I know) You then counter by saying I am
projecting, meaning that I am the one who must be angry. I say, well I am
not angry. You say "non-sequitor."
Have you ever admitted anywhere in the newsgroup that you were wrong? I
doubt it. You strike me as the kind whose ego could bear the shame. I, on
the other hand am wrong all the time and I admit when I am.
> > Questions and a search for knowledge require no references or source
> > material.
>
> Asking questions doesn't but then you do not ask questions so much as you
> base questions on some assumption. This is the 'are you still beating your
> wife' technique that asks a 'questions' but still makes a statement ( that
> you used to beat your wife ).
So, I am making statements because the questions I ask are really statements
phrased as questions. You make me sound very clever. Perhaps you could
take what I say at face value and we could move on from there.
> > I have suggested nothing.
wannabeScienceBoy2000 claim: [Y]ou make many statements of fact for which
you have NO clue.
Taken out of context. You are simply posturing because you can't make your
point with what I said. The above statement notwithstanding, I have
suggested that you are either angry or mean.
Result: wannabeScienceBoy2000 fails to provide a single statement of fact
for which I have no clue.
> "I would like to point out some general criticisms"
Ahah! your point is made! You win! You win! Anyone who would make this
statement...where is the rest of the statement and what was I saying?
Woops! Result: wannabeScienceBoy2000 fails to provide a single statement of
fact for which I have no clue.
> "In case you have not noticed"
Ahah! This proves it! No reasonable person would utter these words unless
clueless fact fever had ravaged his body.
Woops! Result: wannabeScienceBoy2000 fails to provide a single statement of
fact for which I have no clue.
> (Must) "You treat non-scientist like inconsequential children"
First, thank you for altering my sentence. It is often helpful when you are
trying to villify your opponent. This must be one of the pesky questions
that is really a statement. The truth is, as anyone could tell from reading
it in context, this was not a statement but a plea for indulgence.
wannabeScienceBoy2000 fails this appeal for indulgence quite dramatically.
Woops! Result: wannabeScienceBoy2000 fails to provide a single statement of
fact for which I have no clue.
> "Would acknowledging that (M&Ms: presumed by you in the guise of a
> 'question' which you then assume true.) contribution kill you?"
A) You changed my sentence
B) I was trying to bridge a gap between the two sides here.
C) It was a question but a comment could reasonably be taken from it, so I
will deal with it as if it were a statement of FACT.
"acknowledging M&M's contribution"
Fact or fact for which I have no clue. Well, in this case, it is a fact.
My point was based upon a reading of the Nature Corrigendum which said
expressely M&M was at least partly responsible for bringing the problems to
their attention. Why else would they be mentioned at all. It is remotely
possible that I am wrong about this (which I don't think I am) but even if I
were, it is not clueless. That is a fact (if true) that the person I was
responding to seemed unwilling to acknowledge. Clueless indeed. I am happy
to provide you the quote where Mann and company refer to M&M.
AGAIN, Result: wannabeScienceBoy2000 fails to provide a single statement of
fact for which I have no clue.
So. The challenge remains. Have I or have I not made "many" (or even
"any") statement of fact for which I have no clue as you have claimed?
I gave you an out (maybe you have me confused with someone else), you were
too bullheaded to take it. Now. Either you can provide an example of a
statement of FACT for which I have no clue or you cannot. Which is it
wannabeScienceBoy2000? You said there were "many" I want one.
> etc. etc. All troll techniques of long standing along with the
non-sequitors
> and 'you are angry' pop psychology dismissal.
Yeah, _I_ am the troll.
> > I don't know enough about it to
> > have an informed opinion. And my questions have been about process,
> > not the science. I can understand process.
>
> Not so far. Unless you mean the 'process' of 'trolling'.
When are losing an argument, call names and change subjects. Typical
wannabeScienceBoy2000.
Bill
Robert Grumbine
Bill Corden <Cor...@somwhere.com> wrote:
>Ian Claim That he can't back up:
> [Y]ou make many statements of fact for which you have NO clue.
>
>Ian flounders, tries misdirection, muddies the water a bit, but does he site
>a single example? See for yourself. Will he admit he is wrong when he is
>so obviously wrong! Not in a million years. His ego is too frail for it.
>Ian, I dub thee wannabeScienceBoy2000.
rest deleted.
A gentle suggestion: If you're interested in the science, which
is what you said, then following others down a trail that isn't science
is probably not going to get you where you're trying to go. There have
been other, substantive, posts in this thread.