At 2:23 PM -0500 30/11/09, Charles Bennett wrote:
>The "climategate" scandal just gets better and better.
>
>"Programmers all over the world have begun wading through the code and
>they have been stunned by how bad it is. It's quite clearly amateurish
>and nothing but an accumulation of seat-of-the-pants hacks and patches."
well, I don't know about amateurish, but of course it is a
bunch of hacks and patches - most of the code they have been talking
about is run once, to produce one result, a single graph or dataset
or whatever.
Just like if you cook a meal, even if it tastes great it
isn't the same as the way a chef designs a menu knowing he will have
to push out hundreds of plates a night.
><http://chicagoboyz.net/archives/10399.html#more-10399>
>
>The real kicker is in the comments of the second programmer. He
>COULD NOT reproduce the original results on the original data set.
>(WTF? no source code control?)
>
>so, finally.. "Leading British scientists at the University of East
>Anglia, who were accused of manipulating climate change data - dubbed
>Climategate - have agreed to publish their figures in full."
I love the way the argument here goes we should ignore the
consensus from all the different scientific groups around the globe,
because one group (that is notoriously the target of political
pressure and attacks from deniers) turns out to be both not
universally competent, and not universally completely honest and open.
Because you know what is funny thing here? Even with access
to all this illegally obtained private conversation, etc - the
deniers still haven't been able to find evidence of anything actual
published major results that were significantly different.
>"http://www.timesonline.co.uk/tol/news/environment/article6936328.ece";
>
>So not only is the original data NOT going to be released, just the
>"massaged" data, they actually can't even get their hands on the
>original.
>
>"In a statement on its website, the CRU said: "We do not hold the
>original raw data but only the value-added (quality controlled and
>homogenised) data."
>Scientifically, We must say that ALL of the conclusions based in this
>data are therefore invalid.
Nope, what they have remains better than not having any data
at all, and while I'd certainly be very dubious about it if there was
a group that had data that contradicted everyone elses that they
couldn't reproduce, that isn't the case. When their data actually
says much the same as everyone else? Well, you just rely on the other
data a bit more if you have questions.
And you know what? All the other data sources do say the same
damn thing. It is not like every lab in the world was just using
derived data from some other guy. We have plenty of separate
independent data.
The only reason to completely avoid using such data would be
if you basically suspect the other lab of either fraud or some basic
error in data processing. Rationally, we have no reason to suspect
either - the data more or less agrees with other, independent,
source. But this is science, not a court case - you use what you
have, and if you have some issues that mean you don't have full
confidence in it, you can weigh it up against other data and build up
a bigger picture.
But, of course, the wingnuts DO think there was fraud or
such, because they have already decided that it must be wrong, so for
them, the loss of the raw data is a very big deal indeed. Its not.
The loss of the raw data is annoying, and sloppy, and dumb. But not
as big a deal as they think.
> No peer review possible, no data
>available and the comments of the programmer says that the original
>conclusions cannot be duplicated..
Seriously ---
-- there is more than one climate change research group in the world,
more than one set of data, more than one post-doc munging data in a
back room.
-- they all say the same thing. And despite the way their privacy has
been violated by a criminal act and all that, there was no conspiracy
revealed. No global coordinationg between labs to keep the facade up.
Just scientists communicating among themselves trying to make their
publications look good even when their data was not as good as they
would like it to be.
--- yeah, these guys were sometimes sloppy, and sometimes overzealous
in trying to make their work look good. But their published work has
been put under far more scrutiny than any normal science team, and
help up despite it.
>I'm arguing that none of their conclusions can be trusted, and by
>extension any IPCC report that uses their data must be assumed to be
>flawed.
Oh, cease this overdramatic pearl clutching. One lab, one
particular data set among dozens.
>
>Which brings me to this email from the CRU (note the subject line)
>
>"From: Phil Jones
>To: "Michael E. Mann"
>Subject: IPCC & FOI
>Date: Thu May 29 11:04:11 2008
>Mike,
>Can you delete any emails you may have had with Keith re AR4?
>Keith will do likewise. He's not in at the moment - minor family crisis.
>Can you also email Gene and get him to do the same? I don't have his
>new email address.
>We will be getting Caspar to do likewise.
>I see that CA claim they discovered the 1945 problem in the Nature
>paper!!
>Cheers
>Phil
>Prof. Phil Jones"
>
>and
>
>"From Phil Jones:
>"If FOIA does ever get used by anyone, there is also IPR to consider
>as well. Data is covered by all the agreements we sign with people, so
>I will be hiding behind them."
>
>When you ask someone to DELETE emails related to a Freedom Of
>Information request you are likely hiding something or committing fraud.
-- yeah, these guys are defensive, and paranoid, and
sometimes a bit mean and bitchy. Its Manns lab, these guys have been
the victim of a years long campaign of character assassination and
such, a lot of from nasty political operators. They are human beings.
Yeah, they are hiding - they are hiding from people who treat them
like they are defendants in a trial every day of their lives, and
they start to act like it.
That the climate change science has become over politicised
is not news. Seriously, yes, this doesn't look good - but do you
seriously think you wouldn't find stuff as bad or worse if you got
the private email files from pond scum on the denier side like, say,
Steven Milloy?
>When you say that you will "hide behind" agreements to avoid releasing
>data that supposedly proves your claim of AGW you confirm that you
>have no desire to be open and honest with data that might have world
>wide significance.
They have no desire to be open and honest with people like
Steven McIntyre, who will turn any small error they might find in
stats processing etc into a dozen press releases, personal
accusations, etc.
Yeah, climate science is politicised, and people in it tend
to respond to their critics as if they are political enemies rather
than fellow scientists just interested in some peer review. I wonder
why?
>
>Of course this all begs the question, why the fuck does anyone need to
>use the FOI to get the data? If it's good then you should demand to
>publish it where EVERYONE can have a look.
>
>If AGW is a fact, then refusing to share the data and "allowing" the
>controversy to continue while you "hide" it from FOI requests, your
>actions will cost human lives.
The data in question was someone elses data, commercial data,
they didn't have any right to hold onto and reuse it, and explicitly
not republish it, is my understanding.
The raw data paid for by public money may not be held by CRU,
but is available from other, public, sources.
In other words, its a beatup.
>No source code control for the code that takes that data and
>manipulates it?
>
>Are you fucking kidding me?
I've worked with environmental engineers (the ones I worked
with studied lakes mostly, not global climate stuff), and I certainly
found that you had a lot of engineers hacking on fortran code, many
of home were unaware of all sorts of basic computer science results
and tools. Of course, some did know what they were doing, but grab
any individual post-doc or whatever, and there was a pretty
reasonable chance they would be quite clueless.
>If this is indeed "change the world" importance and not just out right
>fraud, then why the hell were the data not backed up?
I know, they threw this stuff away in the 1980s, not knowing
that it would be subject of global debate in the 21st century - fools!
And you know what else really pisses me off? the way the BBC
threw out their tapes of a lot of Hartnell and Troughton Doctor Who
stories in the 1970s, not realising that hordes of 21st century
Doctor Who fans are really wanting to buy them on DVD!
>Oh but it gets EVEN better..
>
>Look at this bit of THEIR code <http://www.qando.net/?p=5936> (look at
>the graphic)
>
>"This, people, is blatant data-cooking, with no pretense otherwise. It
>flattens a period of warm temperatures in the 1940s 1930s - see those
>negative coefficients? Then, later on, it applies a positive
>multiplier so you get a nice dramatic hockey stick at the end of the
>century"
Tim Lambert has, as he so often has, already done a pretty
solid reply to this one
http://scienceblogs.com/deltoid/2009/12/quote_mining_code.php
In short -
- that particular bit of code he is describing sets a variable... and
then the code that uses it is commented out. So it does nothing.
- the code there is used to create a graph for a publication. The
corrected data set that has Erics panties in a knot does not appear
on the graph that was published. Or, as far as we can tell, anywhere
else.
- yeah, is it a highly artificial correction of the data they were
discussing? It sure is. Because it is in a paper discussing a
particular type of data that is known to have previously correlated
with temperature well, but doesn't any more, and how it would need an
artificial correction to be useful. In other words, the context
basically turns it from 'OMG conspiracy' to scientists actually
openly discussing and acknowledging known issues with particular
sorts of data.
- Eric Raymon is a flaming wingnut, and anyone involved in the
computer industry should know that by know. But I've actually met
him, so I can confirm he is a raving wingnut in person, not just on
the internet. He actually, literally, thinks climate change is a
Soviet plot, for starters.
>That Fudge Factor, gentlemen might literally be the smoking gun that
>people will use to claim climate change fraud.
Except the gun wasn't fired. And it was to produce one small
graph, so its a deringer at best. And it was in paper about, not 'the
Hockey stick', but known issues with dendro-chronology, so it wasn't
exactly aimed where you think it was either. And it turns out it
wasn't a gun, but a drill that kind of looked like a gun in bad light.
So, given all the fuss over that one little graph, I think
that you have -
a) been behaving like a hysterical alarmist drama queen over
something that is MUCH less important than you think it is
and
b) been caught taking the poltical ravings of Eric Raymond seriously
in public. I think an appropriate punishment would be to go read
Erics Sex Tips for Geeks http://www.catb.org/~esr/writings/sextips/
as after reading that, you'll probably find it pretty hard to take
him seriously ever again. Or maybe read some of the 'Everybody Loves
Eric Raymond' comic that I used to like, but hardly gets updated
these days.
http://geekz.co.uk/lovesraymond/archives
Cheers
David
_______________________________________________
OSX-Nutters mailing list | [email protected]
http://lists.tit-wank.com/mailman/listinfo/osx-nutters
List hosted at http://cat5.org/
