My Own Private East Anglia
Feb. 13th, 2010 11:50 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
snousleRemember when I said that I wouldn't be making any more posts about climate change? Well, I lied. Here's another.
From the perspective of this po' lil analyst, the unprecedented attack on science by the right-wing noise machine has been disturbing but not personally consequential. I've spent a lot of time soul-searching over the past few months and have seriously considered the possibility of putting real effort into the defense of science - not just posting on blogs but getting involved politically. But short of that, there is no point in getting worked up about something unless I'm going to put real energy into it, and my decision had been to let it go.
Then, night before last, I woke up wide-eyed with one of those "oh, no" moments. First thing in the morning, I checked it out, and "oh no" turned into "oh HELL no". And rather suddenly, the whole "climategate scandal" got a whole lot closer.
Some background: During the summers of 89-90, as an undergraduate, I did some work for the Center for Accelerator Mass Spectrometry at Livermore. John had just started working there himself, and would continue to do so for 15 years. It wasn't nepotism because, that first year, I wasn't paid. The second year, I was, but the management agreed that it still wasn't nepotism since we weren't legally married. Yay for discrimination.
The instrument they run there is used to accurately measure very, very, VERY small amounts of 14C in tiny samples, like one might obtain from a gas bubble in an ice core. At the time, it was being used mostly in archaeology - for artifacts less than 50,000 years old, this technique made it possible to get an accurate radiocarbon date for an organic object using only a speck of material. But there was also earth science research going on, and in particular we had some postdocs that were trying to figure out where all the carbon from fossil fuels was ending up. I was aware of global warming at the time but it was one of many issues and didn't stick out as being especially important.
Anyway, during that time I wrote the first and second versions of two pieces of software. The first one, called Hector, chatted with the detector end of the accelerator and gathered data from these measurements, classifying ~10,000 particle events per second in real time as either 14C or not-14C. (Later work addressed other elements as well.) The second one took these results and normalized them to primary and secondary standards in order to correct for drift in the "tuning" of the accelerator. Both of these applications are still in use today. They have been rewritten in other languages and extended to do more things, but their architecture remains essentially the same.
Although I can't produce a specific example right now, I'm pretty sure that if you followed up on all the references in the IPCC report and found out where their raw data came from, you'd find at least a few - possibly quite a few - results that depended in some way on these measurements. So, somewhere in that report is data that was shaped by my own hand.
Both these applications were quite successful and have been hammered on for so hard and so long that I have complete confidence that they are doing the Right Thing. There's just one little problem with the normalization code.
Its name.
I called it Fudger. Which is analyst slang for "inappropriately manipulating your data". The reason I woke up in a tizzy the other night is because I realized that what I had thought of as an incidental and probably ephemeral piece of code was very likely still in use. On checking up on it, only was I right about that - but it's now been been publicly released. Not with my name on it anymore - later analysts did about 10 times as much work on it as I ever did - but it's still my baby.
I asked John about it and he said yes, the normalization method we used proved quite superior to that used by competing facilities, so quite a few researchers became interested in it and use it at their own institutions. And he confirmed that the name was my idea.
The whole stolen-email flap at the Climate Research Unit is about essentially this - analysts using insider lingo in a way that looks really bad to outsiders. The analysts there are getting death threats over this.
So yeah, it actually is personal. And I didn't even remember this for a couple of months.
Whatever made that name seem like a good idea at the time? Well, I guess we were all intoxicated by the smell of freshly poured concrete and a big expensive shiny machine to play with. I was a cocky undergrad quite delighted to be involved in something so interesting at such a young age. If I had to explain it, I would say that the name was an ironic reaction to the extreme scrutiny the data were subject to. At facilities like this, where one machine or technology supports dozens of research programs, the clients are always very interested in the details of data collection and analysis, often asking for raw, unprocessed data, descriptions of the algorithms used in processing it, and sometimes source code. The number of critical eyes on the process is somewhat daunting. Everyone knows this, so of course calling a program a "fudger" is hysterically funny. Kind of a pressure release valve. What happened, though, is that it went from an inside joke to standard terminology. Normalizing AMS data is now known, conversationally, as "fudging" it. Of course nobody puts this word into scientific papers, but through long exposure, it seemed so natural that they thought nothing of putting it on the Web for other researchers.
So much turns on the presumption of good faith. And so much changes when that's gone. I had never imagined that this code would land in a context where "fudging" was anything but a joke.
It is worth mentioning here that in 20 years of analysis work, I have never been asked to falsify anything. I used to joke that my job title ought to be "Slayer of Hypotheses", because my job was to tell scientists if their results met appropriate standards of statistical significance. And as analyst, when I say no, it means no - the paper does not get published.
You'd think this would have put me under a lot of pressure. It did not. My judgment was never questioned, because despite the stakes, the science transcended any "political" differences. Of course, I had to justify why I would say yes or no to any particular claim, along with answering a whole lot of pointed questions, but these explanations were solid and objective enough that there was no reasonable basis for disagreeing with them. The only time there was any tension or argument at all was in an analysis of the genetic basis of cell migration speeds for a paper now published at PLoS: In Vitro Human Keratinocyte Migration Rates Are Associated with SNPs in the KRT1 Interval. I felt that the results were very marginal and came rather close to asking to be removed from the author list. But we had a (very) long discussion about the role of this data and how it should be communicated, and came to an agreement that I was satisfied with.
As you might imagine, I took these questions with complete seriousness, and developed a great deal of respect for the integrity of scientific research. I can't even begin to tell you how long the path is from experiment to publication, and how severely the results are critiqued at every step along the way. So to witness the arrogance and superficiality of the current attacks on the credibility of this research, where smug intellectual poseurs prance around saying "It's all fraud! They're all crooks! And stupid too!"... well, how can I not take this personally.
And then, to realize that I have, through the smallest, stupidest, most juvenile possible thing, played right into their hands with something from so long ago... well, it leaves me kind of speechless.
It doesn't really matter. I very much doubt that the climate-change denialists will go to the trouble of "connecting dots" as far down the chain as my unfortunately-named creation. But you can understand why this whole thing gets me on edge. I'm part of it, and in a very real way, when the media talks about "climategate", they're talking about me.
The attack on science is not going to stop with climate change. And it may not stop with science, either. The Bush presidency may have been bad, but it never sent chills down my spine in the way this does.
From the perspective of this po' lil analyst, the unprecedented attack on science by the right-wing noise machine has been disturbing but not personally consequential. I've spent a lot of time soul-searching over the past few months and have seriously considered the possibility of putting real effort into the defense of science - not just posting on blogs but getting involved politically. But short of that, there is no point in getting worked up about something unless I'm going to put real energy into it, and my decision had been to let it go.
Then, night before last, I woke up wide-eyed with one of those "oh, no" moments. First thing in the morning, I checked it out, and "oh no" turned into "oh HELL no". And rather suddenly, the whole "climategate scandal" got a whole lot closer.
Some background: During the summers of 89-90, as an undergraduate, I did some work for the Center for Accelerator Mass Spectrometry at Livermore. John had just started working there himself, and would continue to do so for 15 years. It wasn't nepotism because, that first year, I wasn't paid. The second year, I was, but the management agreed that it still wasn't nepotism since we weren't legally married. Yay for discrimination.
The instrument they run there is used to accurately measure very, very, VERY small amounts of 14C in tiny samples, like one might obtain from a gas bubble in an ice core. At the time, it was being used mostly in archaeology - for artifacts less than 50,000 years old, this technique made it possible to get an accurate radiocarbon date for an organic object using only a speck of material. But there was also earth science research going on, and in particular we had some postdocs that were trying to figure out where all the carbon from fossil fuels was ending up. I was aware of global warming at the time but it was one of many issues and didn't stick out as being especially important.
Anyway, during that time I wrote the first and second versions of two pieces of software. The first one, called Hector, chatted with the detector end of the accelerator and gathered data from these measurements, classifying ~10,000 particle events per second in real time as either 14C or not-14C. (Later work addressed other elements as well.) The second one took these results and normalized them to primary and secondary standards in order to correct for drift in the "tuning" of the accelerator. Both of these applications are still in use today. They have been rewritten in other languages and extended to do more things, but their architecture remains essentially the same.
Although I can't produce a specific example right now, I'm pretty sure that if you followed up on all the references in the IPCC report and found out where their raw data came from, you'd find at least a few - possibly quite a few - results that depended in some way on these measurements. So, somewhere in that report is data that was shaped by my own hand.
Both these applications were quite successful and have been hammered on for so hard and so long that I have complete confidence that they are doing the Right Thing. There's just one little problem with the normalization code.
Its name.
I called it Fudger. Which is analyst slang for "inappropriately manipulating your data". The reason I woke up in a tizzy the other night is because I realized that what I had thought of as an incidental and probably ephemeral piece of code was very likely still in use. On checking up on it, only was I right about that - but it's now been been publicly released. Not with my name on it anymore - later analysts did about 10 times as much work on it as I ever did - but it's still my baby.
I asked John about it and he said yes, the normalization method we used proved quite superior to that used by competing facilities, so quite a few researchers became interested in it and use it at their own institutions. And he confirmed that the name was my idea.
The whole stolen-email flap at the Climate Research Unit is about essentially this - analysts using insider lingo in a way that looks really bad to outsiders. The analysts there are getting death threats over this.
So yeah, it actually is personal. And I didn't even remember this for a couple of months.
Whatever made that name seem like a good idea at the time? Well, I guess we were all intoxicated by the smell of freshly poured concrete and a big expensive shiny machine to play with. I was a cocky undergrad quite delighted to be involved in something so interesting at such a young age. If I had to explain it, I would say that the name was an ironic reaction to the extreme scrutiny the data were subject to. At facilities like this, where one machine or technology supports dozens of research programs, the clients are always very interested in the details of data collection and analysis, often asking for raw, unprocessed data, descriptions of the algorithms used in processing it, and sometimes source code. The number of critical eyes on the process is somewhat daunting. Everyone knows this, so of course calling a program a "fudger" is hysterically funny. Kind of a pressure release valve. What happened, though, is that it went from an inside joke to standard terminology. Normalizing AMS data is now known, conversationally, as "fudging" it. Of course nobody puts this word into scientific papers, but through long exposure, it seemed so natural that they thought nothing of putting it on the Web for other researchers.
So much turns on the presumption of good faith. And so much changes when that's gone. I had never imagined that this code would land in a context where "fudging" was anything but a joke.
It is worth mentioning here that in 20 years of analysis work, I have never been asked to falsify anything. I used to joke that my job title ought to be "Slayer of Hypotheses", because my job was to tell scientists if their results met appropriate standards of statistical significance. And as analyst, when I say no, it means no - the paper does not get published.
You'd think this would have put me under a lot of pressure. It did not. My judgment was never questioned, because despite the stakes, the science transcended any "political" differences. Of course, I had to justify why I would say yes or no to any particular claim, along with answering a whole lot of pointed questions, but these explanations were solid and objective enough that there was no reasonable basis for disagreeing with them. The only time there was any tension or argument at all was in an analysis of the genetic basis of cell migration speeds for a paper now published at PLoS: In Vitro Human Keratinocyte Migration Rates Are Associated with SNPs in the KRT1 Interval. I felt that the results were very marginal and came rather close to asking to be removed from the author list. But we had a (very) long discussion about the role of this data and how it should be communicated, and came to an agreement that I was satisfied with.
As you might imagine, I took these questions with complete seriousness, and developed a great deal of respect for the integrity of scientific research. I can't even begin to tell you how long the path is from experiment to publication, and how severely the results are critiqued at every step along the way. So to witness the arrogance and superficiality of the current attacks on the credibility of this research, where smug intellectual poseurs prance around saying "It's all fraud! They're all crooks! And stupid too!"... well, how can I not take this personally.
And then, to realize that I have, through the smallest, stupidest, most juvenile possible thing, played right into their hands with something from so long ago... well, it leaves me kind of speechless.
It doesn't really matter. I very much doubt that the climate-change denialists will go to the trouble of "connecting dots" as far down the chain as my unfortunately-named creation. But you can understand why this whole thing gets me on edge. I'm part of it, and in a very real way, when the media talks about "climategate", they're talking about me.
The attack on science is not going to stop with climate change. And it may not stop with science, either. The Bush presidency may have been bad, but it never sent chills down my spine in the way this does.
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
Think of it as evolution in action
Date: 2010-02-13 09:47 pm (UTC)As for climate change if you Google it there is this great video loop from space of the reduction of arctic sea ice over the past 30 years. If that doesn't convince people give them a metal fork and tell them the toast is stuck.
BestRegards,
Pete
Re: Think of it as evolution in action
Date: 2010-02-13 11:16 pm (UTC)http://svs.gsfc.nasa.gov/vis/a000000/a003400/a003464/index.html
I wonder what the values are since '07. A cursory search has not revealed the data, it must have been censored by the international enviro-marxist conspiracy.
If that doesn't convince people give them a metal fork and tell them the toast is stuck.
I totally need a LOLcat image illustrating that concept!
Re: Think of it as evolution in action
Date: 2010-02-14 05:46 am (UTC)Your post just tells me more and more how this nation is in a decline. I makes me fearful that the sci-fi short story "The Marching Morons" is prophetic.
http://en.wikipedia.org/wiki/The_Marching_Morons
Same kind of sadness I felt when everyone started wearing bluetooth headsets. At first I thought it looked soo cool, so 21st century. Then I was verbally raped by a constant barrage of one half of the most stupid and banal conversations. So this is the future, morons with instant portable communication with other morons.
At least as a planet there are more people getting science/engineering degrees than ever before. More kids in China getting engineering Bachelor degrees than all the Bachelor degrees given in the USA.
Sorry normally I'm an optimist, I'll pull out I think it was the snow storm last week that has me in a funk.
BestRegards,
Pete
no subject
Date: 2010-02-14 03:50 am (UTC)2) Rarely say "never" or "last".
3) This is what I found for "cat fork toaster" search. Not quite adequate.
no subject
Date: 2010-02-14 05:38 pm (UTC)The anti-science crowd will make up stuff, so they don't really need your small opps.
Great explanation and summation of the situation with the email flap.
no subject
Date: 2010-02-15 06:03 pm (UTC)Another one coming up shortly...
Aurrgh
Date: 2010-02-15 04:44 am (UTC)no subject
Date: 2010-02-16 06:31 pm (UTC)no subject
Date: 2010-02-16 06:38 pm (UTC)It also reminds me of the postmodern attacks on science that took place at the same time - they were not so consequential, but as an undergrad, I did have one woman literally laugh in my face because I believed in the existence of DNA. To her, it was merely a social construction reflecting the command-and-control paradigm of the patriarchy, or something like that, and she denied that it corresponded to an actual physical object.
I was traumatized, but got over it. :-P