If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(Smithsonian Magazine)   It seems scientists need to be better about backing up their data; according to new reports, nearly 90% of data in most studies and research are lost within twenty years   (blogs.smithsonianmag.com) divider line 140
    More: Fail, Current Biology, molecular ecology, data store, reproducibility, light-years  
•       •       •

659 clicks; posted to Geek » on 29 Dec 2013 at 9:33 PM (37 weeks ago)   |  Favorite    |   share:  Share on Twitter share via Email Share on Facebook   more»



140 Comments   (+0 »)
   
View Voting Results: Smartest and Funniest

First | « | 1 | 2 | 3 | » | Last | Show all
 
2013-12-30 12:42:14 PM

Jon Snow: THE GREAT NAME: But if (apart from obvious cycles like day/night, summer/winter etc)

What makes it warmer in the daytime vs. the nighttime? What makes it warmer in summer vs. winter?

Who cares? We know that these cycles occur because we have seen them repeatedly and we can (and do) make successful predictions that they will continue. Climatology's predictions of catastrophic warming have not been seen at all and their predictions have not come true.

Jon, you can shimmy up next to things like established meteorogical cycles and try and suggest that maybe your climatology has the same credibility, but it does not. It's a mile away from any kind of objective credibility.


I think you may be kidding yourself. When I point out that climate is unpredictable due to chaotic variability, you claim that the drivers (or forcings, or boundary conditions) make your stuff credible. But science needs predictions. I can predict higher temp next midday versus next midnght. You can predict catastrophic warming by 2100. We both know I will be right, but will you? There is no reason to trust it other than your models, which have all so-far failed.

  You mean like demonstrating the change in the vertical thermal structure of the atmosphere occurring inline with theoretical expectations of enhanced greenhouse warming, which is a completely different behavior than occurs either due to natural variability or due to increases in other kinds of radiative forcing, such as solar?You can prove that wouldn't happen though natural variability? I doubt it.


  You're acting like the climate system is magic and that we don't have physics. You're making a classic fallacious argument from ignorance.

Chaos theory isn't magic. It's real and you have to accept it and stop spreading likes like the white-noise assumption. It would in fact be more like magic if a complex chaotic system like the climate had random behaviour that simulated a series of independent random events - and this would be required for the fluctuations to be white in character, which is required for decadal-scale trends to require anthropogenic cause.
 
2013-12-30 12:46:44 PM

Ishkur: THE GREAT NAME: So, you can model from boundary conditions, but you can never test your model, which is why it is not science.

Not all science is experimentation. In fact, quite a great deal of it isn't. Most science is deductive science: You study the evidence and you build models that account for how the evidence came to be. This is how ALL the historical sciences (including climatology, glaciology, geology, petrology, archeology, paleontology, and yes, even evolution and plate tectonics) operate. Some of them may have some predictive power, but that is not their primary purpose. They are purely explanatory disciplines.


If we frame climatology as a historical science (which you may be trying to do here, not sure) then it would need a HELL of a lot more historical evidence to attribute ANY anthropogenic element to climate behaviour. And it would NEVER predict an acceleration of warming over 21st century LET ALONE try and influence government policy based on it.

 
2013-12-30 12:52:33 PM

THE GREAT NAME: The failure of air temps to rise since 1998


This is the trend in global temps from 1998 to present. The trend is positive.

img.fark.net

The claim that temperature hasn't risen since 1998 is based on cherry picking an extremely warm El Nino year, and using an outdated data set (HadCRUT3) that had known cool bias arising from insufficient coverage of the rapidly warming arctic.

This is no different than looking at the period from 1992-2007 and claiming that we're warming more rapidly than the models and so therefore we've dramatically underestimated the amount of future warming we'll see. It depends on a complete ignorance of the data and climate dynamics.

We are now below the IPCC's lowest predicted temps. It hasn't worked.

Again, the obs are within the model spread. Stop lying.

Its predictions have not come true.

Feel free to refute those I listed above.
 
2013-12-30 12:54:18 PM

THE GREAT NAME: Who cares? We know that these cycles occur because we have seen them repeatedly and we can (and do) make successful predictions that they will continue.


So it's your claim that we have no understanding of why it's warmer during the daytime and summertime than it is during nighttime and wintertime, but rather we simply have observed this to be the case and successfully predicted that it will occur?

In other words, you are claiming that we have no understanding of the physical reasons for this phenomenon?

Yes or no?

So when the summer during 1816 turned out to be cold, that invalidated your naive predictive model and therefore we cannot say anything about whether or not summer will be warmer than winter, let alone what causes this, right?
 
2013-12-30 12:55:03 PM

THE GREAT NAME: Ishkur: THE GREAT NAME: So, you can model from boundary conditions, but you can never test your model, which is why it is not science.

Not all science is experimentation. In fact, quite a great deal of it isn't. Most science is deductive science: You study the evidence and you build models that account for how the evidence came to be. This is how ALL the historical sciences (including climatology, glaciology, geology, petrology, archeology, paleontology, and yes, even evolution and plate tectonics) operate. Some of them may have some predictive power, but that is not their primary purpose. They are purely explanatory disciplines.

If we frame climatology as a historical science (which you may be trying to do here, not sure) then it would need a HELL of a lot more historical evidence to attribute ANY anthropogenic element to climate behaviour. And it would NEVER predict an acceleration of warming over 21st century LET ALONE try and influence government policy based on it.


Maybe an analogy would help.

Say I'm beating you over the head with a baseball bat. Chaos theory, together with physics and biology can be used to predict the effect on your brain. If your calculations don't give a precise, measureable and accurate prediction of the damage to your head, I will deny that beating you with the bat is a bad idea.

That's what you're doing with calls to STOP POLLUTING THE FARKING PLANET.
 
2013-12-30 12:55:11 PM

Jon Snow: THE GREAT NAME: The failure of air temps to rise since 1998

This is the trend in global temps from 1998 to present. The trend is positive.

[img.fark.net image 600x360]

The claim that temperature hasn't risen since 1998 is based on cherry picking an extremely warm El Nino year, and using an outdated data set (HadCRUT3) that had known cool bias arising from insufficient coverage of the rapidly warming arctic.

This is no different than looking at the period from 1992-2007 and claiming that we're warming more rapidly than the models and so therefore we've dramatically underestimated the amount of future warming we'll see. It depends on a complete ignorance of the data and climate dynamics.

We are now below the IPCC's lowest predicted temps. It hasn't worked.

Again, the obs are within the model spread. Stop lying.

Its predictions have not come true.

Feel free to refute those I listed above.


Utter nonsense. Even the IPCC and the UK Met Office accept temps stopped rising. You have just proved that you are a charlatan.
 
2013-12-30 12:57:00 PM

Jon Snow: THE GREAT NAME: Who cares? We know that these cycles occur because we have seen them repeatedly and we can (and do) make successful predictions that they will continue.

So it's your claim that we have no understanding of why it's warmer during the daytime and summertime than it is during nighttime and wintertime, but rather we simply have observed this to be the case and successfully predicted that it will occur?

In other words, you are claiming that we have no understanding of the physical reasons for this phenomenon?


Absolutely not. I said "who cares". You even quoted it. I did not say "we don't know". You made that up. Is making stuff up the only way you can get by? Typical climatologist.
 
2013-12-30 12:59:30 PM

THE GREAT NAME: Jon, you can shimmy up next to things like established meteorogical cycles and try and suggest that maybe your climatology has the same credibility, but it does not. It's a mile away from any kind of objective credibility.


This is pants on head retarded.

The diurnal cycle and seasonal cycles are driven by changes in insolation. This is absolutely no different than climate change. It's a question of energy balance, and has nothing to do with chaos theory.

If chaos theory prevented making meaningful statements about background climatic conditions more than a few weeks in advance, it would prevent us from saying that it will be hotter in summer than winter in the Northern Hemisphere. We would simply not know, on a physical basis.

This is absurd.

You are retreating to the position that we don't need a physical justification for making a claim about the future in cases where we have a record of the behavior in question occurring repeatedly.

So then the End Permian, Triassic-Jurassic, Paleocene Eocene Thermal Maximum, and other various episodes during which a significant carbon pulse drove warming in the paleo record satisfy that absurd position.

But you'll find some reason to once again change your standards to exclude the uncontroversial physics necessitating that increasing GHGs result in an energy imbalance necessitate warming to a higher equilibrium.
 
2013-12-30 12:59:39 PM

MrBallou: THE GREAT NAME: Ishkur: THE GREAT NAME: So, you can model from boundary conditions, but you can never test your model, which is why it is not science.

Not all science is experimentation. In fact, quite a great deal of it isn't. Most science is deductive science: You study the evidence and you build models that account for how the evidence came to be. This is how ALL the historical sciences (including climatology, glaciology, geology, petrology, archeology, paleontology, and yes, even evolution and plate tectonics) operate. Some of them may have some predictive power, but that is not their primary purpose. They are purely explanatory disciplines.

If we frame climatology as a historical science (which you may be trying to do here, not sure) then it would need a HELL of a lot more historical evidence to attribute ANY anthropogenic element to climate behaviour. And it would NEVER predict an acceleration of warming over 21st century LET ALONE try and influence government policy based on it.

Maybe an analogy would help.

Say I'm beating you over the head with a baseball bat. Chaos theory, together with physics and biology can be used to predict the effect on your brain. If your calculations don't give a precise, measureable and accurate prediction of the damage to your head, I will deny that beating you with the bat is a bad idea.

That's what you're doing with calls to STOP POLLUTING THE FARKING PLANET.


So (a) you don't know what chaos theory is and (b) you're a ranting eco-freak and (c) you drew your conclusions about global warming long before you looked at the evidence, if indeed you have ever looked at the evidence.

By the way, if you're SO FARKING WORRIED about pollution, do the world a favour and kill yourself.
 
2013-12-30 01:00:09 PM

THE GREAT NAME: Absolutely not.


Great!

Then what causes it?
 
2013-12-30 01:00:23 PM

THE GREAT NAME: The failure of air temps to rise since 1998 when climatologists said they would. We are now below the IPCC's lowest predicted temps. It hasn't worked.


Oh, I see what your problem is. You think Climatologists are like some sort of long-term weather forecasters and that's all they do (hence your insistence on Meteorology as a better replacement). And that if they make predictions, and those predictions are wrong, it means the entire branch of science is discredited (a fallacy of composition). Of course, that same standard is not applied to weathermen, but I digress.

First of all, you have to understand the difference between the two. To put it briefly, Meteorology studies the "what" and "when" questions concerning the weather. Climatology studies the "how" and "why" questions. It takes a much broader look at interceding conditions at the global level, it considers things like ocean currents, ice caps and ecosystems that have a profound effect on why the weather is the way it is, it analyzes atmospheric compositions and it tracks all of these things globally over lengthy periods of time. It is an incredibly useful science, one of the few that has the temerity to speculate on the ramifications of a number of very large and very involving systems.

These are things that other disciplines just aren't equipped to study, much less understand. But hey, if you still think Meteorology can do all that, and if it did, and Meteorology comes to the same conclusions regarding climate change that the (formerly) climatology did, you'd probably throw out Meteorology as being pseudo-science too. After all, their experts never apologize for being wrong either (especially on the nightly news).
 
2013-12-30 01:01:07 PM

Jon Snow: THE GREAT NAME: Jon, you can shimmy up next to things like established meteorogical cycles and try and suggest that maybe your climatology has the same credibility, but it does not. It's a mile away from any kind of objective credibility.

This is pants on head retarded.

The diurnal cycle and seasonal cycles are driven by changes in insolation. This is absolutely no different than climate change. It's a question of energy balance, and has nothing to do with chaos theory.


In that case, how come we can't predict the weather more than two weeks ahead? It is truly you who is retarded.
 
2013-12-30 01:04:00 PM

THE GREAT NAME: Chaos theory isn't magic.


I don't think you understand what chaos theory actually is. You seem to be handwaving it as some sort of copout exercise in futility. That demonstrates an appalling level of ignorance. To wit, it does not mean "everything is crazy-ass random!". It is random, yes, but in predictable ways. It is still mathematics, after all.

/don't get your science education from Dr. Ian Malcolm in Jurassic Park
 
2013-12-30 01:11:34 PM

dittybopper: How much of that is raw data that you don't need to save long term, and how much of it is processed data that you want to save?


A decent chunk is processed data for my work (maybe around ~70%).  However, generally for our work, we want to keep the raw data as well.  We're always getting requests for data, and oftentimes the effort required to acquire the raw data in the first place makes the data rather valuable.  It's also not uncommon to come up with additional interesting research questions that require the original data to work on a few years later.
 
2013-12-30 01:18:13 PM

THE GREAT NAME: If we frame climatology as a historical science


It is. It is trying to understand how things came to be.

THE GREAT NAME: then it would need a HELL of a lot more historical evidence to attribute ANY anthropogenic element to climate behaviour.


I don't know how much of man's meddling affects the environment, but the science of climatology has these facts to report:

Before the Industrial era, the amount of C02 in the atmosphere was 280 ppm (parts per million).
Today, the amount of C02 in the atmosphere is 392 ppm.

We know that C02 can bring about a greenhouse effect -- one look at blistering Venus, which is 96% C02 and has temperatures hot enough to melt lead -- is clear evidence of this.

And we also know from studying ice cores that the last time the Earth had C02 levels greater than 400 ppm was about 3-5 million years ago, during the Pliocene Epoch. The Earth was much warmer and drier back then. Ocean levels were about 25m higher. There was no arctic ice cap (but Greenland and Antarctica had land ice).

Now, I don't want to jump to conclusions, but given these facts: WHAT DO YOU THINK IS HAPPENING?
 
2013-12-30 01:20:11 PM

THE GREAT NAME: MrBallou: THE GREAT NAME: Ishkur: THE GREAT NAME: So, you can model from boundary conditions, but you can never test your model, which is why it is not science.

Not all science is experimentation. In fact, quite a great deal of it isn't. Most science is deductive science: You study the evidence and you build models that account for how the evidence came to be. This is how ALL the historical sciences (including climatology, glaciology, geology, petrology, archeology, paleontology, and yes, even evolution and plate tectonics) operate. Some of them may have some predictive power, but that is not their primary purpose. They are purely explanatory disciplines.

If we frame climatology as a historical science (which you may be trying to do here, not sure) then it would need a HELL of a lot more historical evidence to attribute ANY anthropogenic element to climate behaviour. And it would NEVER predict an acceleration of warming over 21st century LET ALONE try and influence government policy based on it.

Maybe an analogy would help.

Say I'm beating you over the head with a baseball bat. Chaos theory, together with physics and biology can be used to predict the effect on your brain. If your calculations don't give a precise, measureable and accurate prediction of the damage to your head, I will deny that beating you with the bat is a bad idea.

That's what you're doing with calls to STOP POLLUTING THE FARKING PLANET.

So (a) you don't know what chaos theory is and (b) you're a ranting eco-freak and (c) you drew your conclusions about global warming long before you looked at the evidence, if indeed you have ever looked at the evidence.

By the way, if you're SO FARKING WORRIED about pollution, do the world a favour and kill yourself.


(a) Dude, I have taught courses in chaos theory, so don't try to double-talk your way through to a false result (b) I rarely rant, but I also don't shiat where I eat and (c) 30 years ago, it was obvious to any idiot what would happen if we keep poisoning the planet, so, yeah.

Not sure how your last suggestion would help. I think my bat idea holds more promise.
 
2013-12-30 01:22:10 PM

Ishkur: THE GREAT NAME: The failure of air temps to rise since 1998 when climatologists said they would. We are now below the IPCC's lowest predicted temps. It hasn't worked.

Oh, I see what your problem is. You think Climatologists are like some sort of long-term weather forecasters


Climatologists' predictions of catastrophic anthropogenic global warming are exactly that.

And that if they make predictions, and those predictions are wrong, it means the entire branch of science is discredited (a fallacy of composition).

Until they make some successful predictions, they are not credible as scientists. Religious high priests maybe but not scientists. No fallacy there, Ishkur.

Of course, that same standard is not applied to weathermen, but I digress.

Meteorologists have been seen to produce consistently valid forecasts at below a week timescale. Not always right, but right enough of the time to know there is value in those predictions. This is something climatologists have not achieved.

First of all, you have to understand the difference between the two. To put it briefly, Meteorology studies the "what" and "when" questions concerning the weather. Climatology studies the "how" and "why" questions.

Utter nonsense. Both are concerned with how and why, but both make predictions, i.e. what and when. The only difference is the timescale.

It takes a much broader look at interceding conditions at the global level, it considers things like ocean currents, ice caps and ecosystems that have a profound effect on why the weather is the way it is, it analyzes atmospheric compositions and it tracks all of these things globally over lengthy periods of time.

True, but that hasn't helped in making predictions so far.

It is an incredibly useful science

To whom?

one of the few that has the temerity to speculate on the ramifications of a number of very large and very involving systems.

Temerity as a positive? You're giving away your emotional investment.

These are things that other disciplines just aren't equipped to study, much less understand. But hey, if you still think Meteorology can do all that,

I don't think meteorology can make predictions on a climate timescale. Did I suggest otherwise, or are you putting words in my mouth for the third time in this thread alone?

and if it did, and Meteorology comes to the same conclusions regarding climate change that the (formerly) climatology did, you'd probably throw out Meteorology as being pseudo-science too.

If meteorology was going to do that it would have by now. Instead it accepted the reality of the chaotic system and stopped making claims it could not
back up. Then cliamtology came along and said "we're somehow immune to chaos  and we say the world's about to end".

After all, their experts never apologize for being wrong either (especially on the nightly news).

As a science, it accepted it's limitations. Climatology has not and will not because it isn't really a science anyway.
 
2013-12-30 01:22:54 PM

Ishkur: THE GREAT NAME: Chaos theory isn't magic.

I don't think you understand what chaos theory actually is. You seem to be handwaving it as some sort of copout exercise in futility. That demonstrates an appalling level of ignorance. To wit, it does not mean "everything is crazy-ass random!". It is random, yes, but in predictable ways. It is still mathematics, after all.

/don't get your science education from Dr. Ian Malcolm in Jurassic Park


You certainly don't.
 
2013-12-30 01:25:44 PM

Ishkur: THE GREAT NAME: If we frame climatology as a historical science

It is. It is trying to understand how things came to be.

THE GREAT NAME: then it would need a HELL of a lot more historical evidence to attribute ANY anthropogenic element to climate behaviour.

I don't know how much of man's meddling affects the environment, but the science of climatology has these facts to report:

Before the Industrial era, the amount of C02 in the atmosphere was 280 ppm (parts per million).
Today, the amount of C02 in the atmosphere is 392 ppm.

We know that C02 can bring about a greenhouse effect -- one look at blistering Venus, which is 96% C02 and has temperatures hot enough to melt lead -- is clear evidence of this.

And we also know from studying ice cores that the last time the Earth had C02 levels greater than 400 ppm was about 3-5 million years ago, during the Pliocene Epoch. The Earth was much warmer and drier back then. Ocean levels were about 25m higher. There was no arctic ice cap (but Greenland and Antarctica had land ice).

Now, I don't want to jump to conclusions, but given these facts: WHAT DO YOU THINK IS HAPPENING?


I have no reason to believe any of this implies an imminent climate catastrophe. And yet, that it what climatolgists predict. And they do predict it, even though you say climatology is a historical science (but I don't think you actually believe that, you just backed yourself into a corner, didn't you).
 
2013-12-30 01:28:18 PM

MrBallou: THE GREAT NAME:

(a) Dude, I have taught courses in chaos theory, so don't try to double-talk your way through to a false result


Qualifications please.

(b) I rarely rant, but I also don't shiat where I eat and

People like you always rant.

(c) 30 years ago, it was obvious to any idiot what would ...

30 years ago morons like you were running around screaming about GLOBAL COOLING.
 
2013-12-30 01:35:15 PM

Ishkur: And we also know from studying ice cores that the last time the Earth had C02 levels greater than 400 ppm was about 3-5 million years ago, during the Pliocene Epoch. The Earth was much warmer and drier back then. Ocean levels were about 25m higher. There was no arctic ice cap (but Greenland and Antarctica had land ice).

Now, I don't want to jump to conclusions, but given these facts: WHAT DO YOU THINK IS HAPPENING?


Obviously, time travelers took SUVs back to the Pliocene Epoch and humans caused global warming back then. Because only humans cause warming, and humans are obviously also to blame for the warming before oil started being burned in the 1950s.
 
2013-12-30 01:35:51 PM
So, where is the data for this backup study available from?
 
2013-12-30 01:41:43 PM

THE GREAT NAME: MrBallou: THE GREAT NAME:

(a) Dude, I have taught courses in chaos theory, so don't try to double-talk your way through to a false result

Qualifications please.


Ph.D. Genetics/Population Biology. Yours?

(b) I rarely rant, but I also don't shiat where I eat and

People like you always rant.


You seem to think you know a lot about me. Are you in the NSA?

(c) 30 years ago, it was obvious to any idiot what would ...

30 years ago morons like you were running around screaming about GLOBAL COOLING.


Actually, we were running around screaming that a finite planet cannot continue to support unbridled human expansion forever and that we should think about the long-term instead of just immediate profit. Same as now.
 
2013-12-30 01:47:16 PM

THE GREAT NAME: Until they make some successful predictions, they are not credible as scientists.


Listen: Making predictions and failing is the BEDROCK of science. Falsifying claims is what separates it from every other form of investigative inquiry. Science is supposed to fail. It fails constantly. It fails, stumbles, and falls into success. There are a thousand failed predictions for every successful one. And that is how the more robust theories are developed: They're the ones that stand up to scrutiny and weather the onslaught of millions of studies done every day trying to prove them wrong. Trial and error, throw out what doesn't work, keep what does, each time getting a little better at it, and little more refined, a little more accurate, a little more correct. Being wrong is the hallmark of science. It is a feature, not a bug. The greatest phrase in science is not "Eureka, I've found it!" but rather "Hmm. That's odd. Why did that happen?"

If a Climatologist makes a prediction and it's proven wrong, that doesn't make Climatology wrong (which is kind of a misnomer anyway, because Climatology isn't a thing, it's a process), that makes that particular climatologist's claim wrong. Know the difference.

THE GREAT NAME: I have no reason to believe any of this implies an imminent climate catastrophe.


That's not what I'm asking. I'm asking: WHAT DO YOU THINK IS HAPPENING?

Because these are the facts. Ask a hundred scientists about what they mean, and you'll get 101 predictions. Some may be right. Some may be wrong. But that is why we do the science. Because some will be proven wrong does not mean we throw out the science.

THE GREAT NAME: even though you say climatology is a historical science


Climatology is a historical science. It also makes predictions based on its historical findings. It does both (I never said it didn't).

You are really good at dodging, evading questions, and not posting any facts, and you have an amazing capacity for self-deception. How on Earth do you do that?
 
2013-12-30 01:58:30 PM

THE GREAT NAME: Meteorologists have been seen to produce consistently valid forecasts at below a week timescale. Not always right, but right enough of the time to know there is value in those predictions. This is something climatologists have not achieved.


Climatologists aren't looking at next week's weather. They're looking at global trends over the past 50 years, and extrapolating over the next 50. (or 500, or 5000, or 5 million). They aren't interested in localized specifics.

THE GREAT NAME: I don't think meteorology can make predictions on a climate timescale. Did I suggest otherwise, or are you putting words in my mouth for the third time in this thread alone?


That is indeed what you claimed in this very thread. So if you're changing your answer, then what discipline is better equipped at studying macro changes in the Earth's atmosphere? Give reasons why.

THE GREAT NAME: As a science, it accepted it's limitations. Climatology has not and will not because it isn't really a science anyway.


Those aren't limitations, those are demarcated fields of study that are too broad and byzantine to be encapsulated within a single discipline of science, necessitating fragmentation into specialized areas of research. That's what happens when we end up knowing too much about one thing. We separate it into component parts and dedicate each part to a community of specialists.

You seem to be dodging the issue and I'm going to keep asking you this: If Climatology is not science, as you claim, then what field do you suggest should study the areas of expertise normally covered by Climatology? We've already gone over the fact that Meteorology is woefully ill-equipped at handling the additional obligations. So what else?
 
2013-12-30 02:46:32 PM
Autocad used a 3 1/2" floppy icon for "save."

Should I be worried?

A cloud icon for "save" would worry me too.
 
2013-12-30 02:47:17 PM

THE GREAT NAME: In that case, how come we can't predict the weather more than two weeks ahead


Weather and climate are not the same thing, despite your complete and total failure to understand this.

Weather forecasting is attempting to predict a single realization of a host of stochastic processes which makes it a problem that is incredibly sensitive to initial conditions. Chaos applies here.

Understanding the change to the background state of a system when you change its boundary conditions is a completely different issue.

A single roll of a fair six-sided die is an initial value problem. Realistically, the end result is so dependent on so many stochastic processes that the outcome of a prediction is no better than chance (though chance bounded by the physical characteristics of the die, i.e. 1-6). This is analogous to long term weather forecasting, where skill declines over time until it decays to climatology (which is the meteorological analogy of the 1-6 range).

And yet!

I can confidently tell you that the mean of many rolls of a fair six-sided die will converge over time to 3.5. Even better, I can tell you that the average value would increase to 4.5 if I switched out the six-sided die with a fair eight-sided die. And I could tell you a great many things about other changes to the average statistics that occur with such a change in number of sides. This is analogous to projecting changes in climatology (e.g. average temperatures, average precipitation amounts, frequencies of some kinds of events, large scale features of the general circulation, etc.) due to changes in boundary values, such as radiative forcing at the top of the atmosphere, via increases in GHGs.

You don't get to throw up your hands and shriek "chaos!" as though that's an actual problem for talking about climate change. It's not, and it's an embarrassingly easy concept to grasp.
 
2013-12-30 02:52:02 PM

MrBallou: FTFA: "Some of the time, for instance, it was saved on three-and-a-half inch floppy disks, so no one could access it, because they no longer had the proper drives,"

I can confirm. I even have the 5 1/4" floppy disks holding my Master's degree data, and a bunch of later stuff on Zip Disks. Sorry, Science.

Someday the Cloud will be obsolete and you'll have to transfer all your data to ectoplasm storage via direct mental link, or some damn thing.


I still have my King's Quest and Infocom floppies. They still work but I always keep the 5.25 and 3.5 drives handy for new builds.

The sad thing is I've actually had people ask me what those drives are for.


/I should make a few extra copies or at least re-magnetize/refresh the data
 
2013-12-30 02:59:59 PM
SuperT:

MrBallou: Someday the Cloud will be obsolete and you'll have to transfer all your data to ectoplasm storage via direct mental link, or some damn thing.

this is why every so often, you move your data to the new format.


art.penny-arcade.com
 
2013-12-30 03:00:12 PM

THE GREAT NAME: You can prove that wouldn't happen though natural variability? I doubt it.


Another argument from ignorance.

Just because you don't understand how potential different drivers of surface temperature affect other aspects of the climate system in ways distinct from one another, the variability internal to the system, etc. that does not mean everyone else is equally ignorant.

Increasing the greenhouse effect by increasing atmospheric CO2 levels results in stratospheric cooling, over timescales distinct from any source of natural variability that exists in either the observational record, atmospheric physics and chemistry, or state of the art modeling.

Is it possible that there exists some heretofore unknown kind of natural variability that is simultaneously capable of affecting changes consistent only with increased GHGs as well as simultaneously managing to prevent exactly those changes that fundamental chemistry and physics necessitate that increased GHGs must result in?

It's possible. Just like it's possible that I have an invisible dragon breathing heatless fire in my garage.

Look.

You know nothing about this topic. You scream about pink noise and chaos and cannot understand fundamental ideas like boundary value problems, how GHGs, the sun, and the ocean heat the Earth's surface, any of the successful predictions of climate science, that there is no demarcation between the science underlying our understanding of anthropogenic climate change vs. our understanding of other aspects of the Earth system or general planetary behavior. Hell, you don't even understand that the "70s cooling consensus myth" is complete and utter rubbish, and that even in the 70s we understood that increasing GHGs would lead to warming.

So why do you post in these threads?

It is clearly not to gain knowledge, as you appear to be immune to corrections to your errors/ignorance. It is equally clear that it is not to disseminate knowledge, as you have none on this topic. Is it to seek confrontation?
 
2013-12-30 03:57:42 PM

dittybopper: if the data is important enough that you think it should be preserved for future generations, then the physical storage is really a minor consideration.


The problem is you don't always know what is (or when it will become) important, hence the importance of saving it all.

Some guy taking temperatures in his backyard in the 1800's probably didn't think it would mean a damn thing, and now its center stage in the global weather changing debate. Whole economies are going to shift on incomplete data because many paper records of the era were still lost.

There are scientists gathering data today that don't have a clue what it will mean to someone in a few hundred years time. Most paper isn't being stored in anywhere near ideal conditions, and it isn't being backed up, and it isn't globally accessible.

You can solve all three problems by going to the cloud. The only roadblocks are sorting out formats and access rights.
 
2013-12-30 04:24:16 PM

untaken_name: You don't think they actually want people to try and repeat their results, do you? C'mon, what the hell kind of science is that? You're just supposed to publish as much as possible as quickly as possible under the assumption no one will actually care to understand what you wrote and the grants will keep flowing.


How does the data being kept help with repeating the results? If the data is flawed then you will still get flawed results no matter how often the analysis is redone. If people want to see if the paper is true, then they should set up another experiment based on the methodology. Not simply re-run the numbers. I also fail to see how re-running the numbers aids in understanding the material. It isn't as if clicking a few boxes in some maths program and waiting for the results grants new insights into the hypotheses or the conclusions drawn from them.
 
2013-12-30 04:48:34 PM

MrBallou: FTFA: "Some of the time, for instance, it was saved on three-and-a-half inch floppy disks, so no one could access it, because they no longer had the proper drives,"

I can confirm. I even have the 5 1/4" floppy disks holding my Master's degree data, and a bunch of later stuff on Zip Disks. Sorry, Science.

Someday the Cloud will be obsolete and you'll have to transfer all your data to ectoplasm storage via direct mental link, or some damn thing.


GO AWAY MOM I'M UPSPIRTING MY SCIENCE DATA!!
 
2013-12-30 06:50:49 PM

DerAppie: untaken_name: You don't think they actually want people to try and repeat their results, do you? C'mon, what the hell kind of science is that? You're just supposed to publish as much as possible as quickly as possible under the assumption no one will actually care to understand what you wrote and the grants will keep flowing.

How does the data being kept help with repeating the results? If the data is flawed then you will still get flawed results no matter how often the analysis is redone. If people want to see if the paper is true, then they should set up another experiment based on the methodology. Not simply re-run the numbers. I also fail to see how re-running the numbers aids in understanding the material. It isn't as if clicking a few boxes in some maths program and waiting for the results grants new insights into the hypotheses or the conclusions drawn from them.


Because it helps you decide whether the data is flawed or the analysis that was done on it. It allows separation of the two parts. Otherwise you just go "There may be something wrong but we don`t know where or what"
 
2013-12-30 06:55:33 PM
And in Canada, we just chuck it in the garbage while smiling.
 
2013-12-30 06:58:41 PM

DerAppie: untaken_name: You don't think they actually want people to try and repeat their results, do you? C'mon, what the hell kind of science is that? You're just supposed to publish as much as possible as quickly as possible under the assumption no one will actually care to understand what you wrote and the grants will keep flowing.

How does the data being kept help with repeating the results? If the data is flawed then you will still get flawed results no matter how often the analysis is redone. If people want to see if the paper is true, then they should set up another experiment based on the methodology. Not simply re-run the numbers. I also fail to see how re-running the numbers aids in understanding the material. It isn't as if clicking a few boxes in some maths program and waiting for the results grants new insights into the hypotheses or the conclusions drawn from them.


New analyses are possible on existing/old data. It's not that someone else does the same analysis on the same data (because aside from error-checking that would be a waste of time), but someone else can do a new analysis on that data.

Why would you think that the purpose of keeping the data is simply to allow someone to rerun the same analyses that have already been done? Do you think that scientists have somehow overlooked the pointlessness of that, and want to just redo the same work over and over again, because that will somehow (actually, it won't at all) enable them to roll around in grant money?
 
2013-12-30 07:42:03 PM

HighZoolander: DerAppie: untaken_name: You don't think they actually want people to try and repeat their results, do you? C'mon, what the hell kind of science is that? You're just supposed to publish as much as possible as quickly as possible under the assumption no one will actually care to understand what you wrote and the grants will keep flowing.

How does the data being kept help with repeating the results? If the data is flawed then you will still get flawed results no matter how often the analysis is redone. If people want to see if the paper is true, then they should set up another experiment based on the methodology. Not simply re-run the numbers. I also fail to see how re-running the numbers aids in understanding the material. It isn't as if clicking a few boxes in some maths program and waiting for the results grants new insights into the hypotheses or the conclusions drawn from them.

New analyses are possible on existing/old data. It's not that someone else does the same analysis on the same data (because aside from error-checking that would be a waste of time), but someone else can do a new analysis on that data.


Well, yes. A new analysis can be done on old data. But that has nothing to do with the comment I quoted or the comment I made.

Why would you think that the purpose of keeping the data is simply to allow someone to rerun the same analyses that have already been done? Do you think that scientists have somehow overlooked the pointlessness of that, and want to just redo the same work over and over again, because that will somehow (actually, it won't at all) enable them to roll around in grant money?

I now really hope you quoted the wrong person. If not, do read on:

Why would you think that I think that? Because it is easy to do a superficial reading of a post and comment from up high? Do you think I am a moron? Didn't I say in my comment that doing the same analysis based on the same data was useless? Did you manage to miss that in your self-righteous need to correct people who aren't wrong? Where did I ever mention grants?

Just look at this part of the original quote:

You're just supposed to publish as much as possible as quickly as possible under the assumption no one will actually care to understand what you wrote and the grants will keep flowing.

How does keeping the data around help people understand the research? It doesn't. Which is exactly my point. My quote had very little to do with my understanding of how things work but with my curiosity about the reasoning behind the comment I quoted.

dready zim: Because it helps you decide whether the data is flawed or the analysis that was done on it. It allows separation of the two parts. Otherwise you just go "There may be something wrong but we don`t know where or what"


When a replicated study gives different results, that is when you start looking at whether or not the datasets are correct (the maths should have been checked during peer-review). Without that second dataset it is pretty much impossible to tell whether or not the first dataset is wrong (apart from glaring errors such as values outside of the possible range). And then a third experiment should be done to determine which one was right. So while I do agree that keeping the datasets around definitely is useful, it does not help with replicating a study. Which the person I quoted insinuated and whose reasoning behind his statement I was curious about.
 
2013-12-30 08:01:23 PM

DerAppie: I now really hope you quoted the wrong person


Yep, my bad. I missed that you were responding to untaken_name (who did mention grants, in what you responded to) and conflated the text into one post, hence misreading the thrust of what you said.
 
2013-12-30 08:44:14 PM

WelldeadLink: So, where is the data for this backup study available from?


They tell you right in the paper. Dryad. DOI:10.5061/dryad.q3g37
 
2013-12-30 11:13:39 PM
It's not my responsibility to keep my data for 20 years after I've published.  I have new projects with new deadlines and new papers to write.  If someone wants to store my data for a few decades, they can pay for it.

I run weather models that routinely generate tens of terabytes of data per project.  Have fun.
 
Displayed 40 of 140 comments

First | « | 1 | 2 | 3 | » | Last | Show all

View Voting Results: Smartest and Funniest


This thread is closed to new comments.

Continue Farking
Submit a Link »






Report