Real Climate is claiming that data is available and has this nice link and stuff (given in the slahsdot summary.)
Following their link I noticed that there was no link to raw data for stratospheric temperatures but there was a link to processed data.
I followed the link to the processed data in the hopes that there would be some explanation as to why only processed data was available. I discovered that the processed data wasnt available either, instead the link only pointed to a page with GIF files (graphs.)
Essentially, Real Climate just lied to us about the stratospheric data. Not only is the raw data unavailable, the processed data isnt available either even tho it claims it is available and claims to link to it.
I then clicked around most of the "raw" sites linked to and almost all are fairly devoid of data.
Mr. Jones, the public may buy your bullshit because they might think a GIF file with a graph is relevant "data" but I do not. Mr. Jones, RELEASE YOUR FUCKING RAW DATA.
Their predictions were off by 80%, but you still put value in their prediction method?
Think about it. I'd say that apparently they have no fucking clue whats going to happen next because their models are missing something important, because being wrong by 80% either direction means exactly that.
If the _results_ from the lab in question match up with other independent results, what possible grounds to laymen have to presume the data was deliberately changed? Unless they assume that all independent labs falsified their data in concert, which would be a hell of a conspiracy.
Show me the independent verification of the paper Jones et al 1990.
Lets get this part out of the way real quickly. Firstly is that Jones is the man who runs that Real Climate this article mentions, and is also the big climate cheese at the CRU whos emails were hacked and said all sorts of questionable things in them.
Now, the paper in question is supposedly the definitive attempt the measure the Urban Heat Island effect. Almost two decades worth of Freedom Of Information requests were thwarted by Jones and one of his co-authors, Wang. Of focus here is that the paper claims that they took the raw china temperature data and weeded out the site locations which had unknown site provenance. That is, specifically, that they supposedly only used data from sites which had little to no urbanization or instrumentation changes over the period of the study.
So the site provenance data needs to be available in order to independently verify this paper. Unfortunately, IT DOES NOT EXIST ANY LONGER, and according to the DOE which produced a report (written by Zeng and *Wang*) AT THE SAME TIME as the Jones and *Wang* 1990 paper was being written, STATED EXPLICITLY THAT THIS DATA DID NOT EXIST.
I'll take your independent verification argument seriously when it actually becomes possible to independently verify the works of these fraudsters. Thats right, it is IMPOSSIBLE to even BEGIN to verify some of their work BECAUSE the data they claim to have had DOES NOT EXIST and PROBABLY DIDNT EVEN EXIST WHEN THE WORK WAS DONE.
Seriously, how would you peer review a paper based on data you can't look at?
This is simple. If the paper agrees with your beliefs, give it your stamp of approval.
Remember that along with submitting a paper to a peer reviewed journal, they also submit recommendations as to whom should peer review it.
The problem with the summary (havent read the article) is it fails to mention that Real Climate is run by Jones, the same guy at the CRU who is caught up in this scandal, so that data that is now magically available on the Real Climate site is automatically suspect.
Also, garbage collection only works for memory. If that block of memory holds a resource (file, window,...) you still need to explicitly release the resource.
This is true but fails to mention that these modern managed languages also give you all the tools you need so that when the garbage collector comes around to clean up, it calls your destructor first.
There are many unmanaged languages with destructors.. but the programmer can fail to call them. Managed languages dont fail to call them unless something unrecoverable happens.
Does that work even if you're targeting generic i686, which only goes up to MMX? Does the speed gain outweigh the time overhead to check which instruction sets are available and the space overhead of multiple implementations, one for each instruction set?
When speed matters, runtime processor checking is a triviality. Think about it.
Anyways, runtime cpu checking is normally only done once, often during program initialization.
There is a reason that the fastest implementations of heavy-lifting algorithms (encryption, compression, etc..) are written in assembly, and its not because the compiler is only slightly sub-optimal. Its because the abstract machine model for the language simply doesnt contain the concepts necessary to describe the (presumed to be.. TANSTATFC) optimal process such as explicitly leveraging the x86 flags register and the instructions which manipulate it, which means that you cannot coerce the compiler to emit anything that even resembles the optimal process.
"You can't get there from here."
I'm sure there will be lots of posts about premature optimization. Those arguments apply to application development, but often not to library development.
and even in prehistoric times have never rised this fast.
How do you know that? Seriously.
Instead of processing the claims made by people, try processing their methodology. Can their methodology make the claim you are making? Seriously?
Now, what method of measurement (pick a proxy, think about how its collected and what assumptions must be made) of prehistoric atmospheric CO2 levels will give them better than 100 year resolution back to say 100000 years? How about 1000000 years?
What you have just claimed is that they have better than 100 year proxy resolution all the way back to the beginning of time, when the reality is that for anything back more than a few thousand years they can't even be certain about what century they are measuring when they look at, for example, a section of an ice core (because they make an assumption about precipitation rates that isnt anything better than ballpark even for recent years)
You arm-chair claim-spouters need to shut up with your bullshit.
Well, when it stops warming thats not evidence that the AGW theory is wrong. In fact, when it cools for a decade it still isnt taken as evidence that the theory of AGW is wrong. When there are more hurricanes, thats taken as evidence of AGW. When there are less hurricanes, that is also taken as evidence of AGW. When it rains less in a region, thats taken as evidence of AGW. When it rains more in a region, that too is taken as evidence of AGW.
It is quite clear to me that the theory of AGW is unfalsifiable by design.
The majority of all the 'raw data' is available on line.
No, it is not. Really. There is lots of post-processed data around. For instance, the data from the NOAA was and is post-processed and that post-processing even created a serious problem with the data which incorrectly labeled 1998 as the hottest year on record, instead of 1934.
I don't suppose you remember the problem with the Jones-Wang UHI dataset either. The raw data is available, but important details (the change in the recording sites land usage and position changes over time) were either made up or destroyed intentionally. The Jones-Wang team had claimed that the China data represented an ideal situation in which to measure the Urban Heat Island effect because there were many sites with very detailed land-use records, as well as records on when instrumentation equipment and locations were changed.
After many years (over a decade) trying to get the important data (the land use changes, etc) released so that others could verify the work, it was eventually admitted by Jones that he never had the data, but he claimed that Wang had it. Meanwhile it was discovered that Wang had written a report about the same time as the Jones-Wang paper for the DOE/CAS, in which he claimed that such detailed information wasn't even available.
So Jones was claiming that Wang claimed he had the data needed to conduct the study that they did, while a study authored by Wang for the DOE/CAS claimed the data didn't exist.
Of course, the Jones-Wang paper in still cited as evidence for a low UHI, and that Wang's two-sidedness on this matter has never been reviewed by his university or the DOE.
I've been telling people this for years, that climatology = statistics, that all of their data is collected by the real sciences like geology and oceanography.
All these people do is fuck around with numbers, but they arent actually school experts in fucking around with numbers.
That this important fact is overlooked by just about everybody is a grand statement of how apathetic everyone really is. Nobody seems to care about the actual facts.
When the hack became public and "climate-gate" was unfolding, people were asking on RealClimate.org (one of the sites involved somehow with climategate) for explanations about the numbers and just what the scientists and researchers were discussing when they were talking about tricks in correlating various datum. In the first 250 comments or so, no one brought said anything about global warming/climate change not being real or if it was caused by humans or not.
Note that this is the first 250 ACCEPTED comments. Try to comment there and you will find that ALL comments are moderated, that they DO NOT get posted until going through the human censor weeding out what they don't like.
Depends, what kind of scientist are you? are you a climatologist?
Climatologist eh.. what do you suppose a climatologist does that is scientific?
See, scientists perform this thing called the Scientific Method. They set up experiments, collect results, and all those nifty things.
On the other hand, a climatologist does not perform the Scientific Method. Instead they process data collected by real scientists, such as Geologists and Oceanographers, in a statistical manner. So the really important degree that they should have is one in STATISTICS.
I trust a statistician more than some half-baked earth scientist who couldn't or wouldn't hack it performing actual science. Did you know that these guys hide their data, methods, and results, from people with degrees in statistics? Thats what climatologists do.
Except I have no idea how you can come to a conclusion that cutting energy usage will cause us to die off in the billions.
You heard it here first. The warmers can't even imagine how their policies could ever hurt people. In their eyes, its patently unthinkable that their policy could ever have negative effects, that they couldnt even imagine even the most unlikely scenarios that would cause bad things.
The talking head of the NOAA is most definitely in on some sort of conspiracy. Hansen is the ridiculous mother fucker who was publicly saying how McIntyre was just out to get him, that his data was perfect, that McIntyre's observation that there was a problem with the NOAA data sets was unscientific bullshit.
A year later the NOAA was backtracking and thanking McIntyre for finding Hansens fucking retarded mistakes.
Lets not mention the DECADE of statistical studies that were all based on that bad data, that are still taken as gospel.
Let me repeat that. The NOAA had been passing out bad data to climate scientists, data that greatly exaggerated the recent warming. Climate scientists used this data to support the AGW and Global Warming theories. Two years ago, after a decade of this bad data being used, it was discovered how bad it was. But all of the studied based on the bad data are STILL CITED AS PROOF OF AGW.
Hansen of course claims that the bad data was "inconsequential" and that the results of those studies would be the same with the good data, even though they had to admit that 1998 is NOT actually the warmest year on record. The hottest year on record is actually 1934. Thats right. 1934. Plenty of hot years back then, as it turns out.
It should be pointed out than in any 3+ person game, the optimal strategy is [B]always[/B] to team up. Strive to be in the strongest coalition, destroy the opposition to it, and when dominance is inevitable start in-fighting trying to break off the weakest pieces of the coalition and destroy them to, until its a 2-player game.
Pascal just has a more verbose syntax and C has looser type-coercion rules.
The two reasons that I try to use neither, especially the bit about C's non-explicit type promoting. It eventually leads to a hard-to-find bug in any non-trivial project.
But languages like VB and PHP practically encourage it.
How? Its one thing to say it, and another thing to justify it. In my experience, VB has all the same basic design-freedoms that the C language has (functions, structures, a wide assortment of branching instructions), but also has classes and mostly-enforced garbage collection.
I am wondering why you've picked VB as the language easy to make a mess of things in, when quite clearly its even easier in C.
BASIC is a common language for business applications because 30 years ago people who took a programming course learned this language. You couldn't buy a desktop that didnt come with some sort of BASIC (for many computers it was in ROM), so it made sense to teach it. So there the company is, with many employees but no professional programmers on staff.. but Bob from accounting listed his BASIC programming credentials on his resume. So lets give Bob a few days before crack at it before we go looking for an expensive alternative like contracting the work out.
Twice now I've suggested rewrites from the ground up, arguing that it would take me less time to duplicate the app but with easy capacity for adding features
That program probably did not start as something worthy of a Design Monument.. that thing of beauty that gets the author to pat himself on the back. Also remember that in almost any language, its easier to write code than to read it. Mature code ends up looking ugly, but all those operations have or had a real purpose. That funny looking bit whose purpose you can't understand, thats probably a bug fix and you too will need to do something like it somewhere.. and it too will look just as ugly.
Its easier to write code than to read it. Design Monuments don't last in the real world.
I too was sure that Modula-2 would be something good.. and the reason is because the creator of the language had lots of experience creating languages.
Later I came to realize that NONE of his languages really caught on outside of academic circles or niche markets.
His languages:
PL/0, Euler, Algol W, Pascal, Modula, Modula-2 and Oberon.
Re:Experimental set-up raises a few questions
on
Ants That Can Count
·
· Score: 1
So it makes perfect sence that if the Ant got to where he thought the nest was...
With Obama, any criticism at all is viewed as a racial attack. The "tea baggers" are called racists, for example.
It used to be that calling someone black was a device used to undermine, but NOW the device most used is to call someone racist. Need I remind us all that a white man sat on capital hill calling bankers racists for not lending to those poor disadvantaged black people, and that this ultimately lead to the credit collapse.
Slashdot Mirror
Real Climate is claiming that data is available and has this nice link and stuff (given in the slahsdot summary.)
Following their link I noticed that there was no link to raw data for stratospheric temperatures but there was a link to processed data.
I followed the link to the processed data in the hopes that there would be some explanation as to why only processed data was available. I discovered that the processed data wasnt available either, instead the link only pointed to a page with GIF files (graphs.)
Essentially, Real Climate just lied to us about the stratospheric data. Not only is the raw data unavailable, the processed data isnt available either even tho it claims it is available and claims to link to it.
I then clicked around most of the "raw" sites linked to and almost all are fairly devoid of data.
Mr. Jones, the public may buy your bullshit because they might think a GIF file with a graph is relevant "data" but I do not. Mr. Jones, RELEASE YOUR FUCKING RAW DATA.
Have you ever tried to have a rational discussion about climate change with someone who's either unaware of willfully ignorant of the science?
I got a better question.
How you ever discussed a climate paper where you had access to both the data and methodology used by its authors?
Their predictions were off by 80%, but you still put value in their prediction method?
Think about it. I'd say that apparently they have no fucking clue whats going to happen next because their models are missing something important, because being wrong by 80% either direction means exactly that.
If the _results_ from the lab in question match up with other independent results, what possible grounds to laymen have to presume the data was deliberately changed? Unless they assume that all independent labs falsified their data in concert, which would be a hell of a conspiracy.
Show me the independent verification of the paper Jones et al 1990.
Lets get this part out of the way real quickly. Firstly is that Jones is the man who runs that Real Climate this article mentions, and is also the big climate cheese at the CRU whos emails were hacked and said all sorts of questionable things in them.
Now, the paper in question is supposedly the definitive attempt the measure the Urban Heat Island effect. Almost two decades worth of Freedom Of Information requests were thwarted by Jones and one of his co-authors, Wang. Of focus here is that the paper claims that they took the raw china temperature data and weeded out the site locations which had unknown site provenance. That is, specifically, that they supposedly only used data from sites which had little to no urbanization or instrumentation changes over the period of the study.
So the site provenance data needs to be available in order to independently verify this paper. Unfortunately, IT DOES NOT EXIST ANY LONGER, and according to the DOE which produced a report (written by Zeng and *Wang*) AT THE SAME TIME as the Jones and *Wang* 1990 paper was being written, STATED EXPLICITLY THAT THIS DATA DID NOT EXIST.
Somebody completely made it up (probably Wang) and so far, nothing has been done about the allegations of outright scientific fraud.
I'll take your independent verification argument seriously when it actually becomes possible to independently verify the works of these fraudsters. Thats right, it is IMPOSSIBLE to even BEGIN to verify some of their work BECAUSE the data they claim to have had DOES NOT EXIST and PROBABLY DIDNT EVEN EXIST WHEN THE WORK WAS DONE.
The sites you are linking to used old bad data for warmest years, that was (embarrassingly) corrected by NASA GIS two years ago.
Seriously, how would you peer review a paper based on data you can't look at?
This is simple. If the paper agrees with your beliefs, give it your stamp of approval.
Remember that along with submitting a paper to a peer reviewed journal, they also submit recommendations as to whom should peer review it.
The problem with the summary (havent read the article) is it fails to mention that Real Climate is run by Jones, the same guy at the CRU who is caught up in this scandal, so that data that is now magically available on the Real Climate site is automatically suspect.
They obviously are only allowing questions, instead of rebuttals to their statements. Seems obvious to me.
Also, garbage collection only works for memory. If that block of memory holds a resource (file, window, ...) you still need to explicitly release the resource.
This is true but fails to mention that these modern managed languages also give you all the tools you need so that when the garbage collector comes around to clean up, it calls your destructor first.
There are many unmanaged languages with destructors.. but the programmer can fail to call them. Managed languages dont fail to call them unless something unrecoverable happens.
Does that work even if you're targeting generic i686, which only goes up to MMX? Does the speed gain outweigh the time overhead to check which instruction sets are available and the space overhead of multiple implementations, one for each instruction set?
When speed matters, runtime processor checking is a triviality. Think about it.
Anyways, runtime cpu checking is normally only done once, often during program initialization.
There is a reason that the fastest implementations of heavy-lifting algorithms (encryption, compression, etc..) are written in assembly, and its not because the compiler is only slightly sub-optimal. Its because the abstract machine model for the language simply doesnt contain the concepts necessary to describe the (presumed to be.. TANSTATFC) optimal process such as explicitly leveraging the x86 flags register and the instructions which manipulate it, which means that you cannot coerce the compiler to emit anything that even resembles the optimal process.
"You can't get there from here."
I'm sure there will be lots of posts about premature optimization. Those arguments apply to application development, but often not to library development.
proving that everything is above board and that any errors were accidental and published in good faith.
How can there ever be good faith if they are actively trying to prevent the finding of any errors, accidental OR intentional?
I argue that they cannot be trusted even if the errors are accidental because they obstruct the identification of errors.
and even in prehistoric times have never rised this fast.
How do you know that? Seriously.
Instead of processing the claims made by people, try processing their methodology. Can their methodology make the claim you are making? Seriously?
Now, what method of measurement (pick a proxy, think about how its collected and what assumptions must be made) of prehistoric atmospheric CO2 levels will give them better than 100 year resolution back to say 100000 years? How about 1000000 years?
What you have just claimed is that they have better than 100 year proxy resolution all the way back to the beginning of time, when the reality is that for anything back more than a few thousand years they can't even be certain about what century they are measuring when they look at, for example, a section of an ice core (because they make an assumption about precipitation rates that isnt anything better than ballpark even for recent years)
You arm-chair claim-spouters need to shut up with your bullshit.
If AGW was proven wrong tomorrow
Well, when it stops warming thats not evidence that the AGW theory is wrong. In fact, when it cools for a decade it still isnt taken as evidence that the theory of AGW is wrong. When there are more hurricanes, thats taken as evidence of AGW. When there are less hurricanes, that is also taken as evidence of AGW. When it rains less in a region, thats taken as evidence of AGW. When it rains more in a region, that too is taken as evidence of AGW.
It is quite clear to me that the theory of AGW is unfalsifiable by design.
The majority of all the 'raw data' is available on line.
No, it is not. Really. There is lots of post-processed data around. For instance, the data from the NOAA was and is post-processed and that post-processing even created a serious problem with the data which incorrectly labeled 1998 as the hottest year on record, instead of 1934.
I don't suppose you remember the problem with the Jones-Wang UHI dataset either. The raw data is available, but important details (the change in the recording sites land usage and position changes over time) were either made up or destroyed intentionally. The Jones-Wang team had claimed that the China data represented an ideal situation in which to measure the Urban Heat Island effect because there were many sites with very detailed land-use records, as well as records on when instrumentation equipment and locations were changed.
After many years (over a decade) trying to get the important data (the land use changes, etc) released so that others could verify the work, it was eventually admitted by Jones that he never had the data, but he claimed that Wang had it. Meanwhile it was discovered that Wang had written a report about the same time as the Jones-Wang paper for the DOE/CAS, in which he claimed that such detailed information wasn't even available.
So Jones was claiming that Wang claimed he had the data needed to conduct the study that they did, while a study authored by Wang for the DOE/CAS claimed the data didn't exist.
Of course, the Jones-Wang paper in still cited as evidence for a low UHI, and that Wang's two-sidedness on this matter has never been reviewed by his university or the DOE.
I've been telling people this for years, that climatology = statistics, that all of their data is collected by the real sciences like geology and oceanography.
All these people do is fuck around with numbers, but they arent actually school experts in fucking around with numbers.
That this important fact is overlooked by just about everybody is a grand statement of how apathetic everyone really is. Nobody seems to care about the actual facts.
When the hack became public and "climate-gate" was unfolding, people were asking on RealClimate.org (one of the sites involved somehow with climategate) for explanations about the numbers and just what the scientists and researchers were discussing when they were talking about tricks in correlating various datum. In the first 250 comments or so, no one brought said anything about global warming/climate change not being real or if it was caused by humans or not.
Note that this is the first 250 ACCEPTED comments. Try to comment there and you will find that ALL comments are moderated, that they DO NOT get posted until going through the human censor weeding out what they don't like.
Depends, what kind of scientist are you? are you a climatologist?
Climatologist eh.. what do you suppose a climatologist does that is scientific?
See, scientists perform this thing called the Scientific Method. They set up experiments, collect results, and all those nifty things.
On the other hand, a climatologist does not perform the Scientific Method. Instead they process data collected by real scientists, such as Geologists and Oceanographers, in a statistical manner. So the really important degree that they should have is one in STATISTICS.
I trust a statistician more than some half-baked earth scientist who couldn't or wouldn't hack it performing actual science. Did you know that these guys hide their data, methods, and results, from people with degrees in statistics? Thats what climatologists do.
Except I have no idea how you can come to a conclusion that cutting energy usage will cause us to die off in the billions.
You heard it here first. The warmers can't even imagine how their policies could ever hurt people. In their eyes, its patently unthinkable that their policy could ever have negative effects, that they couldnt even imagine even the most unlikely scenarios that would cause bad things.
The talking head of the NOAA is most definitely in on some sort of conspiracy. Hansen is the ridiculous mother fucker who was publicly saying how McIntyre was just out to get him, that his data was perfect, that McIntyre's observation that there was a problem with the NOAA data sets was unscientific bullshit.
A year later the NOAA was backtracking and thanking McIntyre for finding Hansens fucking retarded mistakes.
Lets not mention the DECADE of statistical studies that were all based on that bad data, that are still taken as gospel.
Let me repeat that. The NOAA had been passing out bad data to climate scientists, data that greatly exaggerated the recent warming. Climate scientists used this data to support the AGW and Global Warming theories. Two years ago, after a decade of this bad data being used, it was discovered how bad it was. But all of the studied based on the bad data are STILL CITED AS PROOF OF AGW.
Hansen of course claims that the bad data was "inconsequential" and that the results of those studies would be the same with the good data, even though they had to admit that 1998 is NOT actually the warmest year on record. The hottest year on record is actually 1934. Thats right. 1934. Plenty of hot years back then, as it turns out.
These guys are hacks with an agenda.
It should be pointed out than in any 3+ person game, the optimal strategy is [B]always[/B] to team up. Strive to be in the strongest coalition, destroy the opposition to it, and when dominance is inevitable start in-fighting trying to break off the weakest pieces of the coalition and destroy them to, until its a 2-player game.
Pascal just has a more verbose syntax and C has looser type-coercion rules.
The two reasons that I try to use neither, especially the bit about C's non-explicit type promoting. It eventually leads to a hard-to-find bug in any non-trivial project.
But languages like VB and PHP practically encourage it.
How? Its one thing to say it, and another thing to justify it. In my experience, VB has all the same basic design-freedoms that the C language has (functions, structures, a wide assortment of branching instructions), but also has classes and mostly-enforced garbage collection.
I am wondering why you've picked VB as the language easy to make a mess of things in, when quite clearly its even easier in C.
BASIC is a common language for business applications because 30 years ago people who took a programming course learned this language. You couldn't buy a desktop that didnt come with some sort of BASIC (for many computers it was in ROM), so it made sense to teach it. So there the company is, with many employees but no professional programmers on staff.. but Bob from accounting listed his BASIC programming credentials on his resume. So lets give Bob a few days before crack at it before we go looking for an expensive alternative like contracting the work out.
Twice now I've suggested rewrites from the ground up, arguing that it would take me less time to duplicate the app but with easy capacity for adding features
That program probably did not start as something worthy of a Design Monument.. that thing of beauty that gets the author to pat himself on the back. Also remember that in almost any language, its easier to write code than to read it. Mature code ends up looking ugly, but all those operations have or had a real purpose. That funny looking bit whose purpose you can't understand, thats probably a bug fix and you too will need to do something like it somewhere.. and it too will look just as ugly.
Its easier to write code than to read it. Design Monuments don't last in the real world.
I too was sure that Modula-2 would be something good.. and the reason is because the creator of the language had lots of experience creating languages.
Later I came to realize that NONE of his languages really caught on outside of academic circles or niche markets.
His languages:
PL/0, Euler, Algol W, Pascal, Modula, Modula-2 and Oberon.
So it makes perfect sence that if the Ant got to where he thought the nest was ...
With Obama, any criticism at all is viewed as a racial attack. The "tea baggers" are called racists, for example.
It used to be that calling someone black was a device used to undermine, but NOW the device most used is to call someone racist. Need I remind us all that a white man sat on capital hill calling bankers racists for not lending to those poor disadvantaged black people, and that this ultimately lead to the credit collapse.
In many cases they do leave the higher price up. The original price can only be had via coupons.
Its a no-brainer here. Apply simple supply and demand to a situation where a referrer site is very successful at bringing in lots of customers.