[QUOTE=GIGObuster]
Quote: Originally Posted by intention
If you believe this, you’ve obviously never tried to replicate the results of a study.
- While it is sometimes possible to reverse engineer the code that they are using if their code is 100% correct, if they have made any errors it is not possible.
Give it up, that was no code but data.
[/quote]
There seem to be several misunderstandings here. The first one is that what Hanson is reporting is raw data of some sort.
It is not. Hanson’s gridded global temperature dataset is the result of a number of computer operations on individual station data. What computer operations? Well … we don’t know, do we, because Hanson won’t reveal the computer code.
[QUOTE=GIGObuster]
Quote: Originally Posted by intention
2) Without the code, some incorrect procedures may be masked by other correct procedures, and not be discoverable.
They did anyhow.
[/quote]
This is another misunderstanding. While an error was discovered, other errors (which I called “some incorrect procedures” above) may exist which have not been discovered. And without access to the code, those errors may not be discoverable. The fact that one error was discovered means nothing about the other possible errors.
[QUOTE=GIGObuster]
Quote: Originally Posted by intention
3) The lack of the code makes it very slow and difficult to discover if there are errors.
Once again, they did. And the bottom line remains: the overall conclusion showed up again.
[/quote]
Clearly, you think that since one error has been discovered, none remain. Perhaps you could let us know how you determined this.
[QUOTE=GIGObuster]
Quote: Originally Posted by intention
This all misses the central issue, however. Science is about transparency and replicability. While one error has been found, Hansen’s refusal to reveal the code makes it hard or impossible to find all of the errors that may have been made.
Hansen’s work is publicly funded, and has no commercial value or business secrets. He’s asking us to risk billions of dollars on his results, but he won’t say how he got the results.
The thing that you keep missing is that the slight error, after an attempt to find an even bigger error, ends up confirming the early research.
[/quote]
Confirming the early research? After years of saying “1998 was the warmest year ever in the US”, we now find that the warmest year in the US was 1934. Please let us know how that “confirms” the previous finding that 1998 was the warmest year.
[QUOTE=GIGObuster]
Quote: Originally Posted by intention
You keep thinking that we need to come up with reasons for him to reveal the code, but the opposite is true. He is obligated by both scientific practice and public disclosure laws to reveal the code, unless he can come up with some sound reason not to.
Perhaps you could let us know what you think that sound reason for hiding his work might be?
w.
The simplest reason: the data is available, Hansen got results, someone replicated the effort with different coding, and ended virtually confirming what Hansen got.
The point of “hiding the code” is moot and it remains a smoke screen.
[/QUOTE]
Oh, I see. To you, transparency in science means that if you hide your code, and as a result someone can only find obvious errors in your work, you are therefore justified in hiding your code … please tell me you’re not a scientist.
You see, the way this “science” thing works is that
-
A scientist uses some procedure to come up with a result.
-
The scientist publishes his procedures and data.
-
Other scientists examine his work, to see if they can find any errors in either the procedures or the results.
-
If they cannot find any errors, the work is considered to be established, until such time as may be overturned in the future.
Now, this funny thing called “science” can only work if the scientist publishes both the procedures (mathematical steps, laboratory procedures, computer code, etc.) and the data. Absent that, we are left with an apocryphal claim of unknown validity.
Your idea that if no one can find an error in work whose procedures are kept hidden, that inability to discover errors thus justifies the hiding of the procedures, is completely backwards from the way science actually works.
w.