The BBC reports (shockingly), that the journal Nature is going to begin requiring a reproducibility checklist of authors, based on a survey performed last year where at least 70% of respondents (self-selected, of course) indicated that they were unable to reproduce expected results. As the ability to replicate studies is what allows science to demonstrate meaningfulness and continue moving the body of knowledge forward, it is surprising that it has taken this long for top of the line journals to more strongly encourage replication to establish validity.
Wha wha wha whaaaatttt ? How DARE they apply the scientific method to settled science ! Blasphemy !
I think AGW just got bitch-slapped.
Here is an interesting (though technical) article on running climate models.
https://chiefio.wordpress.com/2017/01/18/gcm-modele-elephants-playing-mozart/
The money quote:
“Never The Same Code Twice
First off, it follows in the tradition of GISStemp where every time you run the program
it is recompiled. This means it is never the same code twice.”
Then follows a discussion of why that is and about all the parameters that can get tweaked…
The new policy from _Nature_ that all results must be reproducible means they can never
report on Climate Models!
–Bad News
Free at last! Free at last! Thank God Almighty science is free at last!
“…it is surprising that it has taken this long for top of the line journals to more strongly encourage replication to establish validity.”
It is only surprising if you forget that they are PUBLISHERS, not scientists. They get paid based on how many people buy the journal, not how good the science is.
Here is an interesting (though technical) article on running climate models.
Whoever wrote that drivel is dangerously ignorant of how computers work. Trust nothing he says.
You wrote:
“Whoever wrote that drivel is dangerously ignorant of how computers work. Trust nothing he says.”
You need to be more specific. The article is full of very good advice about how to avoid software traps. Where is the “drivel”?
The author writes “I ran a supercomputer center for 7+ years…” and I believe him, based on my 30+ years in IT. What is your background?
–Bad News
@Bad News is correct. Labeling something as drivel, without substantiating your opinion, is…… drivel.
I don’t know what a .pm file type is, so I looked inside the RegressionTools one. It looks like a bunch of parameter setting. (Is pm short for parameters?)
Immediately disqualified from being taken seriously about *nix programming.
There is literally not a single paragraph in that blog post that does not contain ample evidence the author knows only enough to be dangerous.