> our vulnerability to the peer review process and how a storied journal retracts its support of an article

A word in everyone's ear from the physics wing.

The "peer review" process here has a limited ability to referee what is claimed to be a one-off report of a sighting of a mythical beastie in the wild.  Either one prints it or not, as superconductors at room temperature are certainly possible, but actually baking one that truly works has turned out to be a trial-and-error process, more akin to the alchemists of old, than the kind of work to which we are accustomed.  (That said, a lot more progress has been made by people saying "WTF?" than has ever been made by people shouting "Eureka!", so even the failures are well-worth documenting.

"Nature" is of course, not a magazine, but a refereed journal that man people would kill to get published in.

Materials science is incredibly complex.  Man already knows virtually "everything" about the discovered elements, and the holes in the periodic table are there precisely because we know which elements have yet to be manufactured by brute force from the properties of the subatomic particles.  All this is so well-known, it is called "the Standard Model".  

BUT - things like superconductors are always alloys - and alloys can behave in very unexpected ways.  So, a new superconductor, such as the OTHER newly-announced but they retracted room-temperature superconductor of 2023 (yes, there were two) named "LK-99", which turned out to have a less-then consistent conductivity across a range of temperatures when  other labs tried to make the same alloy (or amalgamation), so the "dose/effect" was so variable as to be clearly an error.  That also led to a retraction.

In the case of the "Dias" paper, there were multiple co-authors, and THEY called for the paper to be retracted, because the final version of the paper "as published" went to far in its claims in their views.  This points out the fact that in today's world, there are no solo researchers that will ever be able to accomplish much in any field.  Teams are needed.  Technicians are needed, often even specialists in pure statistical analysis to make sure that the data gets handled appropriately.   Not all the members of the team participate in writing the paper.  In this case, the team as a whole called foul on the "writer" members of the team.  Yes, Dias has a (ahem) colorful reputation, but there was a larger team, all well-respected and with excellent track records, so there were no raised eyebrows until the co-authors saw the paper in genuine print.

This means that the system worked just fine, as no one wants to have their good family name associated with anything lacking a solid supporting dataset with rigorous statistics. 

In the area of superconductors, moreso claimed "room temperature" superconductors, no one takes much seriously until someone else can also make it work.  It’s a lot like seeing a comet - first you call another observatory, and ask them to point at the coordinates and verify the location, vectors, and velocity.

To compare and contrast, Google and read about "FOGBANK", a still-classified material critical to the US nuclear weapons program.  We seem to "forget" as a country how to make this very important stuff sometime during the nuclear test-ban treaty.  It should be obvious to even the casual observer that the reason we "forgot" is not that anyone lost the recipe, but that the more recent attempts to make the stuff following the same process failed miserably.  The reason was simple - more modern efforts resulted in more "pure" ingredients", and the "purity" meant that a contaminant was left out.  The contaminant turned out to be a mission-critical ingredient.

This kind of thing is happening all the time.  

The only more complex and "messy" branch of science I can think of is hydrology - the Army Corps of Engineers has a group called ERDC (http://erdc.usace.army.mil), were they still build scale terrain models of rivers and dams and levees at a facility in Vicksburg, MS because we simply do not know enough to be able to "model" any river or stream in a computer simulation well enough to predict floods accurately, so the precision scale models, when fed scale amounts of water, accurately reflect the behavior of real rivers.  These same folks ran a scale model of the entire Mississippi river in Jackson MS that covered 200 acres, and saved thousands of lives by being able to predict where water would be highest, and focus sandbag placement, and keep the river from flooding so much.  That model was abandoned in the 1990s as it was replaced with a set of smaller more local models.

But water - we still do not understand moving water very well.  The further one gets into science, the more things one finds that are indistinguishable from pure magic.  Like... what the water does in any river or stream you care to name.  We dunno.

             ***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software.  For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html