The role of formal scientific processes in climate science appear to be under threat as never before.
Last year, physicist Prof Richard Muller and colleagues published - in the sense of posting material on their website - results from a new project analysing the Earth's temperature record.
The Berkeley Earth (BEST) project basically backed up established temperature records from Nasa and others; the world is indeed warming, and by about as much as we previously thought, it concluded.
Prof Muller was attacked in some quarters for not waiting for the formal process of peer review in a scientific journal before launching the data publicly.
He responded that his method - to put the draft out there openly and let everyone respond who wants to - is increasingly the norm in physics and indeed has always been the norm in string theory, that most arcane of disciplines.
In his view, it's the right way to do things.
A couple of weeks ago, in a New York Times article accompanying the release of five more BEST papers that are being submitted to scientific journals, Prof Muller went further, saying that the majority of 20th Century warming could be laid at the door of greenhouse gas emissions.
By contrast, analysis by established bodies such as the Intergovernmental Panel on Climate Change (IPCC) holds that only after mid-Century did greenhouse gases drive the warming - prior to that, it was predominantly down to natural causes such as solar cycles and a decline in the frequency of large volcanoes.
The original BEST study particularly got up the nose of meteorologist turned sceptic blogger Anthony Watts.
It dismissed the claim he'd made that US weather stations gave an unreliable temperature record because many were badly sited - in places where the extent of heat-reflecting tarmac, for example, had expanded over time.
Also a couple of weeks back, Mr Watts launched a new analysis purporting to show that BEST had it all wrong.
BEST had used an out-of-date methodology for assessing station quality, he argued; use the right one, and you find that US temperatures have risen by only half as much over the last 30 years as Prof Muller and others say it has.
This paper too has been released web-first, on the wattsupwiththat blog, with the aim of formal publication later.
It referred to a scientific paper out this week in which he calculates how the incidence of extreme weather events has changed since the middle of the last century.
Using simple statistics rather than computer models, he shows that the frequency of "extreme anomalies" - for the statistically-minded, defined as more than three standard deviations from the mean - has increased 10-fold.
Without climate change, it concludes, last year's drought in Texas and Oklahoma, the 2010 Moscow heatwave, and the 2003 heatwave centred on France wouldn't have happened.
(The article's appearance induced the journal publishing the paper, Proceedings of the National Academy of Sciences, to lift the embargo for reporters, but it doesn't appear to be on their website as yet - sometime this week, presumably.)
Prof Hansen's paper has had a mixed reaction from other researchers.
"Rather than say, 'is this because of climate change?' That's the wrong question.
"What you can say is, 'how likely is this to have occurred with the absence of global warming?' It's so extraordinarily unlikely that it has to be due to global warming."
Prof Myles Allen, the Oxford University climate modeller who has spent 10 years developing the science of climate attribution, said it was "broadly in line" with previous analyses, but that the interpretation "goes further than many scientists are comfortable with".
What's perhaps more remarkable about Prof Hansen's paper is the style.
Rarely if ever have I seen a published scientific paper that states the rationale for its existence so baldly in terms of public perception - specifically, "the need for the public to appreciate the significance of human-made global warming".
"Actions to stem emissions of the gases that cause global warming are unlikely to approach what is needed until the public recognises that human-made climate change is underway and perceives that it will have unacceptable consequences if effective actions are not taken to slow the climate change," the authors write.
You'd have to be from another planet not to realise that climate science has been the subject of extraordinarily intense political forces over the last few years.
And many scientists involved feel passionately about it.
At its core, though, climate science has been able to retain its identity partly because researchers generally don't give in to passion, instead sticking to formal processes - publication in peer-reviewed journals and the presentation of data and conclusions in strictly academic terms.
It's rapidly becoming more blurred. And the question arises: is this a good thing?
Prof John Christy, the University of Alabama scientist who has taken a position sceptical of "climate catastrophism" down the years while working in the mainstream discipline of compiling temperature records, believes it could be.
Two years ago, he suggested replacing the monolithic procedures of the IPCC with a "wiki" approach.
And he tells me now that he got involved with the Anthony Watts exercise partly because it "would be an interesting experiment for me in which the paper was 'cloud reviewed' and then rewritten to accommodate important new information before being submitted [to an academic journal]... I'm wondering if this is the way 'review' in the digital age will unfold as time goes on."
Prof Christy makes the distinction - crucial to scientists - between draft papers for discussion and final, complete ones that go into academic journals and become part of the formal literature of science.
But how clear is that distinction to the public that Mr Watts, Prof Hansen and Prof Muller are trying to influence?
And if it's not clear, how does the new model benefit public understanding?
Peer review is far from perfect - especially in a politicised arena such as climate science where some journals exist with a specific, directed slant on the issue.
Energy and Environment, for example, proclaims itself "a forum for more sceptical analyses of 'climate change'".
Creationists have attempted to clothe themselves in scientific garb down the years by establishing publications designed to look and feel like scientific journals.
The Journal of Creation, for example, says it is a peer-reviewed journal but clearly comes with a specific aim - to combat the problem that "creationists cannot publish their creationist ideas in secular journals because the evolutionary worldview has a stranglehold on scientific publishing".
Well, clearly the "evolutionary worldview" ought to dominate scientific journals - because a vast amount of evidence testifies to the fact it's real.
But you can create a parallel world where it isn't, if you really try.
With all its flaws, publication in mainstream peer-reviewed journals is the best mechanism science has yet devised for ensuring that the findings and conclusions reaching the public ear remain above a certain quality threshold.
String theorists can perhaps afford to take a different tack, because - with all due respect - it doesn't make any practical difference to anyone in the wider world who's right and who's wrong in that particular discipline.
But with climate science, it does. It matters a lot.
Is it really time to throw the traditions away? And if it is, whose interests would that serve?