Time is running out – literally, says scientist

[I]f time gradually slows “but we naively kept using our equations to derive the changes of the expansion with respect of ‘a standard flow of time’, then the simple models that we have constructed in our paper show that an “effective accelerated rate of the expansion” takes place.”
While the change would be infinitesimally slow from an ordinary human perspective, from the grand perspective of cosmology – in which scientists study ancient light from suns that shone billions of years ago – this temporal slowing could be easily measured.

Interesting hypothesis. Though the article mentions this as a “radical suggestion” the headline is much more certain, which has become a peeve of mine. Conjecture is one of the things that scientists do. We play a lot of “what if” games, and most of this gets shot down when we realize a conflict exists with existing observations, or someone points them out when you compare notes with a colleague. This happens a lot in the lab when faced with a novel set of data — what was going on here? Is this a problem with the equipment or some new effect? The former is much, much more likely than the latter, but until you run tests and replicate the results reliably, you are faced with a mystery, and chasing the solution is both frustrating and somewhat intoxicating. (The scariest scenario is when the anomalous signal just disappears and you can’t replicate it). But you can also do it with the models you build. What happens if a particular term in an equation is or isn’t small, when the opposite usually holds? What if there is some additional effect? You play with the equations and see where it goes.

If you don’t get tripped up by these “slain by an ugly fact” obstacles, you can formulate a model that could possibly be tested. Eventually, you present it for others so that they, too can comment on and think about it. Many of these ideas never pan out, at each stage of this evolution and distillation. These authors have an idea that has gotten to this point. It doesn’t appear it has yet been rigorously tested to see of it holds up — one needs to know what specific predictions it makes that distinguish it from the current models. It’s just not obviously wrong after having had some level of scrutiny.

I think that the headline editors and journalists do a disservice when they attach much more certainty to (in this case) as-yet untested ideas that show up in the journals, or any single peculiar experimental result that pops up.