Fossil shows mammal sinking teeth into dinosaur

I'm pretty sure my little dog would do something similar if she encountered a wild boar in the forest. I'm quite sure that if I were successfully chased down by a velociraptor, I would not just lay down and wait to be eaten.

> Researchers said the discovery challenged a long-held view of early mammals as “fodder” for dinosaurs.

I don't really get how that "challenges" any view at all. As if dinosaurs never got bitten nor scratched by stuff they were trying to eat. And with an erupting volcano nearby, I'd probably fight harder just for the hell of it and to mess with the palaeontologists.

1. Science as an institution of academics, some of whom have demonstrably prioritized their career over doing their job (and even when that's not the case, has been shown to be wrong if you go back far enough to various scientific theories that were disproved by later discoveries), and...

2. Science as a process of inquiry that begins with an observation which leads to a hypothesis which is then put to the test through experimentation. This needn't be a process limited to academic science, either, but can be exercised in many areas of life.

I suspect we all generally agree that Science #1 should usually be taken with some salt (even in the best scenario, once you take science at its word, you've stopped doing science, though there are times, like gravity, when for practical purposes you can accept a high degree of technically-non-perfect certainty is close enough), and Science #2 is generally good for either reinforcing or correcting our understanding of the world.

I'm not anti-science. I'm saying we screwed up bigtime trusting people within the institutions we put up to manage science. Especially in the soft sciences.

We let things go without serious peer review or replication far too long, and allowed far too dodgy statistics, resulting in the replication crisis we see today. Science is amazing, but we haven't been doing science in certain areas for a number of years. We're going to be sorting this out for at least the next decade. We should start again at some point, and that starts with clearing out all the crank theories that have made their way into journals that either don't replicate or have no predictive power.

We shouldn't have been relying on trust in the beginning. We should have been doing replications.

"The victim of its attack was Psittacosaurus lujiatunensis, a bipedal, plant-eating beaked dinosaur"

It doesnt fully exclude the "random attack" behaviour, but makes it more unlikely.