[CraigS]

Originally Posted by renormalised

However, it's also equally correct to say that the authors themselves should be careful not to be so sure of their own conclusions, based on their mathematics and their assumptions. Maths can be made to say and prove anything.

Originally Posted by marki

I would say the opposite is true here. It's people who do know what they are doing that make it correct, just check out any of the latest political fights, both sides always seem to be backed by irrefutable stats based on in depth studies by proffessional bodies. How can they both be right when they directly oppose each other? The old saying lies, dam lies and statistics still holds true. Mathematics like any piece of written work is always open to bias and manipulation.

The entire point of this paper, is to provide an analysis to highlight the sensitivity of the outcome to the initial assumptions. The conclusion is, necessarily, that choice of initial assumptions, (in this case the probability parameter of relatively short-span abiogenesis), has the dominant influence on the outcome.

Further, they restate the commonly held fact, (made so by abundant empirical evidence), that life began on this planet fairly soon after the Earth became habitable. This is then, is entirely consistent with a low probability of abiogenesis. Using this as a basis for another iteration of the Bayesian calculation, necessarily drives the conclusion of life being arbitrarily rare in the Universe.

I find the application of the cliches like "lies, dam lies and statistics", and inferences of "mathematics being manipulated for a particular outcome", as small-valued opinions, adding little weight to a meaningful scientific discussion.

Bayesian inference in statistics is specifically used to estimate if a hypothesis is true. It makes use of iterations when fresh evidence (raw data) appears, to repeatedly modify the initial hypothesis. This technique carries a lot of weight in scientific probability estimation circles, as it is clearly designed to expose the dependency of outcomes on initial assumptions, for all to see. (Ie: falsifiability incarnate).

The authors specifically make the point that a discovery of exo-life would alter the outcome of future iterations of the Bayesian Inference (due to the sensitivity to initial assumptions). Like it or not, this is exactly the butterfly effect in action.

I for one, choose to not dismiss outright, a value-laden analysis, purely because of prior opinionated bias, coming form some prior belief system.