It makes perfect sense to me. There is indeed a division between the metabolic first hypothesis, which posits the spontaneous assembly of autocatalytic metabolic networks under the correct conditions (lots of incoming useful energy + abundant simple molecules for autocatalysis). The problem with this view is it doesn't realistically allow for natural selection (even putting aside the difficulty of achieving a biochemically plausible cycle in early earth conditions). The replication first hypothesis allows for natural selection but in turn faces the problem of how the raw material of sufficient complexity (nucleotides) to allow reliable template based synthesis could have plausibly come about. The linked to article addresses one hypothesis for one possible path for arriving at the building blocks of RNA given that early conditions were close enough to its assumptions.

I enjoy works which look for scenarios where both could have emerged simultaneously. Reading the introduction of this paper/article provides an accessible exposition to the topic: http://www.mdpi.com/1422-0067/10/4/1838/htm

If you are curious about under what conditions self-organization might more effectively dissipate entropy I suggest reading anything by Ilya Prigorine. But this short minute physics video narrated by Sean Carroll is good for a start: https://youtu.be/HxTnqKuNygE?t=2m07s