I've been thinking about the synchronisation issue, and the comment that we need (coal-based) baseload to provide a steady 50Hz for the renewables to hook in to.
My gut feeling was that providing synchronisation in a highly distributed system without coal is certainly more complex, but not impossible.
So I did a bit of looking into the current technical articles on this matter, and found a squillion (not a technical term) articles looking in to this exact thing.
So, basically, as long as there is enough generation and storage so that the intermittent nature of wind, solar etc is not an issue, synchronisation shouldn't be an issue either.
This is a fairly short one (3 pages) that seems to explain the issues and some basic definitions, as well as comparing a couple of techniques. For those with some technical understanding.
ABSTRACT Distributed Generation (DG) System based on renewable energy sources provides a good substitute to the centralized power generation systems which consumes fossil fuels and has imposed several problems on the environment. The safe operation of DG systems requires proper synchronization with power grids by using power electronics converters. The use of Phase Locked Loop (PLL) is an important technique for grid synchronization along with control optimization. This paper presents a comparative study between Dual Synchronous Reference Frame (DSRF) PLL based on synthesis circuit and Second Order Generalized Integrator (SOGI) based PLL. The comparative study has been made on the basis of how they track the phase angle during different abnormal grid conditions like voltage sag, line to ground fault and frequency deviations. The results are obtained from MATLAB/SIMULINK environment and are analyzed for feasibility of the study.

