At my Wikipedia-minus-math level of understanding, the problem with any resolution of "Olbers' Paradox" through a fractal distribution (such as the "Cantor set" depicted in the Wikipedia article of that name) of stars or star clusters, rather than the alternative of a beginning of our multiverse (or universe) in time, is such a distribution's fractal dimension of approximately 0.6, rather than the dimension of 2 or more that is reportedly favored by most cosmologists. It seems to me that the favored dimension implies some preferred size for the proportion of the multiverse taken by our observable region, which definitely does not seem to be supported by observation. If this lack of support does leave such an alternative to any beginning open, a secondary problem might be the improbability of such fractally linear alignments of star clusters. (I'm assuming that they'd be even larger than the clusters of galaxies whose relations would reflect universal increases in the amounts of space between them.) In that case, might the pressure or other averaged effects of gravitational or other radiation on the evolution of such clusters,or of such components of them as the binary systems accounting for a large proportion of observed stars, perhaps account for such linear arrangements of star clusters over the periods of time during which they would form across larger regions of our local universe than the observable one, or would some anthropomorphic factor have to account for them?