I was just reading a 1995 paper by Lee Smolin arXiv:gr-qc/9505028 "Linking topological quantum field theory and non-perturbative quantum gravity" not such a catchy title, but contains some surprising things We all know the story about how Einstein put Lambda (cosmological constant) in the quations in 1916 and then found it embarrassing to have done so. And Lambda went into eclipse for many years. The standard story is that in 1998 it suddenly made a comeback because of supernova Type Ia data which showed accelerating expansion. then positive Lambda became necessary to make things fit the data. Cosmology underwent a revolution and came out looking quite different. Before 1998 people would put Lambda in sometimes and try the model but it wasn't showcased, it was just an optional doodad. Or so the story goes. But it turns out that Smolin developed a "spin-networks" QG approach in 1995 that absolutely depends on positive Lambda. Indeed Lambda is quantized and (in natural units c=G=hbar=1) its reciprocal has to be a positive integer divided by 6pi. You remember how the present radius of the observable universe is about 40 billion light years (about 3 times 13.8 billion). the area of that horizon is also quantized, in units of the planck area. It sounds incredible I know. The mind reels. But this paper is turning out to be an important one. The conclusions are seemingly still valid and play a role today. Both the event horizon of a black hole and the expanding boundary of the observable universe are quantized areas and the number of planck area units comprised has some information theoretic meaning. Very hard to comprehend.