There are a number of different threads here that are worth following up on. The most obvious is the question of the original formation of nucleic acids. There is equivocated evidence in favor of the view that this took place more than once; mitochondrial DNA uses a slightly different genetic code than eukaryotes use, and there is a theory called the endosymbiotic theory that is substantiated by genetic analysis showing that their ancestor was a proteobacterium, probably an alpha-proteobacterium; it may have been related to rickettsia, but this last is somewhat controversial. However, it is worth noting that even with such variations (and the mitochondrial variation is only the first among many), the code is so similar across all of Earth's species that it is unlikely to have developed in separate events.
In addition, as I have pointed out elsewhere (and I was not the first to note it), there are many possible genetic codes. How likely an alternate genetic code is to be more (or less!) prone to point defects could be a characteristic subject to natural selection; as could speed of expression, or ease of replication, or other factors that we currently know nothing of. The fact that there are any alternates at all at least hints that this and potentially (likely?) many other factors may have militated toward the code most life forms on Earth use now.
It is also important to keep in mind that the fact that we (that is, living things on Earth) use the same genetic code implies in turn that almost all primary proteins, that is, proteins created directly from amino acids by expression of the genes through transcription into mRNA and protein synthesis from that mRNA by tRNA and ribosomes, are made from the same twenty amino acids. Now, these proteins, depending on their structure, can theoretically create secondary proteins, utilizing other amino acids than the twenty; but by and large, to the best of my knowledge, this process is not used. If someone has an example, I'd love to hear about it.
Now, one of the things to keep in mind about RNA is that it functions in at least three different ways in the eukaryotic cell. First, there is mRNA, made by RNA polymerase from the DNA (and note that the RNA so created is not a duplicate of that DNA strand, but its complement- in other words, a copy of the other side (assuming that the DNA is double-stranded in that spot)). Second, there is rRNA, making up the ribosomes, the "protein factories" of the cell. Third, there is tRNA, which attracts a particular amino acid, and is in turn attracted by a particular codon (i.e., trio of bases) when it is inside the ribosome; tRNA, therefore carries the amino acids to the "factory" where each amino acid is added to the end of the growing protein molecule being "manufactured" by the ribosomes.
But it turns out if you look into it, that unlike DNA, in nature there are many alternate bases used in RNA; quite literally hundreds, in fact. It is by far the more complexly used molecule of the two nucleic acids. RNA also doesn't generally exist as single strands; the attraction for other nucleotides, and the weakness of the single strand, apparently militate against this. The strands are far more than the simple double-helix, however; they encompass the ability for RNA to form complex shapes, shapes which are capable of acting as enzymes (so-called "ribozymes") by doubling or further foldings of the strands onto one another.
So this is not some rare, docile worker molecule, with only one possible purpose; instead, it is a dynamic, active, highly common molecule that has functions woven all through cellular metabolism and reproduction. Molecular biologists are slowly coming to understand how many different roles it can take on. And this is the reason for the proposal of "RNA world." This molecule might, by itself, be capable of forming an autocatalytic network, without any proteins. Such a network would "eat" nucleotides and create new copies of its member RNA strands; this is the minimal functioning of life. I grant that this is currently only a hypothesis, but the evidence it explains is widely spread, and it is extremely plausible. The strongest two pieces of evidence are:
1. While the Urey-Miller experiments did not produce nucleotides, only amino acids, later (and probably more accurate) laboratory simulations of conditions on the early Earth (notably that of Joan Oro) did produce nucleotides.
2. Nucleotides also have been found inside meteorites. This indicates that they may have been formed in the solar nebula, or that they can form from the heat of passage through the Earth's atmosphere at the Earth's orbital speed. In either case, it is clear that they would have been present very early in Earth's history.
Finally, even if it was not RNA itself, but a precursor, such as PNA, or GNA, the above facts still work; and once you have viruses using RNA, and a virus switches to DNA to protect its genes from attack during the cell-invasion-virus-replication process, then you have cells that have DNA in them. And you can't use DNA directly to make protein, it's too delicate; but you CAN use it to make RNA, and of course the virus would have to provide a mechanism for that, encoded in RNA, if it were to survive (since the cell, of course, would have no DNA before then).
So much for "problems" with the plausibility of the RNA world hypothesis.
I'll also point out that autocatalysis among amino acids and the resulting proteins need not be affected by the presence or absence of this "RNA world;" the two need not be combined until a great deal of time had passed.
And finally, it appears that lipid spheres form spontaneously upon exposure to water. That would be why they are so important in the formation of cell membranes. Such lipid spheres are present in carbonaceous chondrites, which are meteorites with a great deal of carbon in them; they are also present in Oro's experiments.
So what do we have?
1. An abundant source of novel amino acids.
2. A source, perhaps not quite so abundant, but perhaps more so, of nucleic acids. (Oro's first major discovery was proving that adenine could be synthesized from ammonium cyanide, a common chemical found in stellar nebulae, and presumably also on the early Earth and in early comets).
3. An abundant source of lipids.
These are the most important ingredients for life. Pardon me for paraphrasing several generations of biologists, but what this looks to me like is a very nice "life soup;" put it on simmer and stir regularly for a few hundred million years.
