The Ghadiri peptide, at 32 amino acids, is the smallest known self-replicating peptide (AFAIK). Your response is to complain that its replication is not general enough. <shrug>
Of course, when the replication essentially consists of the formation of one single bond. That is no "replication". That is a tinker-toy.
Here is an opinion on these type reactions.
If you want to get two molecules to react, positioning them in the right way could get you there quite efficiently. Using this simple chemical insight and some clever retrosynthetic thinking, Reza Ghadiri and his group at the Scripps Research Institute, La Jolla, California, created a completely new research field some six years ago. Knowing an existing biological structure, eg the coiled coil (two a-helices wrapped around each other), they described one strand of it as the template. This strand helps the positioning of the other one, which we could call the target strand. They cut the target strand in half and activated the ends so that the halves can react to form the whole. Then they demonstrated that the template can serve to position the halves, thus speeding up their reaction to form the complete target strand.
...
The availability of such peptides provides a unique opportunity to study complex molecular behaviour in a simple system. Some of the processes involved will be spookily reminiscent of things happening in the living cell. For example, Ghadiri's group reported the emergence of 'symbiosis' - two distinct self-replicators enhancing each other's success - in their early work on peptide hypercycles.1 However, one should not be tempted to transfer these findings to the still largely mysterious field of the origin and pre-cellular evolution of life. Biological molecules do not actually replicate themselves, but rather replicate each other. Furthermore, most researchers see RNA as a more promising candidate for the principal role in the early molecular stages of evolution. Thus, self-replicating peptides may have little to teach us about the roots of the tree of life, but they do add some interesting new branches to the tree of chemistry.
To overcome this fundamental problem, Chmielewski's group has now systematically reduced the stability of the coiled-coil complex by as much as they could without endangering the binding of the fragments necessary for positioning. They achieved this goal by shortening their self-replicating peptide E1E2 to a length of 26 residues, which they believe to be minimal for the desired reaction to occur.3 Studying the self-replicating capacity of the new peptide, called RI-26, they observed catalytic efficiency (catalysed rate constant:uncatalysed rate constant) of 100,000, which is more than 20 times higher than the previous record for self-replicating molecules. This efficiency approaches the range found in natural enzymes.
Andrew is of the opinion that things are not worth knowing unless you can know tham right now. Biogenesis is not worth pursuing because it might take decades, or even centuries. In the meantime he can sit on the sidelines and snipe.
On a completely unrelated subject, the Japanese are inventing completely unexpected metal alloys, products no one thought possible. You might think this a bit ironic.