The post Universality of interaction combinators and chemical reactions ends with the idea that Lafont universality notion, for interaction systems, may be the right one for chemical computing.
These days are strange, every one comes with some call from one of my old projects. (About new ones soon, I have so many things.) Today is even more special because there were two such calls.One of them was from what I wrote in A project in chemical computing page from april 2015. It ends with:
- If you examine what happens in this chemical computation, then you realise that it is in fact a means towards self-building of chemical or geometrical structure at the molecular level. The chemlambda computations are not done by numbers, or bits, but by structure processing. Or this structure processing is the real goal!
- Universal structure processing!
There is even this video about an Ackermann function molecular computer I forgot about.
The idea is that the creation of a real molecule to compute Ackermann(2,2) would be the coolest thing ever made in chemical computing. If that is possible then as possible as well would be an Ackermann goo made from Ackermann(4,4):
In Graphic lambda calculus and chemlambda (III) I comment again on Lafont:
- Lafont universality property of interaction combinators means, in this pseudo-chemical sense, that
the equivalent molecular computer based on interaction combinators reactions (though not the translations) works
- for implementing a big enough class of reactions which are Turing universal in particular (Lafont shows concretely that he can implement Turing machines).
In the series about Lafont interaction combinators and chemlambda (1) (2) (3), as well as in the paper version of the article Molecular computers, an effort is made to reconnect chemlambda research with much older work by Lafont. [UPDATE: I retrieved this, I forgot about it, it’s mostly chemlambda v1 to chemlambda v2, see also this post ]