Wednesday, 30 June 2010

lo.logic - Degrees many-one below 0omega

Sorry for answering my own question, but I'm still hoping for a characterization of some kind.



A partial answer: Let S be the result of a normal jump inversion forcing for 0omega with added requirements that the nth column is 0n modulo a finite amount. This yields a degree many-one below 0omega, even many-one above each 0n, whose jump is 0omega.



The reason S is many-one below 0omega is that we know that the algorithm gives uniformly an answer for the nth bit from 0n, since no higher requirements have been initialized by that point. So, running the turing algorithm from 0omega is the same as running the corresponding algorithm from 0n. The outcome is an arithmetical fact, so can be checked in one query of 0omega.



It appears that the many-one degrees below 0omega (many-one) above each 0n is a rich enough structure, since it contains [S,0omega]m.



It seems possible that similarly intertwining requirements, we can get a set whose double-jump is 0omega above each 0n.

No comments:

Post a Comment