Defect theorems with compatibility relations. (Q2480767): Difference between revisions

The well known `defect theorem' of words states that if a set of \(n\) words in the free semigroup \(A^+\) generated by the letters of an alphabet \(A\) satisfies a nontrivial relation, then these words can be expressed simultaneously as products of at most \(n-1\) words. Putting restrictions on the \(n-1\) words give variants of the defect theorem. Another formulation of the defect theorem is the following: Let \(X\subseteq A^+\) be a finite set of words and let \(B\) be the base of the `free hull' of \(X\), that is, the smallest free semigroup containing \(X\). Then \(|B|\leq|X|\) (where \(|B|,|X|\) denote the cardinalities of \(B,X\) respectively), and \(|B|=|X|\) if and only if \(X\) is a code. In this paper, the authors study defect theorems with `word relations' \(R\subseteq A^+\times A^+\). More specifically, they generalize the above mentioned concepts with respect to these relations \(R\), which are reflexive and symmetric relations induced by their restrictions on the letters of \(A\). These word relations were introduced by the authors in order to generalize the notion of `partial word'.