0

In one of the works about tree languages I met the following definition:

Let $\mathcal{N}$ be the set of nonnegative integers and let $U$ be the free monoid generated by $\mathcal{N}$ with operation $\cdot$ and identify $0$. Let $a\leq b$ for $a,b\in U$ iff $\exists x\in U\ni a\cdot x=b$, let $a$ and $b$ incomparable iff $a\nleq b$ and $b\nleq a$, and let $a<b$ if $a<b$ and $a\neq b$.

This is exact citation from the paper (from ethical reasons I do not know should I give exact link to it or not).

It seems to me that there are 3 inaccurates in this definition. I just want to be sure that it is not my misunderstanding, but possible misprint in the paper. Could you please confirm my assumption or explain why it is wrong?

  1. It should be not "indetify" but "identity"

  2. The definition of $a\leq b$ implies that $a\neq b$, because if $x$ exists, then $a\cdot x\neq a$?

  3. The definition of the strict inequality $a<b$ seems to me totally confusing. It probably should be: $a<b$ iff $a\leq b$ and $a\neq b$, but according to p.2 it seems to be useless.

  • 1
    There is a further inconsistency in that they have $0$ be one of the generators of the free monoid but also have it as the identity. – Tobias Kildetoft Jul 27 '17 at 18:53
  • @TobiasKildetoft would you please be so kind and expand this point, may be as answer (I will definitely +1 it), because it is something that I do not quite understand. – Andremoniy Jul 27 '17 at 18:55
  • There is not that much to expand on. The free monoid on a set is the set of all finite strings of symbols from that set. The identity is the empty string, which is in particular not an element of the original set. – Tobias Kildetoft Jul 27 '17 at 18:56
  • May be they meant that it is a free monoid with* identity $0$ and generated with operation $\cdot$*? – Andremoniy Jul 27 '17 at 18:58
  • @TobiasKildetoft Could you please give a clue, what would be correct form of this definition regarding identity? – Andremoniy Jul 27 '17 at 19:05

1 Answers1

1

(1) Yes, it should say "identity".

(2) No, because $x$ could be the identity element $0$. In that case, $a\cdot x=a$.

(3) Yes, it should say "let $a<b$ if $a\leq b$ and $a\neq b$".

Eric Wofsey
  • 330,363