Skip to main content
Corrections
Source Link
Salvo Tringali
  • 10.5k
  • 2
  • 29
  • 64

I'd like to prove that we can get rid of Conditions 2 and 4 from your listNot an answer, but a question based on a previous attempt of wishesfinding a solution, for then,which Benjamin Steinberg pointed out a foolish mistake in order to conclude, it will be enough to consider your preferred finite commutative band and extend it, if necessary, according to the stepscomments below.

For, suppose that we have somehow built Is there an "enlarge-and-shrink recipe" to extend a semigroup $\mathbb A = (A, \cdot)$, either finite or infinite, for which Conditions 1 and, 3 and 4 hold true. First, we unitize $\mathbb A$ by forcing an identity even if $\mathbb A$ is already unital. It is seen that the outcome is still to a larger semigroup, let me denote it by $\mathbb A^{(1)}$$(S, \cdot)$ for which Conditions 1 and, 3 and 4 continue to be true (unless I'm missing something). More interestingly, we have a gainbut in the process, because both of the regular representations of $\mathbb A^{(1)}$ are now faithfuladdition (in the sense of your definition). This shows that we can assume from$S^2 \ne S$? If the outset that $\mathbb A$answer is a monoid fulfilling Conditions 1yes, 3 and 4 (as forthen the latter, evenproblem is solved in a stronger form).

At this point, pick an element $\infty \notin A$ and extend $A$ to the set $S := A \cup \{\infty\}$ and $\cdot$ to a binary operation on $S$, here still denoted bypositive: Start with your preferred self-dual band $\cdot$$\mathbb A$, unitize it by taking $\infty \cdot \infty := e$ and $\infty \cdot x := x \cdot \infty := x$ for $x \in S$, where $e$ is theadjoining an identity ofonly if $\mathbb A$. Then, $(S, \cdot)$ is still a semigroup for which Conditions 1, 3 and 4 are verifiednot already unital (again, unless I'm missing something), but in addition $S \cdot S = A \subsetneq S$.

Added later. As forin such a concrete example (based onway that the above)unitization is still a self-dual band, but we can just considerhave a set $S$ with three elements $1$, $e$ and $\infty$ and let $\cdot$ begain in the binary operation on $S$ given byprocess, since now the following (Cayley) table:

$$\begin{array}{c|ccc} \cdot & 1 & e & \infty \\ \hline 1 & 1 & e & 1 \\ e & e & e & e \\ \infty & 1 & e & e \end{array}.$$ The recipeoutcome is: We start with a semigroup whose regular representations are both faithful, regardless as to whether or not this was already the trivial groupcase with carrier $\{1\}$, then we unitize it by forcing a new identity $e \ne 1$$\mathbb A$), and lastly we adjoin a further element $\infty \notin \{1, e\}$ and extendfinally use the multiplication table by taking $\infty \cdot \infty := e$ and $\infty \cdot x = x \cdot \infty = x$ for $x = 1$ or $e$enlarge-and-shrink recipe to conclude.

I'd like to prove that we can get rid of Conditions 2 and 4 from your list of wishes, for then, in order to conclude, it will be enough to consider your preferred finite commutative band and extend it, if necessary, according to the steps below.

For, suppose that we have somehow built a semigroup $\mathbb A = (A, \cdot)$, either finite or infinite, for which Conditions 1 and 3 hold true. First, we unitize $\mathbb A$ by forcing an identity even if $\mathbb A$ is already unital. It is seen that the outcome is still a semigroup, let me denote it by $\mathbb A^{(1)}$ for which Conditions 1 and 3 continue to be true (unless I'm missing something). More interestingly, we have a gain in the process, because both of the regular representations of $\mathbb A^{(1)}$ are now faithful (in the sense of your definition). This shows that we can assume from the outset that $\mathbb A$ is a monoid fulfilling Conditions 1, 3 and 4 (as for the latter, even in a stronger form).

At this point, pick an element $\infty \notin A$ and extend $A$ to the set $S := A \cup \{\infty\}$ and $\cdot$ to a binary operation on $S$, here still denoted by $\cdot$, by taking $\infty \cdot \infty := e$ and $\infty \cdot x := x \cdot \infty := x$ for $x \in S$, where $e$ is the identity of $\mathbb A$. Then, $(S, \cdot)$ is still a semigroup for which Conditions 1, 3 and 4 are verified (again, unless I'm missing something), but in addition $S \cdot S = A \subsetneq S$.

Added later. As for a concrete example (based on the above), we can just consider a set $S$ with three elements $1$, $e$ and $\infty$ and let $\cdot$ be the binary operation on $S$ given by the following (Cayley) table:

$$\begin{array}{c|ccc} \cdot & 1 & e & \infty \\ \hline 1 & 1 & e & 1 \\ e & e & e & e \\ \infty & 1 & e & e \end{array}.$$ The recipe is: We start with the trivial group with carrier $\{1\}$, then we unitize it by forcing a new identity $e \ne 1$, and lastly we adjoin a further element $\infty \notin \{1, e\}$ and extend the multiplication table by taking $\infty \cdot \infty := e$ and $\infty \cdot x = x \cdot \infty = x$ for $x = 1$ or $e$.

Not an answer, but a question based on a previous attempt of finding a solution, for which Benjamin Steinberg pointed out a foolish mistake in the comments below. Is there an "enlarge-and-shrink recipe" to extend a semigroup $\mathbb A = (A, \cdot)$ for which Conditions 1, 3 and 4 hold true to a larger semigroup $(S, \cdot)$ for which Conditions 1, 3 and 4 continue to be true, but in addition $S^2 \ne S$? If the answer is yes, then the problem is solved in the positive: Start with your preferred self-dual band $\mathbb A$, unitize it by adjoining an identity only if $\mathbb A$ is not already unital (in such a way that the unitization is still a self-dual band, but we have a gain in the process, since now the outcome is a semigroup whose regular representations are both faithful, regardless as to whether or not this was already the case with $\mathbb A$), and finally use the enlarge-and-shrink recipe to conclude.

Added a three element example
Source Link
Salvo Tringali
  • 10.5k
  • 2
  • 29
  • 64

I'd like to prove that we can get rid of Conditions 2 and 4 from your list of wishes, for then, in order to conclude, it will be enough to consider your preferred finite commutative band and extend it, if necessary, according to the steps below.

FirstFor, suppose that we have somehow built a semigroup $\mathbb A = (A, \cdot)$, either finite or infinite, for which Conditions 1 and 3 hold true. First, we unitize $\mathbb A$ by forcing an identity even if $\mathbb A$ is already unital. It is seen that the outcome is still a semigroup, let me denote it by $\mathbb A^{(1)}$ for which Conditions 1 and 3 continue to be true (unless I'm missing something). More interestingly, we have a gain in the process, because both of the regular representations of $\mathbb A^{(1)}$ are now faithful (in the sense of your definition). This shows that we can assume from the outset that $\mathbb A$ is a monoid fulfilling Conditions 1, 3 and 4 (as for the latter, even in a stronger form).

At this point, pick an element $\infty \notin A$ and extend $A$ to the set $S := A \cup \{\infty\}$ and $\cdot$ to a binary operation on $S$, here still denoted by $\cdot$, by taking $\infty \cdot \infty := e$ and $\infty \cdot x := x \cdot \infty := x$ for $x \in S$, where $e$ is the identity of $\mathbb A$. Then, $(S, \cdot)$ is still a semigroup for which Conditions 1, 3 and 4 are verified (again, unless I'm missing something), but in addition $S \cdot S = A \subsetneq S$.

EditsAdded later. SorryAs for a concrete example (based on the above), I was writing while Boris posted his answerwe can just consider a set $S$ with three elements $1$, $e$ and $\infty$ and let $\cdot$ be the binary operation on $S$ given by the following (Cayley) table:

$$\begin{array}{c|ccc} \cdot & 1 & e & \infty \\ \hline 1 & 1 & e & 1 \\ e & e & e & e \\ \infty & 1 & e & e \end{array}.$$ The recipe is: We start with the trivial group with carrier $\{1\}$, then we unitize it by forcing a new identity $e \ne 1$, and lastly we adjoin a further element $\infty \notin \{1, e\}$ and extend the multiplication table by taking $\infty \cdot \infty := e$ and $\infty \cdot x = x \cdot \infty = x$ for $x = 1$ or $e$.

I'd like to prove that we can get rid of Conditions 2 and 4 from your list of wishes, for then, in order to conclude, it will be enough to consider your preferred commutative band and extend it, if necessary, according to the steps below.

First, suppose that we have somehow built a semigroup $\mathbb A = (A, \cdot)$, either finite or infinite, for which Conditions 1 and 3 hold true. First, we unitize $\mathbb A$ by forcing an identity even if $\mathbb A$ is already unital. It is seen that the outcome is still a semigroup, let me denote it by $\mathbb A^{(1)}$ for which Conditions 1 and 3 continue to be true (unless I'm missing something). More interestingly, we have a gain in the process, because both of the regular representations of $\mathbb A^{(1)}$ are now faithful (in the sense of your definition). This shows that we can assume from the outset that $\mathbb A$ is a monoid fulfilling Conditions 1, 3 and 4 (as for the latter, even in a stronger form).

At this point, pick an element $\infty \notin A$ and extend $A$ to the set $S := A \cup \{\infty\}$ and $\cdot$ to a binary operation on $S$, here still denoted by $\cdot$, by taking $\infty \cdot \infty := e$ and $\infty \cdot x := x \cdot \infty := x$ for $x \in S$, where $e$ is the identity of $\mathbb A$. Then, $(S, \cdot)$ is still a semigroup for which Conditions 1, 3 and 4 are verified (again, unless I'm missing something), but in addition $S \cdot S = A \subsetneq S$.

Edits. Sorry, I was writing while Boris posted his answer.

I'd like to prove that we can get rid of Conditions 2 and 4 from your list of wishes, for then, in order to conclude, it will be enough to consider your preferred finite commutative band and extend it, if necessary, according to the steps below.

For, suppose that we have somehow built a semigroup $\mathbb A = (A, \cdot)$, either finite or infinite, for which Conditions 1 and 3 hold true. First, we unitize $\mathbb A$ by forcing an identity even if $\mathbb A$ is already unital. It is seen that the outcome is still a semigroup, let me denote it by $\mathbb A^{(1)}$ for which Conditions 1 and 3 continue to be true (unless I'm missing something). More interestingly, we have a gain in the process, because both of the regular representations of $\mathbb A^{(1)}$ are now faithful (in the sense of your definition). This shows that we can assume from the outset that $\mathbb A$ is a monoid fulfilling Conditions 1, 3 and 4 (as for the latter, even in a stronger form).

At this point, pick an element $\infty \notin A$ and extend $A$ to the set $S := A \cup \{\infty\}$ and $\cdot$ to a binary operation on $S$, here still denoted by $\cdot$, by taking $\infty \cdot \infty := e$ and $\infty \cdot x := x \cdot \infty := x$ for $x \in S$, where $e$ is the identity of $\mathbb A$. Then, $(S, \cdot)$ is still a semigroup for which Conditions 1, 3 and 4 are verified (again, unless I'm missing something), but in addition $S \cdot S = A \subsetneq S$.

Added later. As for a concrete example (based on the above), we can just consider a set $S$ with three elements $1$, $e$ and $\infty$ and let $\cdot$ be the binary operation on $S$ given by the following (Cayley) table:

$$\begin{array}{c|ccc} \cdot & 1 & e & \infty \\ \hline 1 & 1 & e & 1 \\ e & e & e & e \\ \infty & 1 & e & e \end{array}.$$ The recipe is: We start with the trivial group with carrier $\{1\}$, then we unitize it by forcing a new identity $e \ne 1$, and lastly we adjoin a further element $\infty \notin \{1, e\}$ and extend the multiplication table by taking $\infty \cdot \infty := e$ and $\infty \cdot x = x \cdot \infty = x$ for $x = 1$ or $e$.

Completed a previous, incomplete answer
Source Link
Salvo Tringali
  • 10.5k
  • 2
  • 29
  • 64

This is not a complete answer, but it is too long for a comment. I'd like to prove that we can get rid of Conditions 2 and 4 from your list of wishes, for then, in order to conclude, it will be enough to consider your preferred commutative band and extend it, if necessary, according to the steps below.

ForFirst, suppose that we have somehow built a semigroup $\mathbb A = (A, \cdot)$, either finite or infinite, for which Conditions 1 and 3 hold true. First, we unitize $\mathbb A$ by forcing an identity even if $\mathbb A$ is already unital. It is seen (unless I'm missing something) that the outcome is still a semigroup, let me denote it by $\mathbb A^{(1)}$, for which Conditions 1 and 3 continue to be true (unless I'm missing something). More interestingly, we have a gain something in the process, because both of the regular representations of $\mathbb A^{(1)}$ are now faithful (in the sense of your definition). This shows that we can assume from the outset that $\mathbb A$ is a monoid fulfilling Conditions 1, 3 and 4 (as for the latter, even in a stronger form).

At this point, pick an element $\infty \notin A$ and extend $A$ to the set $S := A \cup \{\infty\}$ and $\cdot$ to a binary operation on $S$, here still denoted by $\cdot$, by taking $\infty \cdot \infty := e$ and $\infty \cdot x := x \cdot \infty := x$ for $x \in S$, where $e$ is the identity of $\mathbb A$. Then, $(S, \cdot)$ is still a semigroup (again, unless I'm missing something), for which Conditions 1, 3 and 4 are verified (again, unless I'm missing something), but in addition $S \cdot S = A \subsetneq S$.

EditEdits. Sorry, I was writing while Boris posted his answer.

This is not a complete answer, but it is too long for a comment. I'd like to prove that we can get rid of Conditions 2 and 4 from your list of wishes.

For, suppose that we have somehow built a semigroup $\mathbb A = (A, \cdot)$, either finite or infinite, for which Conditions 1 and 3 hold true. First, we unitize $\mathbb A$ by forcing an identity even if $\mathbb A$ is already unital. It is seen (unless I'm missing something) that the outcome is still a semigroup, let me denote it by $\mathbb A^{(1)}$, for which Conditions 1 and 3 continue to be true. More interestingly, we gain something in the process, because both of the regular representations of $\mathbb A^{(1)}$ are now faithful (in the sense of your definition). This shows that we can assume from the outset that $\mathbb A$ is a monoid fulfilling Conditions 1, 3 and 4 (as for the latter, even in a stronger form).

At this point, pick an element $\infty \notin A$ and extend $A$ to the set $S := A \cup \{\infty\}$ and $\cdot$ to a binary operation on $S$, here still denoted by $\cdot$, by taking $\infty \cdot \infty := e$ and $\infty \cdot x := x \cdot \infty := x$ for $x \in S$, where $e$ is the identity of $\mathbb A$. Then, $(S, \cdot)$ is still a semigroup (again, unless I'm missing something), for which Conditions 1, 3 and 4 are verified, but in addition $S \cdot S = A \subsetneq S$.

Edit. Sorry, I was writing while Boris posted his answer.

I'd like to prove that we can get rid of Conditions 2 and 4 from your list of wishes, for then, in order to conclude, it will be enough to consider your preferred commutative band and extend it, if necessary, according to the steps below.

First, suppose that we have somehow built a semigroup $\mathbb A = (A, \cdot)$, either finite or infinite, for which Conditions 1 and 3 hold true. First, we unitize $\mathbb A$ by forcing an identity even if $\mathbb A$ is already unital. It is seen that the outcome is still a semigroup, let me denote it by $\mathbb A^{(1)}$ for which Conditions 1 and 3 continue to be true (unless I'm missing something). More interestingly, we have a gain in the process, because both of the regular representations of $\mathbb A^{(1)}$ are now faithful (in the sense of your definition). This shows that we can assume from the outset that $\mathbb A$ is a monoid fulfilling Conditions 1, 3 and 4 (as for the latter, even in a stronger form).

At this point, pick an element $\infty \notin A$ and extend $A$ to the set $S := A \cup \{\infty\}$ and $\cdot$ to a binary operation on $S$, here still denoted by $\cdot$, by taking $\infty \cdot \infty := e$ and $\infty \cdot x := x \cdot \infty := x$ for $x \in S$, where $e$ is the identity of $\mathbb A$. Then, $(S, \cdot)$ is still a semigroup for which Conditions 1, 3 and 4 are verified (again, unless I'm missing something), but in addition $S \cdot S = A \subsetneq S$.

Edits. Sorry, I was writing while Boris posted his answer.

added 65 characters in body
Source Link
Salvo Tringali
  • 10.5k
  • 2
  • 29
  • 64
Loading
Rollback to Revision 1
Source Link
Salvo Tringali
  • 10.5k
  • 2
  • 29
  • 64
Loading
added 45 characters in body
Source Link
Salvo Tringali
  • 10.5k
  • 2
  • 29
  • 64
Loading
Source Link
Salvo Tringali
  • 10.5k
  • 2
  • 29
  • 64
Loading