Date: Sep 18, 2013 3:39 AM
Author: William Elliot
Subject: Re: Homomorphism of posets and lattices
On Wed, 18 Sep 2013, quasi wrote:

> William Elliot wrote:

>

> >What's your opinion about the following except from a manuscript?

>

> You should credit your source.

> Manuscript by who?

Victor.

> Published when and where ?

Preprint; on his web site.

> >I consider the Definition 2.92 to be in error unless the domain

> >of the function is a linear order. For example, the identity

> >function from a two point antichain { a,b } to the chain a < b.

> >

> >Comments? Are not antichains preserved by order isomorphisms?

>

> Yes, of course.

>

> As your counterexample clearly shows, the author's definition

> 2.92 is flawed.

>

> >2,1,12 Homomorphism of posets and lattices

> >

> >Definition 2.90. A monotone function (also called order

> >homomorphism) from a poset A to a poset B is such a function f

> >that x <= y --> f(x) <= f(y).

> >

> >Definition 2.91. Order embedding is an injective monotone

> >function.

> >

> >Definition 2.92. Order isomorphism is an surjective order

> >embedding (= bijective monotone function).

>

> As noted, the above definition is flawed, but the fix is easy.

>

> Definition 2.92 (possible revision):

>

> An order isomorphism is a bijective function such that both it

> and its inverse are monotone functions.

Assume f and f^-1 are monotone and f(x) = f(y)

Then f(x) <= f(y) and f(y) <= f(x). Whence,

x = f^-1f(x) <= f^-1f(y) = y and similarily,

y <= x, so x = y.

Your definition is equivalent to a bimonotone surjection.

How would you revise order embedding? A bimonotone map?

My suggestions for the definitions, are in my recent thread "Order Embedding".

> >Order isomorphism preserves properties of posets, such as

> >order, joins and meets, etc.

>

> With the above suggested revision of definition 2.92, those

> properties _are_ preserved.

Even without knowing what a straight map

may be, what's your opinion of this excerpt?

Proposition 3.4. Let f be a monotone map from a meet-semilattice K

to some poset L. If

Aa,b in K: f(a meet b) = f(a) meet f(b)

then f is a straight map.