How about you read what has been written.
Is there more variation between great danes and poodles than there is within either population?
I have explained quite carefully in my last post why the genetic differences between humans and neanderthals along with the similarities between neanderthals and chimps that humans don't share make neanderthals a separate but closely related branch on the hominid tree.
You didn't address the question at the time, and are now asking questions that indicate that you either didn't understand what I wrote or didn't bother reading it. Unless you can show that there's similar genetic evidence indicating a branch of Great Danes and Poodles away from domesticated dogs (as there clearly is between Neanderthals and humans), then that question is a huge red herring. Do you really not see why?
Neanderthals are very closely related to humans; not surprising based on how close our last common ancestor was supposed to be based on mitochondrial dna evidence. Nevertheless, they are a separate species.
And, as I have pointed out to you twice now, the definition of species is fuzzy because the real world is complex, and there are a lot of ways that closely related organisms can become separate species, and the more closely related a species is, the fuzzier the boundaries get. As I have said to you before, this is how things actually are. In fact, they're what you'd expect to see in the evolutionary paradigm - some speciation events may lead to further and further divergence between populations, and others may fail as populations re-merge. One species may share variable amounts of genetic material with one or more closely related species, yet still maintain a separate identity.
So why do we have the term 'species' or 'subspecies' if there is this continuum? Because populations do differ sufficiently at the genetic and morphological level to require classification (or reclassification) into one species or another. That's why I keep pointing out that even creationists identify organisms using their Latin names at the species level - it's more specific and more useful to be able to talk about wolves, coyotes and domestic dogs than to only be able to refer to them as canid kind, right? Particularly since that as all the kinds have not yet been completely or effectively defined, the term is of limited utility when compared to conventional taxonomy.
Birds singing a new song might be.
People speaking a new language definitely isn't.
If sexual selection in humans was as rigorous and based on a single behavioral characteristic (plumage or song, for example) as is the case in many species of birds, you might have a point. But it isn't, and so you don't.
There is no set rule. It's left open and malleable so evolutionists can call everything they feel comfortable with "evolution" because, look see, it changed!
It's malleable because nature is complicated and messy. It's messy on the gene level, it's messy on the species level. This is reality. You don't have to like it, but there it is.