Re: Phylogenetic question for John Harshman



On Mon, 15 Sep 2008 07:52:49 -0700 (PDT), spintronic
<spintronic@xxxxxxxxxxx> wrote in
<f2783fbe-7583-46fd-b7e2-821d8477aa5c@xxxxxxxxxxxxxxxxxxxxxxxxxxxx> :

On 15 Sep, 15:37, Augray <aug...@xxxxxxxxxxxx> wrote:
On Mon, 15 Sep 2008 06:07:08 -0700 (PDT), spintronic

You say *skin* "YES/NO" is a useful trait.

When I say "useful", I mean useful for constructing a phylogenetic
tree, which is what this thread is all about. "A plug that works in
this car" is *not* useful for constructing a phylogenetic tree.


Why? In the context of this group.

Because it's so vague that it's virtually meaningless. What did you
mean when you wrote "A plug that works in this car"?


But *skin*;  "black/white" is also a useful trait in a certain
"group",
namely homosapiens.

Perhaps, at least if you're constructing a phylogenetic tree of Homo
sapiens.

Ahh.

Are you saying that in the *construction* of a "specific" group,
small characteristics matter?

That really depends on what you mean by "small characteristics". But
obviously, when constructing specific phylogenetic trees, some
characters are more relevant than others. For instance, "Feathers:
Yes/No" is useless when constructing a tree for birds.


Don't answer, *WE* already know.

Too late.


In our **************GROUP************* of cars.

"w5dc" is as useful as "Black; yes/no".

and

"w7dc" as useful as "White; yes/no".

*Sigh*

Sigh all you like.

Don't worry, I will.


Should we have a vidual, Or a holiday?

If I knew what a "vidual" was, I might be able to answer that.


How many times do I have to explain that "w5dc" or "w7dc" are not
informative traits because they're not a requirement for any
particular vehicle to have?



You have to explain, why "black/whit" skin is informative.

I didn't say that it was, I said that *perhaps* it was. After all,
you're the one who brought it up, and who knows what context you had
in mind when you wrote it?


Yet "w5dc/ w7dc" Is not

Because it's not a consistent characteristic. I mean, how many times
do I have to explain that the presence or absence of w5dc plugs does
not relevant to whether or not a car is a 280 SE Coupé???


In certain "groups".

As far as I can tell, in *any* group.


"Colour" is pretty vague, you know.

Yes, I know. So why did you bring it up? Why this obsession with vague
characters?

.



Relevant Pages

  • Re: Phylogenetic question for John Harshman
    ... I mean useful for constructing a phylogenetic ... this car" is *not* useful for constructing a phylogenetic tree. ... Sigh all you like. ... informative traits because they're not a requirement for any ...
    (talk.origins)
  • Re: about tree
    ... beginers of c writes: ... i want a program for constructing a tree in C. ... And also i wish to traverse the tree.pls write a program for it ...
    (comp.lang.c)
  • Re: HTML Parser
    ... > screenshot, I can see that it is event-based. ... it leaves me with the job of constructing ... > and filling all the data for the object tree I need to have. ... Speed wise, it's way snappy.. ...
    (borland.public.delphi.thirdpartytools.general)
  • help on graph algorithms - constructing a tree with a mixed criterion
    ... undirected graph. ... to construct a tree T which minimizes the cost with the format ... minimum spanning tree (MST) in graph G, ... As constructing MST and SPT both are in complexity P, ...
    (sci.math)
  • help on constructing a tree with a mixed criterion
    ... undirected graph. ... to construct a tree T which minimizes the cost with the format ... minimum spanning tree (MST) in graph G, ... As constructing MST and SPT both are in complexity P, ...
    (comp.theory)