Re: The Promise of Forth

Guy Macon wrote:
John Doty wrote:
Guy Macon <http://www.guymacon.com/> wrote:

...is a proper subset of the null hypothesis...
...OK, so you *finally* state that your hypothesis is in fact not a "null" hypothesis at all...

Nice touch putting quotes around "null" as if it was some weird
term I invented instead of part of standard scientific methodology.

The null hypothesis is, of course, part of standard scientific methodology. But what *you* call the "null hypothesis" is a different thing.


Let me know if you ever decide to address what I actually wrote
instead of making things up, putting the words in my mouth, and
then knocking down the straw man that you have created.

As long as you choose to take clear statements such as "a proper subset of the null hypothesis" and to pretend that they mean
"not a null hypothesis at all", you have made it impossible to
have any sort of real discussion. I could spend a lot of time
correcting fibs like the one above, but I doubt that doing so would in any way deter you.

As long as you claim an untestable hypothesis qualifies as a null hypothesis in the scientific sense, it is impossible to reason with you.




John Doty wrote:

Bruce, if you believe Forth's sorry situation is a consequence of irreversible and/or unknowable factors, why do you bother with this conversation at all? It doesn't concern you. This conversation is for optimists who have hope that something might still be salvaged of Forth's beautiful ideas.

comp.lang.forth is an unmoderated USENET newsgroup. You don't get to decide who "this conversation is for" or whether "it doesn't concern" someone.

No, but I do get to point out when someone's concerns make no sense.


Besides, pointing out logical fallacies like not considering the
null hypothesis and attempting to educate those who appear to be
totally ignorant as to what the null hypothesis is are always welcome in the sci.* and comp.* newsgroups.

You are indeed totally welcome.





I have concluded, based on your posts, that your repeated ignoring
of the null hypothesis,

You have never stated a null hypothesis. A statistical null hypothesis requires a quantitative statistical model. The way one uses it is that one calculates, given the null hypothesis, the likelihood the observations deviate as much from expected as they are seen to. I have gone down this road many, many, times in my career.

Where's your quantitative statistical model?

repeated replacing of the null hypothesis
with a straw man, and repeated claims that nobody has mentioned
the null hypothesis as an alternative to yours is quite deliberate
and is simply a cheap debating trick.

Calling en empty, untestable assertion a "null hypothesis" isn't even a debating trick.


Your argument that there exists no other hypothesis

No independent *testable* hypothesis has been given.

It is always possible to create untestable hypotheses. Maybe Dennis Ritchie resorted to voodoo to curse Forth to the margins. That's just as good a hypothesis as yours.

The deadliest insult in physics is Pauli's "It isn't even wrong", used to describe a vacuous, untestable hypothesis.

There are also hypotheses on the table that are not independent of the hypothesis of a weakness in Forth: they merely add detail.

and that therefor yours must be correct is an invalid argument, and I think that you are well aware that it is an invalid argument.

I'll ask you, then, why do you care? Your "null" hypothesis has no useful consequences. It offers no way to repair the damage. That's a peculiar position for an engineer to take. Usually, a good engineer seeks a way to do better rather than looking for excuses for an unsatisfactory status quo.


You refuse to consider the hypothesis that there is no cause
and effect relationship and that the correlation is either simply the product of random chance or a product of a cause
or causes unknown (AKA the null hypothesis).

I refuse to consider an untestable hypothesis, especially when I have one that:

1. Actually stands up to the facts as I see them.

2. Offers the possibility of doing better in the future.


Or, as Bruce McFarling wrote to you recently:

"The *point* of a null hypothesis is that it refrains from being one of the competing alternate hypothesis. The null hypothesis is, 'there's nothing unusual here that needs explaining'.

"The null hypothesis in this case is that there is nothing surprising about a computer language being in the position that Forth is in.

"Since you can't contradict the null in favor of your working hypothesis except by replacing it with a flimsy straw horse version, there's no particular reason as of yet to take your working hypothesis seriously."

I obviously can't contradict an untestable hypothesis. Come up with a testable one.

--
John Doty, Noqsi Aerospace, Ltd.
http://www.noqsi.com/
--
History teaches that logical consistency is neither sufficient nor necessary to establish practical, real world truth. Those who attempt to use logic for that purpose are abusing it.
.