Re: What it would take...

On 13/05/2025 00:58, Ben Bacarisse wrote:

Richard Heathfield <rjh@cpax.org.uk> writes:
 

On 12/05/2025 18:21, Ben Bacarisse wrote:

Richard Heathfield <rjh@cpax.org.uk> writes:
>

The HHH code doesn't exactly invite confidence in its author, and his theory
is all over the place, but a thought experiment suggests itself.
>
If we were not all wasting our time bickering with a career bickerer... if
we were to really /really/ try, could we patch up his case and send him on
to his Turing Award? And if so, how?

Eh?

>
Do you know the term 'steelmanning'?
https://en.wikipedia.org/wiki/Straw_man#Steelmanning

 Yes.  That is, as it happens, how I address cranks.  I don't usually
argue against them but try to get them to say, as clearly and as
unambiguously as possible, what they are trying to say.  After a lot of
back and forth I got PO to be clear and unambiguous about what he was
saying.  For example, I asked
 | Here's the key question: do you still assert that H(P,P) == false is
| the "correct" answer even though P(P) halts?
 and PO replied "Yes that is the correct answer even though P(P) halts".
 Ref: Message-ID: <c8idnbFAF6C8QuP8nZ2dnUU7-avNnZ2d@giganews.com>
      Date: Sun, 31 Oct 2021 13:23:29 -0500
 This, to my mind, is the steel man that people should be addressing.
What is the point in countering anything other than this clear position?
 Similarly, he was clear that false (does not halt) was correct for his
sketched simulator because of what /would happen/ if the code was
altered by removing the line that makes the simulation stop.
 The current nonsense avoids the need for this because PO does not show
the full simulation.  Mike Terry "steelmanned" the simulation by showing
the parts PO hides.
 On the other hand, you are spending a lot of time arguing about his
knowledge and use of C.  Yes, it's awful.  He knows very little C and
the code is crap, but that /is/ a straw man -- it's the simplest part of
his argument to fix.  Someone, for example, you, could rewrite it all
clearly and neatly so that the real error (declaring false to be correct
for some halting computations) was even more clear.  There is nothing
unfixable about the simulation.
 

ISTR that there is suspected to be a theoretical window for him, so I
suppose what I'm asking is what sort of boathook we would need to poke that
window a little wider.

What on Earth do you mean?  What window?

>
Well, you know the history better than I do and I'm not about to trawl
through a month's worth of back-messages, so maybe I'm talking nonsense, but
I was under the impression that the line he was taking to attack on Linz's
argument could conceivably have merit.

...

Mr Olcott seems to have (correctly) spotted a minor flaw in
the proof published by Dr Linz.

>
Maybe I grasped the wrong end of that stick.

 I don't know.  There is, indeed, a small error in the notation that is
easy to fix.  I offered to explain the details but you said you had
paint drying that needed to be watched.  As far as I can see, both ends
of the stick say "easy to fix minor flaw".  What end did you grasp?
 ...

(Rather belatedly, it occurs to me that you might be being sarcastic.

>
Moi?!? Perish the thought!
>
But no, I just thought that Mr Olcott is obviously not good at presenting
his case, and it occurred to me that we could probably do a far better
job. We could fix his code, clean up his reasoning, all that, and see what
falls out the bottom.

 As explained, I (and others) have done a lot of that.  For me the steel
man is that, to PO, false is correct from some class of halting
computations.  I assumed you were engaging with him just for the fun of
it, not that you thought there might be some merit in it.
 

Even if it's only ash and clinker.

 False is the correct answer for some halting computations.  Is that ash
and clinker?  I don't know, but I am puzzled by why so many posts don't
address this.  (For the record, I refuse to talk to him because of his
inexcusable rudeness to me.  I can take a lot, but his comments were
finally beyond the pale.)

I think it's just that PO has stopped talking about that point, and instead embarked on a multi-year project to "prove" his argument tiny point by tiny point.  For the last year or so he has been stuck on his first step, focussing on whether the /simulation/ of his DD performed by HHH progresses as far as DD's return.  People just respond on the specific points PO posts about.
Mind you it does seem to have gone mad the last month or so.  It seems there are only about 2 or 3 actual variations of what PO is saying and all the rest is several thousand repeats by both PO and responders...
Mike.