Re: Surrogate Factoring Solution

jstevh_at_msn.com
Date: 03/11/05 

Date: 10 Mar 2005 17:44:30 -0800

Tim Peters wrote:
> [JSH]
> >> My apologies up front, as I have the solution now,
>
> [David C. Ullrich]
> > Here's a hint: Even before you posted the "oops, that was
> > wrong too", _every_ person reading this, except you, was
> > _certain_ that you were wrong,
>
> Nope -- I wasn't. I would have bet on it, but not as much as you'd
think.
> There is a polynomial factorization here with the square of the
target
> number (M) on one side, and it was always possible that _some_ method
would
> (in effect) end up taking the gcd of M with, say, each integer less
than M,
> one at a time. It could have found a factor at any choice of j then.

> Indeed, Nora Baron gave a related method that was in fact guaranteed
to find
> a factor ... eventually.
>

I find this post intriguing, as I think it's one of the more well
written posts I've seen from someone else trying to reason with one of
the sci.math'ers who obsessively reply in my threads.

I feel a slight urge to turn it into a propaganda post for myself, but
I don't think that would be fair, so I'll mainly address some issue
Peters brought up, and point out just a bit, how a sci.math'er like
Ullrich operates, and how he normally gets away with it.

> I would have bet a whole lot more that any such algorithm would end
up being
> no more effective than trial division, but I wouldn't have bet your
life on
> that either -- let alone mine <wink>.
>
> > and most people thought it was pretty likely that you'd be _saying_
> > no you were wrong very quickly.
>
> Keep that in mind, though: James didn't accuse anyone with an actual

> counterexample of lying throughout this multi-month affair. Ya, he
went way
> over the line in hoping that Rick Decker burned in hell _if_ his
claim of a
> counterexample was "a lie", but thereafter when a counterexample was
given
> he moved on at once. So there's a reason people this time _ended up_

> expecting he'd be saying "oops" very quickly. They didn't expect
that when
> this was still new (and because counterexamples were slower in coming
then,
> and so "oops" wasn't heard as frequently then).
>

When I accuse a person of lying I like to have quite a bit of
substantiation.

Not surprisingly, posters who obsessively reply to me who do lie, will
often lie about being caught in lies, and as they do most of the
talking, others seem to just kind of go along with them, and accept
that I just call people liars all the time.

Reality is that I test out ideas. If someone can shoot down an idea,
then it's just shot down.

Getting upset doesn't change facts, and neither does denial or just
calling someone a liar.

> > And they were all right. Except you, of course.
>
> Most were right just by playing the track record odds -- that's a
rational
> reason, but not a very satisfying one. It was longer-lasting fun to
find
> counterexamples than to chortle in triumph in anticipation of one.
>

David Ullrich is a math professor, and I think that some of his posts
are his teaching style, with a lot of psychological stuff thrown in,
where I think in this post readers can see that it looks like he's
trying to continually insist something to me.

However, brainstorming is a process that generates a lot of junk ideas.

The odds are, as Peters says in reply to Ullrich, that I would be
wrong.

It's like baseball.

On any given night, with any given game, when that batter steps up to
the plate, more than likely he will not make it to first base.

The posters like Ullrich who try to make a big deal out of my errors
are like people who don't get baseball.

> > Why not do what Tim suggested, and actually _test_ these things
> > before announcing you had a solution?
>
> That remains the major mystery to me, especially as M=15 at j=14 was
a
> counterexample over and over again. Having been burned by the same
tiny
> example repeatedly, who wouldn't check at least that much?!
>

Good question. I think that as I'd already had several algorithms that
I had believed were perfect, and had failed repeatedly, I had a sinking
feeling about the entire thing, and just started throwing out as much
as I could, in blind hope.

That basically never works, or in my experience it's never worked, but
I find myself at times doing it anyway.

Deep down I have a bad feeling about an idea, but even I, at times,
find myself holding on to it, versus just junking ideas that fail.

But usually I'm a good chunker.

> > You say you're "brainstorming", but what you're actually doing is
> > just posting random modifications, with _no_ reason to think that
> > they're going to fix the problem,
>
> I've said before that it looked to me like James thought he had "good

> reasons" for most of the modifications he suggested. They appear to
have
> been based on misjudging what actually followed from the equations,
but he
> was usually (always?) quite clear about _saying_ "well, just try
this, I
> don't know" when he was truly picking a direction at random. The
latter
> were the posts that didn't start with two pages of algebraic
manipulations I
> couldn't follow.
>

Brainstorming.

If you don't understand the process go get a book and read up.

Also, towards the end, I was just delaying admitting the obvious.

Holding on for emotional reasons with a lost cause.

> > and not even bothering to test them before announcing that you've
> > Solved the Problem.
> >
> > See, it makes you look like an idiot.
>
> The lack of even teensy testing remains a mystery.
>

I've explained above.

I do admit that at times I too will hold on to an idea long past the
time to do the chunking.

Chunking ideas is a valuable skill that you can lose if you don't keep
at it.

My chunking skills seem to be a little flawed lately.

If an idea fails, chunk it.

> ...
> >> Well, my dad did call me paranoid, and maybe he's right. I can
keep
> >> telling myself all kinds of scary things, but knowledge is
valuable,
> >> and I have to believe that the truth is good.
> >>
> >> I believe that the truth is the best.
>
> > And over and over you call the people who say you're wrong liars.
>
> In all fairness, he didn't call me a liar during any of this. Well,
ya, in
> other threads about other topics, but I said "wrong" more often than
anyone
> in these threads and was never called a liar _here_. I never said
"wrong"
> without evidence in hand, though.
>

Exactly. And posters who I do catch in lies tend to just repeat their
lies or claim they're not lying, and they post a LOT.

The people who dominate the threads repeating the same old lies, over
and over, and over again, tend to shift the point of view of others to
their position.

I think some readers naively believe that if someone lies about you on
Usenet you should answer forcefully.

Not if you want a life you don't as these people can keep up a campaign
morning, noon and night, indefinitely.

So some posters or poster lies about you, and you reply, so they reply
back, and it goes on and on, and often it won't just be them, but a
little gang as you stumble into a nest, and before long, all you're
doing is replying to posters who gleefully reply back, until you tire
out, and they declare victory.

> > No matter how many times it turns out they were actually right.
>
> Ya, but many responses of the "you're wrong, wrong, wrong" kind could
just
> as well have been posted by bots keying off the author header. James
does
> have a dedicated following of posters who just enjoy slashing at him,

> regardless of what he's talking about. While he's probably earned
that
> following, just being right isn't a ticket to heaven either <wink>.

Hey, a while back some person actually used a bot, and when a poster
replied that it didn't seem fair to use a robot program to reply to me
all the time, some sci.math'ers jumped on him in posts, and he backed
down.

Conforming behavior.

The reality is that posters who reply obsessivly to me are trying to
control my behavior, and they're too stupid to realize it doesn't work.

I think that when I stop posting for a while they congratulate
themselves telling themselves that they must be why.

Then I start posting again, like when I'm brainstorming, and they
furiously start up yet again.

The process has played out over and over again over a period of years.

Some readers seem to think that I'm the guy who can't get it, when I
get what I want, time after time.

It's the posters who reply to me to control the newsgroups who
continually fail.

I get what I want, airing out of ideas that I'm brainstorming.

So I'm not the one who is continually playing a losing game, even if
every time I come up to bat, the odds are against me.

As while I may strike out more often than I hit, at the end of the
ninth inning, I always have the winning score, as I got what I set out
to get in the brainstorming process.

James Harris

Relevant Pages

  • Re: Surrogate Factoring Solution
    ... Peters brought up, and point out just a bit, how a sci.math'er like ... Not surprisingly, posters who obsessively reply to me who do lie, will ... brainstorming is a process that generates a lot of junk ideas. ... And posters who I do catch in lies tend to just repeat their ...
    (sci.crypt)
  • Re: USELESS PEOPLE (Few exceptions)
    ... What is still a joke is that they think that any of their lies really ... nobody gives a damn on what this most redundant of posters has to say. ... Who cares about this group? ... You and your non-existant therapist buddies is one of your many lies. ...
    (rec.music.opera)
  • Re: for all you good guys out there
    ... lies and begin posting those lies and fabrications about peoples wives ... In common with many other posters, I have been sucked into hurling - ... although on reflection that should be returning - abuse. ... anybody on Usenet) so reputations are irrelevant really. ...
    (uk.people.disability)
  • Re: Sally Sue goes RL
    ... her photo. ... I'm not in the top this month at AMF but I'm on the top all time posters. ... What's the matter, you feel guilty or something, just because I gave a link to report fraud? ... If a person were to make a list of all of your lies and your asinine insinuations, it would reach from one coast to the other. ...
    (alt.support.chronic-pain)
  • Re: Done something.
    ... Newbies to the group are unaware of the personalities of the posters. ... and having trolls coming here posting lies about the posters ... The problem with posters like prot is that they lie about who to trust, ... What he gains is what all trolls gain - joy at the disruption he causes, ...
    (uk.people.support.depression)