JSH: Under review
jst...@msn.com
verified receipt.
A version of the paper is at
http://groups-beta.google.com/group/extrememathematics/web/MultiPrime.pdf
and in keeping with the philosophy of extreme mathematics, I would
appreciate comments.
The paper is under review as Princeton verified receipt. Uh, yeah that
surprised even me.
This thread is for comments on the paper.
And yes, this paper could end it all. If Princeton does what it should
then it doesn't matter what any of you say, as you know and I know that
Princeton trumps every last one of you.
I especially welcome comments from posters who argue with me regularly.
I am curious though. Read the paper, and come back and post as
confidently as you've done.
What gives you that faith?
James Harris
Larry Hammick
>I wrote another paper and sent it to the Annals of Mathematics which
> verified receipt.
sci-dot-math (again).
jst...@msn.com
No. I know and you know that Princeton doesn't give a damn what you
think.
There is only one case where sci.math definitely interfered and killed
a journal and that is with SWJPAM.
There is no way that Princeton would fall in the same way.
You people can't kill the Annals of Mathematics.
I dare you to try. Send your emails. Just try.
James Harris
Tim Peters
> I wrote another paper and sent it to the Annals of Mathematics
Congratulations at least on taking an appropriate action.
> which verified receipt.
>
> A version of the paper is at
>
> <http://groups-beta.google.com/group/extrememathematics/web/MultiPrime.pdf>
>
> and in keeping with the philosophy of extreme mathematics, I would
> appreciate comments.
>
> The paper is under review as Princeton verified receipt. Uh, yeah that
> surprised even me.
>
> This thread is for comments on the paper.
>
> And yes, this paper could end it all.
"Should", actually -- but won't.
> If Princeton does what it should then it doesn't matter what any of
> you say, as you know and I know that Princeton trumps every last one
> of you.
Except Princeton doesn't trump /you/, right? When the paper is rejected,
nothing will change, except you'll make up bizarre "reasons" for why
Princeton didn't do "what it should".
> I especially welcome comments from posters who argue with me regularly.
No, you don't. You know perfectly well that you'll just hear the same
points raised against this 100 times before. If you were honest, you'd say
instead that you welcome 100 excuses to recycle olds rants of yours in
response.
Here's a hint you've been given many times before: if it's true that
there's "a breakthrough" here, you need to use your claimed results to
/prove/ something not already known. That paper doesn't even attempt such a
thing, beyond merely /asserting/ that it's so:
And quickly with the world's first multi-variable prime
counting function many questions can be answered.
BTW, it's dishonest to the point of deliberate fraud to claim:
and a numerical integration of that partial derivative then
answers the final questions about the relation between the prime
distribution, and continuous functions.
when you've said flat-out here:
The real puzzle for me, still a puzzle, has been the difficulty
I've had in getting that numerical integration properly done.
> I am curious though. Read the paper, and come back and post as
> confidently as you've done.
OK.
> What gives you that faith?
Knowing something about the topic, earned by legitimate study.
mensa...@aol.com
jst...@msn.com wrote:
> I wrote another paper and sent it to the Annals of Mathematics which
> verified receipt.
>
> A version of the paper is at
>
> http://groups-beta.google.com/group/extrememathematics/web/MultiPrime.pdf
>
> and in keeping with the philosophy of extreme mathematics, I would
> appreciate comments.
You didn't really send them that document, did you?
Don't be surprised if they send it back complaining
about lack of "polish".
That is just about the worst .pdf file I've ever seen.
What did you do to make it so crappy looking?
Forget to embed the non-standard font you used?
>
> The paper is under review as Princeton verified receipt. Uh, yeah that
> surprised even me.
Uh, yeah, extreme mathematics is not the way
to present a paper.
>
> This thread is for comments on the paper.
Isn't your abstract a bit light? Aren't you supposed
to explain in English what the breakthrough is?
>
> And yes, this paper could end it all. If Princeton does what it should
> then it doesn't matter what any of you say, as you know and I know that
> Princeton trumps every last one of you.
Could, but won't. When they reject it, you'll continue
to prattle about it for years.
>
> I especially welcome comments from posters who argue with me regularly.
>
> I am curious though. Read the paper, and come back and post as
> confidently as you've done.
This is the last sentence in the paper:
<quote>
And quickly with the world's rst multi-variable prime counting func-
tion many questions can be answered.
</quote>
Where's the rest of the paper?
What are the questions and how have you answered them?
>
> What gives you that faith?
I'll leave it to others to shoot holes through the math.
Next time ask for comments BEFORE you send off
the paper.
Why don't you go read MY paper and learn a few
things about how to present your ideas.
>
>
> James Harris
jst...@msn.com
> [jst...@msn.com]
> > I wrote another paper and sent it to the Annals of Mathematics
>
> Congratulations at least on taking an appropriate action.
>
> > which verified receipt.
> >
> > A version of the paper is at
> >
> > <http://groups-beta.google.com/group/extrememathematics/web/MultiPrime.pdf>
> >
> > and in keeping with the philosophy of extreme mathematics, I would
> > appreciate comments.
> >
> > The paper is under review as Princeton verified receipt. Uh, yeah that
> > surprised even me.
> >
> > This thread is for comments on the paper.
> >
> > And yes, this paper could end it all.
>
> "Should", actually -- but won't.
>
> > If Princeton does what it should then it doesn't matter what any of
> > you say, as you know and I know that Princeton trumps every last one
> > of you.
>
> Except Princeton doesn't trump /you/, right? When the paper is rejected,
> nothing will change, except you'll make up bizarre "reasons" for why
> Princeton didn't do "what it should".
>
Well, they already surprised me. I didn't expect them to acknowledge
the paper at all, especially after the last couple.
Oh, um, yeah, I've had one paper solidly rejected by the Annals of
Mathematics, and they were right as it was a factoring paper, where
later I found out that the method was crap.
I don't mind rejection for math that does not work.
I've had a lot of it.
> > I especially welcome comments from posters who argue with me regularly.
>
> No, you don't. You know perfectly well that you'll just hear the same
> points raised against this 100 times before. If you were honest, you'd say
> instead that you welcome 100 excuses to recycle olds rants of yours in
> response.
>
Brainstorming still escapes you.
Why don't you take a course on modern problem solving techniques and
learn something?
Under brainstorming I am a man possessed.
I walk the line of insanity in the pursuit of knowledge with full
awareness of what I am doing.
But I come back across that line to test the results.
It may be a Faustian bargain, but it is one I take.
And at the end of the day, if the math results are wrong, then I
discard the ideas.
> Here's a hint you've been given many times before: if it's true that
> there's "a breakthrough" here, you need to use your claimed results to
> /prove/ something not already known. That paper doesn't even attempt such a
> thing, beyond merely /asserting/ that it's so:
>
> And quickly with the world's first multi-variable prime
> counting function many questions can be answered.
>
> BTW, it's dishonest to the point of deliberate fraud to claim:
>
> and a numerical integration of that partial derivative then
> answers the final questions about the relation between the prime
> distribution, and continuous functions.
>
> when you've said flat-out here:
>
> The real puzzle for me, still a puzzle, has been the difficulty
> I've had in getting that numerical integration properly done.
>
Is it?
We'll see.
> > I am curious though. Read the paper, and come back and post as
> > confidently as you've done.
>
> OK.
>
Still confident, eh?
> > What gives you that faith?
>
> Knowing something about the topic, earned by legitimate study.
So why then, did Princeton still acknowledge the paper?
They have a history with me as well.
Would you have?
Answer that question alone, would you have acknowledged that paper?
Or would you have simply told me in reply that it was not of merit for
your journal?
Remember, we're talking about the Annals of Mathematics.
I dare you to answer that question fully and honestly.
James Harris
Gib Bogle
Proginoskes
jst...@msn.com wrote:
> I wrote another paper and sent it to the Annals of Mathematics which
> verified receipt.
>
> A version of the paper is at
>
> http://groups-beta.google.com/group/extrememathematics/web/MultiPrime.pdf
>
> and in keeping with the philosophy of extreme mathematics, I would
> appreciate comments.
And keeping with the philosophy of JSH, he'll threaten to kill you and
call you a liar.
> The paper is under review as Princeton verified receipt. Uh, yeah that
> surprised even me.
Not really. "Under review" simply means they got it and (maybe) have
sent it out to a few referees. _Every_ paper starts off "under review".
The referees look at the paper, then write up a brief report, and send
it back to the journal. This is the phase called "Required Reviews
Completed". Every paper goes through this stage as well.
--- Christopher Heckman
Proginoskes
jst...@msn.com wrote:
> Tim Peters wrote:
> > [jst...@msn.com]
> [...]
> > > I especially welcome comments from posters who argue with me regularly.
> >
> > No, you don't. You know perfectly well that you'll just hear the same
> > points raised against this 100 times before. If you were honest, you'd say
> > instead that you welcome 100 excuses to recycle olds rants of yours in
> > response.
> >
>
> Brainstorming still escapes you.
>
> Why don't you take a course on modern problem solving techniques and
> learn something?
I _teach_ a course in modern problem solving techniques. I _know_
brainstorming. I have used it to get results (results that are still
published, btw; for instance, a recursive construction for the 3/8
independence ratio of triangle-free planar graphs with max degree <=
3). As such, I am qualified to say that JSH is not using the entire
brainstorming procedure, only the first two steps. He's NOT throwing
out the junk.
> Under brainstorming I am a man possessed.
>
> I walk the line of insanity in the pursuit of knowledge with full
> awareness of what I am doing.
(1) No argument here.
(2) And yet JSH complains when people diagnose him psychologically.
> But I come back across that line to test the results.
Bullmuffins. You post it without testing the results.
Surrogate Factoring? Never tested. Efficiency of the "new" formula for
the prime function? Never tested. In both cases, JSH admitted he had
_not_ tested either idea, even though either one would only require ten
or 15 minutes worth of computer programming, which he claims to be
proficient in.
> It may be a Faustian bargain, but it is one I take.
A nickel's worth of free advise: If you're going to sell your soul to
the Devil for something, make sure it's for something that's worth it.
> And at the end of the day, if the math results are wrong, then I
> discard the ideas.
Nope. They make it to Usenet. JSH has posted before, saying that he
types it in his computer as the ideas come to him. So either he was
lying before, or he's lying now.
The very fact that the results are debatable means they haven't been
filtered.
> > Here's a hint you've been given many times before: if it's true that
> > there's "a breakthrough" here, you need to use your claimed results to
> > /prove/ something not already known. That paper doesn't even attempt such a
> > thing, beyond merely /asserting/ that it's so:
> >
> > And quickly with the world's first multi-variable prime
> > counting function many questions can be answered.
> >
> > BTW, it's dishonest to the point of deliberate fraud to claim:
> >
> > and a numerical integration of that partial derivative then
> > answers the final questions about the relation between the prime
> > distribution, and continuous functions.
> >
> > when you've said flat-out here:
> >
> > The real puzzle for me, still a puzzle, has been the difficulty
> > I've had in getting that numerical integration properly done.
> >
>
> Is it?
>
> We'll see.
>
> > > I am curious though. Read the paper, and come back and post as
> > > confidently as you've done.
> >
> > OK.
> >
>
> Still confident, eh?
>
> > > What gives you that faith?
> >
> > Knowing something about the topic, earned by legitimate study.
>
> So why then, did Princeton still acknowledge the paper?
Because they didn't lose it. (See my other post in this thread.)
> They have a history with me as well.
Not a good one, though.
> Would you have?
>
> Answer that question alone, would you have acknowledged that paper?
>
> Or would you have simply told me in reply that it was not of merit for
> your journal?
That's not for the editor to say; that's for the referees to say. So,
if I were an editor, the answe would be no, but if I were a referee, I
would read it and then decide.
> Remember, we're talking about the Annals of Mathematics.
>
> I dare you to answer that question fully and honestly.
They have to analyze _any_ paper that is submitted to them. Even those
written by crackpots, like Archimedes Plutonium.
--- Christopher Heckman
Tim Peters
>>> I wrote another paper and sent it to the Annals of Mathematics
[Tim Peters]
>> Congratulations at least on taking an appropriate action.
>>> which verified receipt.
>>>
>>> A version of the paper is at
>>>
>>> <http://groups-beta.google.com/group/extrememathematics/web/MultiPrime.pdf>
>>>
>>> and in keeping with the philosophy of extreme mathematics, I would
>>> appreciate comments.
>>>
>>> The paper is under review as Princeton verified receipt. Uh, yeah
>>> that>surprised even me.
>>>
>>> This thread is for comments on the paper.
>>>
>>> And yes, this paper could end it all.
>> "Should", actually -- but won't.
>>> If Princeton does what it should then it doesn't matter what any of
>>> you say, as you know and I know that Princeton trumps every last one
>>> of you.
>> Except Princeton doesn't trump /you/, right? When the paper is
>> rejected,nothing will change, except you'll make up bizarre "reasons"
>> for why Princeton didn't do "what it should".
> Well, they already surprised me. I didn't expect them to acknowledge
> the paper at all, especially after the last couple.
It's unclear what "verified receipt" means. What, exactly, did they say to
you?
> Oh, um, yeah, I've had one paper solidly rejected by the Annals of
> Mathematics, and they were right as it was a factoring paper, where
> later I found out that the method was crap.
Huh? You previously said, right here, that the paper previously rejected by
the Annals was the "Advancing Polynomial Factorization" paper later
published in Chiaroscuro:
http://www.etienne.nu/isis/Chiaroscuro8.pdf
You agree "the method was crap" there?
> I don't mind rejection for math that does not work.
>
> I've had a lot of it.
>>> I especially welcome comments from posters who argue with me
>>> regularly.
>>
>> No, you don't. You know perfectly well that you'll just hear the
>> same points raised against this 100 times before. If you were honest,
>> you'd say instead that you welcome 100 excuses to recycle olds rants
>> of yours in response.
> Brainstorming still escapes you.
Hold on. What does brainstorming have to do with a paper submitted to the
/Annals/? While you'll get you no argument from me that the paper reads
like typical JSH brainstorming gibberish (full of obscurities and
unjustified wild claims), are you saying that this paper is "just
brainstorming" to you too? If so, why on Earth would you imagine a serious
journal would even dream of publishing it? Or, if you're not saying it's
"just brainstorming", why are you bringing up brainstorming here?
> Why don't you take a course on modern problem solving techniques and
> learn something?
As above, this looks like an attempt to just start ranting again. To answer
your question, I've made a nice living solving real-world problems for
decades, and get all the "positive feedback" I need about that in real life.
I have learned from you, but, alas, mostly reminders about what happens when
hopes are allowed to overrule realities. I've learned a lot more from a few
general problem-solving, and many field-specific, seminars and conferences
over the years.
> Under brainstorming I am a man possessed.
>
> I walk the line of insanity in the pursuit of knowledge with full
> awareness of what I am doing.
>
> But I come back across that line to test the results.
Huh. That's a stage I've never seen, then.
> It may be a Faustian bargain, but it is one I take.
>
> And at the end of the day, if the math results are wrong, then I
> discard the ideas.
Sometimes. It would be a lot more effective if you could cut years off the
process. Like, when everyone tells you "you're wrong" about a technical
point, it's always the case that you are (although it sometimes takes you
years to catch up -- and on some such points you're still in denial).
>> Here's a hint you've been given many times before: if it's true that
>> there's "a breakthrough" here, you need to use your claimed results to
>> /prove/ something not already known. That paper doesn't even attempt
>> such a thing, beyond merely /asserting/ that it's so:
>>
>> And quickly with the world's first multi-variable prime
>> counting function many questions can be answered.
Lack of response implies you still don't "get" that part. The paper's
reviewer certainly will get it.
>> BTW, it's dishonest to the point of deliberate fraud to claim:
>>
>> and a numerical integration of that partial derivative then
>> answers the final questions about the relation between the prime
>> distribution, and continuous functions.
>>
>> when you've said flat-out here:
>>
>> The real puzzle for me, still a puzzle, has been the difficulty
>> I've had in getting that numerical integration properly done.
> Is it?
Yes. Presenting a mathematical claim as unqualifiedly true when you know
you can't prove it is fraud. I expect this happens to everyone on Usenet
from time to time, given the general low standards and time pressures, but
in a paper submitted for publication it would destroy your reputation and
end your career. Of course you're immune on those points, but it's fraud
all the same.
> We'll see.
Already did.
>>> I am curious though. Read the paper, and come back and post as
>>> confidently as you've done.
>> OK.
> Still confident, eh?
Of course:
>>> What gives you that faith?
>> Knowing something about the topic, earned by legitimate study.
> So why then, did Princeton still acknowledge the paper?
As above, I don't know what "acknowledge" means, and I certainly don't speak
for Princeton regardless. As someone with relevant training and hard-earned
knowledge, I can tell you it would be beyond merely miraculous if a journal
with such high standards published that paper. Sorry, but that's a fact.
> They have a history with me as well.
>
> Would you have?
Would I have /what/? Counterfactual hypotheticals suck regardless.
> Answer that question alone, would you have acknowledged that paper?
If I were in charge of receiving papers, I'd certainly /acknowledge/ receipt
of every paper received. Sounds like a basic job requirement.
> Or would you have simply told me in reply that it was not of merit for
> your journal?
The person in charge of receiving papers generally isn't empowered to make
such judgments. If you sent that paper scribbled in crayon on stained
cocktail napkins, maybe.
> Remember, we're talking about the Annals of Mathematics.
>
> I dare you to answer that question fully and honestly.
Then I dare you to make more sense ;-) Start by quoting /exactly/ what
Princeton told you, not a paraphrase. We've seen how even polite two-line
brush-offs (like Minsky's) get interpreted by you as "top mathematicians
found no problems in my work". Your capacity for self-flattering delusion
is unparalleled in my experience of the world.
The Last Danish Pastry
news:1164345377.1...@j44g2000cwa.googlegroups.com...
>I wrote another paper and sent it to the Annals of Mathematics which
> verified receipt.
>
> A version of the paper is at
>
> http://groups-beta.google.com/group/extrememathematics/web/MultiPrime.pdf
>
> and in keeping with the philosophy of extreme mathematics, I would
> appreciate comments.
>
> The paper is under review as Princeton verified receipt. Uh, yeah
> that
> surprised even me.
>
> This thread is for comments on the paper.
James... it's perfect!! A gem of a paper!! I am sure it will be in the
Annals very soon.
[And thanks for explaining the purpose of this thread.]
The paper is concise, yet _full_ of mathematics!!
It has those things that look like a capital M on its side. And square
roots!!
But you saved the best bit to near the end: those curly d things!!
They look great!!
Excellent!!
--
Clive Tooth
http://www.shutterstock.com/cat.mhtml?gallery_id=61771
Dave Parker
looked like that (in other words, shitty), we would glance
at it for .04 ms before tossing it in the garbage.
Learn how to use document-producing tools.
Rupert
All you have done in this paper is the following:
You have given your algorithm for evaluating the prime counting
function, which is correct but not of sufficient interest to be
published in the Annals of Mathematics.
You have written down a partial differential equation and proved
absolutely nothing about it.
As Tim mentioned, it is extremely dishonest of you to say "numerical
integration of that partial differential equation then answers the
final questions" when you haven't performed this numerical integration.
Richard Henry
Maybe you hit them at the right time of their cycle. It may be time to
gather articles for their April issue.
Richard Tobin
<jst...@msn.com> wrote:
>I would appreciate comments.
No you wouldn't. You'd just denounce them as lies.
-- Richard
--
"Consideration shall be given to the need for as many as 32 characters
in some alphabets" - X3.4, 1963.
Larry Hammick
jst...@msn.com wrote:
> Larry Hammick wrote:
> > <jst...@msn.com>
> > >I wrote another paper and sent it to the Annals of Mathematics which
> > > verified receipt.
> > Conjecture: When they reject it, you will blame the sinister cadre of
> > sci-dot-math (again).
>
> No. I know and you know that Princeton doesn't give a damn what you
> think.
>
> There is only one case where sci.math definitely interfered and killed
> a journal and that is with SWJPAM.
>
> There is no way that Princeton would fall in the same way.
will reject your paper, and I said you will then blame the people in
this room. And I say it again.
sg...@hotmail.co.uk
> I did not say that the Annals would go out of business. I said they
> will reject your paper, and I said you will then blame the people in
> this room. And I say it again.
"This room"? Damnit, man, we're supposed to be keeping up the pretense
of being a large number of posters operating independently around the
world. Now you go and let slip about the JSH Replies office in the
cabal headquarters.
-Rotwang
David C. Ullrich
If the paper is "brainstorming" then it's extremely inappopriate
to try to get it published as fact.
Not that it matters, since there's no chance it will appear.
What _is_ the excuse going to be for this rejection, I wonder?
>Why don't you take a course on modern problem solving techniques and
>learn something?
>
>Under brainstorming I am a man possessed.
>
>I walk the line of insanity in the pursuit of knowledge with full
>awareness of what I am doing.
>
>But I come back across that line to test the results.
>
>It may be a Faustian bargain, but it is one I take.
>
>And at the end of the day, if the math results are wrong, then I
>discard the ideas.
>
>> Here's a hint you've been given many times before: if it's true that
>> there's "a breakthrough" here, you need to use your claimed results to
>> /prove/ something not already known. That paper doesn't even attempt such a
>> thing, beyond merely /asserting/ that it's so:
>>
>> And quickly with the world's first multi-variable prime
>> counting function many questions can be answered.
>>
>> BTW, it's dishonest to the point of deliberate fraud to claim:
>>
>> and a numerical integration of that partial derivative then
>> answers the final questions about the relation between the prime
>> distribution, and continuous functions.
>>
>> when you've said flat-out here:
>>
>> The real puzzle for me, still a puzzle, has been the difficulty
>> I've had in getting that numerical integration properly done.
>>
>
>Is it?
Yes, it is. Dishonest to the point of deliberate fraud is exactly
right - in the paper you say that numerical integration _answers_
questions, while in fact you haven't done the numerical integration.
Again, not that it matters. Nobody's going to ooh and ahh over your
statement that you can "answer questions" unless you give _some_
hint _what_ questions you're answering.
>We'll see.
Um, regarding the question of whether this is fraud, we've _seen_.
The paper _is_ fraudulent. Supposing that pigs fly and the
paper is published by the Annals, that wouldn't change the fact
that it's fraudulent.
Even if someone does the numerical integration soon and it
_does_ answer questions, that _still_ would not change the
fact that the paper is fraudulent.
>> > I am curious though. Read the paper, and come back and post as
>> > confidently as you've done.
>>
>> OK.
>>
>
>Still confident, eh?
>
>> > What gives you that faith?
>>
>> Knowing something about the topic, earned by legitimate study.
>
>So why then, did Princeton still acknowledge the paper?
>
>They have a history with me as well.
They have a history of not acknowledging submissions from you?
If so my guess is that the reason they acknowledged this one
was to shut you up for a while - they're hoping if they say
they got this one you won't send them any more nonsense until
they've got around to formally rejecting this one.
>Would you have?
>
>Answer that question alone, would you have acknowledged that paper?
>
>Or would you have simply told me in reply that it was not of merit for
>your journal?
>
>Remember, we're talking about the Annals of Mathematics.
>
>I dare you to answer that question fully and honestly.
Not that you asked me, but if _I_ were the editor of a journal,
even a low-class one, not something like the Annals, I'd send
you a note saying that the paper is not suitable. I don't know
what excuse I'd use in the note, but the real reason would be
that the paper is _obviously_ simply nonsense.
That's a full and honest answer to your question. Happy?
>James Harris
************************
David C. Ullrich
jst...@msn.com
BUT I blame sci.math only for one journal misadventure which is with
SWJPAM.
There sci.math'ers emailed the editors attacking my paper, the editors
yanked it, and a few months later the journal shut down.
So you're deluded if you think you have any weight with editors at
Princeton, as if that scenario--which is the only scenario by which I
would blame the sci.math newsgroup--would play out again.
There is just no way that my paper could get published by the Annals
and retracted because of some emails from any or all of you, so there
is no basis for your concern.
There is no other case where I have blamed sci.math at all with regard
to a journal, so your repetition is just bizarre, as if, as if you
people could sway Princeton editors the way you did the editors at the
Southwest Journal of Pure and Applied Mathematics.
You just could not do it.
You people are not the problem here. Mathematicians at universities
have the real power, and you got the jump on some editors at a small
electronic journal in Oklahoma.
That was just a lucky strike. Princeton is immune to fringe Usenet
groups.
James Harris
Arturo Magidin
Tim Peters <tim...@comcast.net> wrote:
[...]
>It's unclear what "verified receipt" means. What, exactly, did they say to
>you?
At a guess, something along the usual lines:
Dear XXXX
We are in receipt of the manuscript entitled "XXXXX" by X, Y, and
Z, which you have submitted to the "Annals of Mathematics."
The manuscript has been sent to an editor/referee for
handling/review. We will contact you when we have their report.
Thank you for submitting your work to the "Annals."
Yours faithfully,
XXXXX
--
======================================================================
"It's not denial. I'm just very selective about
what I accept as reality."
--- Calvin ("Calvin and Hobbes" by Bill Watterson)
======================================================================
Arturo Magidin
magidin-at-member-ams-org
mensa...@aol.com
What makes you think the editors haven't already
been e-mailed?
Eddie Journal Spork
<jst...@msn.com> wrote in message
news:1164380773....@45g2000cws.googlegroups.com...
>
> Larry Hammick wrote:
> > jst...@msn.com wrote:
> > > Larry Hammick wrote:
> > > > <jst...@msn.com>
> > > > >I wrote another paper and sent it to the Annals of Mathematics
which
> > > > > verified receipt.
> > > > Conjecture: When they reject it, you will blame the sinister cadre
of
> > > > sci-dot-math (again).
> > >
> > > No. I know and you know that Princeton doesn't give a damn what you
> > > think.
> > >
> > > There is only one case where sci.math definitely interfered and killed
> > > a journal and that is with SWJPAM.
You're welcome.
The same will happen soon with "Annals of Mathematics".
We will be contacting their editors on Monday morning to warn
them about crank submissions.
jst...@msn.com
The replies in this thread reek of fear.
What I actually did was go from a prime counting sieve function, to a
constrained summation of a partial difference equation that counts
primes, to the partial differential equation that follows from it.
The three forms connect the dots from the discrete count of primes to a
continuous function directly, for the first time in mathematical
history, showing how the count of primes connects to continuous
functions, and accomplishing what notables like Gauss and Riemann set
out to do so long ago.
That is done in just two pages, and is an accomplishment at the
pinnacle of human achievement in mathematics, but it's also an
accomplishment members of the sci.math newsgroup have enviously and
jealously fought to disparage.
It is one of the greatest accomplishments in the history of human
thought.
It is more than a great enough accomplishment to merit publication in
one of the world's top math journals.
James Harris
Rupert
Tee hee hee.
The replies in this thread are the same as they have always been.
> What I actually did was go from a prime counting sieve function, to a
> constrained summation of a partial difference equation that counts
> primes, to the partial differential equation that follows from it.
>
> The three forms connect the dots from the discrete count of primes to a
> continuous function directly, for the first time in mathematical
> history, showing how the count of primes connects to continuous
> functions, and accomplishing what notables like Gauss and Riemann set
> out to do so long ago.
>
Yes, but you haven't actually done anything with the partial
differential equation. You haven't even proved that it has a solution.
All you've done is write it down. There is no reason to think it leads
to anything interesting.
> That is done in just two pages, and is an accomplishment at the
> pinnacle of human achievement in mathematics, but it's also an
> accomplishment members of the sci.math newsgroup have enviously and
> jealously fought to disparage.
>
I'm afraid just writing down a partial differential equation doesn't
count as the pinnacle of human achievement in mathematics. As far as
the partial differential equation goes you've done absolutely nothing,
you haven't proved a single thing about it. What you've done is give
the recurrence relation on which your algorithm for counting primes is
based. And I'm afraid that's not at the pinnacle of human achievement
either.
> It is one of the greatest accomplishments in the history of human
> thought.
>
> It is more than a great enough accomplishment to merit publication in
> one of the world's top math journals.
>
You reckon?
>
> James Harris
fishfry
> I wrote another paper and sent it to the Annals of Mathematics which
> verified receipt.
Only one 'n' needed.
Tim Peters
>> I wrote another paper and sent it to the Annals of Mathematics which
>> verified receipt.
[fishfry]
> Only one 'n' needed.
Possibly, but ragging on James for using Princeton's spelling is of dubious
value when there's so much of substance you /could/ object to instead ;-)
Patrick Hamlyn
Possibly you are operating under the humour-impaired model commonly used by
denizens of the US and A.
'Fishfry' is not seriously challenging the spelling of annals, he is implying
that a different spelling would have been a more appropriate address for James'
paper.
In other countries we call this humour and appreciate it 'in and of itself'.
------------ And now a word from our sponsor ---------------------
For a secure high performance FTP using SSL/TLS encryption
upgrade to SurgeFTP
---- See http://netwinsite.com/sponsor/sponsor_surgeftp.htm ----
junoexpress
1) Abstract contains boast, but no actual statement of what is done.
2) Intro is confusing
3) No references,
4) no derivations,
5) grandoise claims,
6) no conclusion
and most importantly,
7) no theorems and no results!
Yeah, I'm sure Princeton is running the presses overtime as we speak.
Matt B
jst...@msn.com wrote:
> I wrote another paper and sent it to the Annals of Mathematics which
> verified receipt.
>
Tim Peters
>>>> I wrote another paper and sent it to the Annals of Mathematics which
>>>> verified receipt.
[fishfry]
>>> Only one 'n' needed.
[Tim Peters]
>> Possibly, but ragging on James for using Princeton's spelling is of
>> dubious value when there's so much of substance you /could/ object
>> to instead ;-)
>>
>> http://annals.princeton.edu/
[Patrick Hamlyn]
> Possibly you are operating under the humour-impaired model commonly
> used by denizens of the US and A.
>
> 'Fishfry' is not seriously challenging the spelling of annals, he is
> implying that a different spelling would have been a more appropriate
> address for James' paper.
>
> In other countries we call this humour and appreciate it 'in and of
> itself'.
Ah, excellent point! It's true that I have no sense of humor whatsoever. I
did get a tattoo of the Queen Mother on my left bicep to remind me to
/pretend/ to have one, but, alas, I was too weary of JSH threads to flex by
the time I saw fishfry's reply, and in its relaxed state the tattoo
resembles nothing so much as a frowning pale sausage. Now that you've
pointed it out, I'm delighted to report that flexing the tattoo in the
direction of fishfry's posting clearly shows the Queen Mother appearing to
wink at at. That's a sure sign of rich English humor after all :-)
Tim Peters
http://groups-beta.google.com/group/extrememathematics/web/MultiPrime.pdf
]
> A perfect JSH paper.
>
> 1) Abstract contains boast, but no actual statement of what is done.
> 2) Intro is confusing
> 3) No references,
> 4) no derivations,
> 5) grandoise claims,
> 6) no conclusion
> and most importantly,
> 7) no theorems and no results!
Bingo! Excellent summary. The astonishing thing is that JSH can't see why
any one of them would be enough to merit rejection by a serious journal.
> Yeah, I'm sure Princeton is running the presses overtime as we speak.
I expect they'll devote an entire issue to it, perhaps 100 copies of the
article just to fill the page quota ;-)
Gib Bogle
>>And at the end of the day, if the math results are wrong, then I
>>discard the ideas.
>
>
> Nope. They make it to Usenet. JSH has posted before, saying that he
> types it in his computer as the ideas come to him. So either he was
> lying before, or he's lying now.
There is another option ...
Davie
<jst...@msn.com> wrote in message
news:1164345377.1...@j44g2000cwa.googlegroups.com...
>I wrote another paper and sent it to the Annals of Mathematics which
> verified receipt.
>
>
> http://groups-beta.google.com/group/extrememathematics/web/MultiPrime.pdf
>
that is not a paper, it should be an abstract, nothing is in it. an empty
jar, nada, zero, blank, zip, poof,
You submitted that 2 page piece of no-dung to them? tossed in the trash, no
comment.
You have given Extreme Mathematics a bad name, fool.
Larry Hammick
the harmless abstract sense, as in "chat room". But you're thinking of
The Room -- the secret chamber in which the Mathematicians in Black
periodically congregate to decide the fate of civilisation over brandy
and cigars. The location of The Room is so secret that not even Jack
Sarfatti has been able to determine it, even with the help of his
clairvoyant business partner Uri Geller and their associates on other
planets.
LH
Davie
"The Last Danish Pastry" <cli...@gmail.com> wrote in message
news:4snjriF...@mid.individual.net...
come on ! who is sh*tting who, here ? But he asked for
comments.........
Proginoskes
And you think Princeton is immune to "us"? Bwa ha ha ha ha!
"What a maroon." -- Bugs Bunny.
--- Christopher Heckman
Proginoskes
Patrick Hamlyn wrote:
> "Tim Peters" <tim...@comcast.net> wrote:
>
> >[jst...@msn.com]
> >>> I wrote another paper and sent it to the Annals of Mathematics which
> >>> verified receipt.
> >
> >[fishfry]
> >> Only one 'n' needed.
> >
> >Possibly, but ragging on James for using Princeton's spelling is of dubious
> >value when there's so much of substance you /could/ object to instead ;-)
> >
> > http://annals.princeton.edu/
>
> Possibly you are operating under the humour-impaired model commonly used by
> denizens of the US and A.
>
> 'Fishfry' is not seriously challenging the spelling of annals, he is implying
> that a different spelling would have been a more appropriate address for James'
> paper.
>
> In other countries we call this humour and appreciate it 'in and of itself'.
I saw the humor in it, and I can even spell "humor" correctly.[1]
If I wasn't overweight, no one would believe that I was an American.[2]
They usually place me as being from somewhere around the eastern
Mediterranian.
--- Christopher Heckman
[1] and [2]: These are, for the humor-impaired people in the crowd,
jokes. [1] reminds me of when I was talking with our department head --
who is British -- about printing up certificates for the top Putnam
winners at ASU. He decided on the final wording: "We'll say 'honorable
mention', and leave out the 'u'." I came very close to saying, "Sure,
and I'll leave out the 'z' as well."
Proginoskes
junoexpress wrote:
> A perfect JSH paper.
>
> 1) Abstract contains boast, but no actual statement of what is done.
> 2) Intro is confusing
> 3) No references,
> 4) no derivations,
> 5) grandoise claims,
> 6) no conclusion
> and most importantly,
> 7) no theorems and no results!
>
> Yeah, I'm sure Princeton is running the presses overtime as we speak.
There was a paper written about "crank proofs" of some result (I don't
remember which one, but it wasn't the 4CT; it might have been FLT). In
the opening, the author (whose name I don't remember, either) said that
he is obviously unable to refer to specific proofs, since he would have
to ask the authors: "Would you mind if I use your proof as an example
of junk?" Maybe Princeton has found a way around this legal obstacle.
--- Christopher Heckman
P.S. Did anyone notice that "Princeton" is an anagram of "PRINT ONCE"?
Or "INEPT CORN"? Or "INTERN COP"?
The Last Danish Pastry
news:4567c92d$0$97214$892e...@authen.yellow.readfreenews.net...
I have never sh*tted anybody!!
But I do think those curly d gizmos are WAY KEWL. AND he has a little
triangle thingy in there too... That is REAL mathematics!!
--
Clive Tooth
http://www.shutterstock.com/cat.mhtml?gallery_id=61771
The Last Danish Pastry
news:1164410130....@f16g2000cwb.googlegroups.com...
>A perfect JSH paper.
Yes, total perfection.
~~~~
When anybody is writing something they often imagine it being read by
a particular type of person. James, I am fairly sure, imagines his
work being read by a professional mathematician (PM) who has never
heard of James Harris.
As James lovingly crafts his prose, he imagines PM's face, and he
tries (often successfully) to induce the following sequence of
expressions...
# A neutral expression. [About two words into the piece.]
# A very slight frown of puzzlement. [Usually about six words into the
piece.]
# The big frown. [In this paper I think James is going for "the big
frown" moment when he defines the P(,) thing without actually saying
what it is. But I think the big frown may actually have come a little
sooner.]
# The really annoyed look.
# Then... that magical tipping point... when the first trace of a
smile flickers across PM's face.
# PM reads on, entranced. The smile becomes bigger and more confident.
PM is quite enjoying it now.
James continues to pile on the humor and PM may go through the
following phases:
# Laughing out loud.
# Beer spurted over computer screen.
# Tears running down face.
# etc.
Of course, after that first read PM will no longer be a "James Harris
virgin" and may never be able to recapture those first-time magical
moments. But, such is James' skill in these matters, many people
return again and again to his work.
Maximilian von Krank
> http://groups-beta.google.com/group/extrememathematics/web/MultiPrime.pdf
Laughable.
> It is one of the greatest accomplishments in the history of human
> thought.
>
> It is more than a great enough accomplishment to merit publication in
> one of the world's top math journals.
Harris, are you even AWARE of what a total ass you make of yourself, over
and over?
David C. Ullrich
Guffaw. They reek of fear in your imagination.
What did you think was going to happen when you _asked_ for
comments on the paper? You thought people were going to say
how great it was, when people have explained over and over
how what little material there is in the paper is just nonsense?
>What I actually did was go from a prime counting sieve function, to a
>constrained summation of a partial difference equation that counts
>primes, to the partial differential equation that follows from it.
>
>The three forms connect the dots from the discrete count of primes to a
>continuous function directly, for the first time in mathematical
>history, showing how the count of primes connects to continuous
>functions, and accomplishing what notables like Gauss and Riemann set
>out to do so long ago.
>
>That is done in just two pages, and is an accomplishment at the
>pinnacle of human achievement in mathematics, but it's also an
>accomplishment members of the sci.math newsgroup have enviously and
>jealously fought to disparage.
>
>It is one of the greatest accomplishments in the history of human
>thought.
>
>It is more than a great enough accomplishment to merit publication in
>one of the world's top math journals.
I _am_ glad you sent it to the Annals. Because when you make these
grandiose claims I'm never certain whether you're serious...
marcus_b
> I wrote another paper and sent it to the Annals of Mathematics which
> verified receipt.
>
> A version of the paper is at
>
> http://groups-beta.google.com/group/extrememathematics/web/MultiPrime.pdf
>
> and in keeping with the philosophy of extreme mathematics, I would
> appreciate comments.
>
> The paper is under review as Princeton verified receipt. Uh, yeah that
> surprised even me.
>
> This thread is for comments on the paper.
>
> And yes, this paper could end it all. If Princeton does what it should
> then it doesn't matter what any of you say, as you know and I know that
> Princeton trumps every last one of you.
>
> I especially welcome comments from posters who argue with me regularly.
>
> I am curious though. Read the paper, and come back and post as
> confidently as you've done.
>
> What gives you that faith?
>
The Annals ordinarily does not publish papers which contain
freshman-level mistakes in calculus.
Your expression for the differential equation is wrong. You
have used the chain rule incorrectly. You start with the
difference equation
delta(P(k, k) = P(k, sqrt(k)) - P(k - 1, sqrt(k - 1)),
and, passing to the limit (replacing the "-1" by "dk" and
then replacing "k" by "y", in your last equation, you end
up with
dP(y, sqrt(y))/dy.
However, the derivative of P(k,sqrt(k)) should be
(dP/dx1) + (dP/dx2)*(.5*y^(-.5)),
where dP/dx1 and dP/dx2 represent partial derivatives of P with
respect to the first and second arguments respectively. and these
are evaluated at x1 = y, x2 = y.
So your implied claim to have numerically integrated to find
a solution is obviously false. For this reason alone the paper
should be withdrawn.
Marcus.
>
> James Harris
Tim Peters
http://groups-beta.google.com/group/extrememathematics/web/MultiPrime.pdf
]
> The Annals ordinarily does not publish papers which contain
> freshman-level mistakes in calculus.
So you might think -- but then you're just some fringe yahoo on Usenet,
while the editors at the Annals are Top Mathematicians who will be so bowled
over by the paper's brilliance they'll /kill/ for a chance to correct what
are, after all, just typos ;-)
> Your expression for the differential equation is wrong. You
> have used the chain rule incorrectly. You start with the
> difference equation
>
> delta(P(k, k) = P(k, sqrt(k)) - P(k - 1, sqrt(k - 1)),
Does he? It's clear as mud to me what he thinks he's doing. /If/ he really
started from "the difference equation" you give there, then the appearance
of, e.g., P(x/y, y) in his claimed PDE wouldn't even seem to make "Harris
Sense".
What I /guess/ he did is take his entire "difference equation" expression
for P(x, y), and compute P(x, y) - P(x, y-1) to get exactly (all but one of
the terms in the expansions of the two sums cancel out):
P(x, y) - P(x, y-1) =
-[P(x/y, y-1) - P(y-1, sqrt(y-1)] * deltaP(y, y)
Or perhaps he started from an exact expression for P(x, y+1) - P(x, y)
instead.
Regardless, he then tries computing the partial derivative wrt y, and
various forms of magical thinking permit ignoring the chain rule in the part
you identified, but also ignoring the chain rule in the P(x/y, y-1) part;
and in the P(y-1, sqrt(y-1)) part (repeated in the expansion of the
deltaP(y, y) part); ignoring the differences between y and y-1; and, for the
"coup de grace", wishing away that he /knows/ deltaP(y, y) was torturously
contrived to be an indicator function (0 if y composite, 1 if y prime), so
that the notion he's going to get a lick of sense out of this by presuming
to differentiate it is exceedingly dubious on the face of it.
> and, passing to the limit (replacing the "-1" by "dk" and
> then replacing "k" by "y", in your last equation, you end
> up with
>
> dP(y, sqrt(y))/dy.
>
> However, the derivative of P(k,sqrt(k)) should be
>
> (dP/dx1) + (dP/dx2)*(.5*y^(-.5)),
>
> where dP/dx1 and dP/dx2 represent partial derivatives of P with
> respect to the first and second arguments respectively. and these
> are evaluated at x1 = y, x2 = y.
>
> So your implied claim to have numerically integrated to find
> a solution is obviously false.
Hard to say because he hasn't shown any of his work. As David Ullrich has
repeatedly pointed out to him over a span of years now, numerical
integration can deliver all sorts of nonsense depending not just on the
function, but also on the method and step size chosen. It's quite
conceivable to me that James thrashed at random until he stumbled into some
misunderstanding of some numeric method using an inappropriae step size
that, e.g., got close to 4 when he crunched the numbers at x=10. Woo hoo!
HUGE result.
What is certain is that he hasn't /proved/ a damn thing about it, so that
his:
and a numerical integration of that partial derivative then
answers the final questions about the relation between the prime
distribution, and continuous functions.
is pure bluff, which he often does on Usenet. In the context of
publication, "fraud" is a more accurate term.
> For this reason alone the paper should be withdrawn.
Well, let's not go overboard ;-)
jpaw...@gmail.com
jst...@msn.com wrote:
> I wrote another paper and sent it to the Annals of Mathematics which
> verified receipt.
>
> A version of the paper is at
>
> http://groups-beta.google.com/group/extrememathematics/web/MultiPrime.pdf
>
> and in keeping with the philosophy of extreme mathematics, I would
> appreciate comments.
>
> The paper is under review as Princeton verified receipt. Uh, yeah that
> surprised even me.
>
> This thread is for comments on the paper.
>
> And yes, this paper could end it all. If Princeton does what it should
> then it doesn't matter what any of you say, as you know and I know that
> Princeton trumps every last one of you.
>
> I especially welcome comments from posters who argue with me regularly.
>
> I am curious though. Read the paper, and come back and post as
> confidently as you've done.
>
> What gives you that faith?
>
>
I read the abstract before realizing this was over my head. Its just
too EXTREME for me.
It doesn't appear you've mentioned extreme mathematics in your paper.
Why not? Do even you realize that it makes you sound 100% ridiculous?
marcus_b
> [marcus_b, on
> http://groups-beta.google.com/group/extrememathematics/web/MultiPrime.pdf
> ]
> > The Annals ordinarily does not publish papers which contain
> > freshman-level mistakes in calculus.
>
> So you might think -- but then you're just some fringe yahoo on Usenet,
> while the editors at the Annals are Top Mathematicians who will be so bowled
> over by the paper's brilliance they'll /kill/ for a chance to correct what
> are, after all, just typos ;-)
>
I know, I know, I am risking all by defying the Powers That Be at
Princeton. Even as we speak this paper may be roaring through
the faculty like a bolt from on high, and once they see the /truth/
of it, our careers are over. I am just rolling the dice here on the
off-chance that Harris overlooked some little error ... maybe some
of the fame will rub off ...
> > Your expression for the differential equation is wrong. You
> > have used the chain rule incorrectly. You start with the
> > difference equation
> >
> > delta(P(k, k) = P(k, sqrt(k)) - P(k - 1, sqrt(k - 1)),
>
> Does he? It's clear as mud to me what he thinks he's doing.
There's no doubt he does include this equation. It has to be the
basis for what he does later.
> /If/ he really
> started from "the difference equation" you give there, then the appearance
> of, e.g., P(x/y, y) in his claimed PDE wouldn't even seem to make "Harris
> Sense".
>
It makes no sense on any scale whatsoever. I just picked out
one little piece.
> What I /guess/ he did is take his entire "difference equation" expression
> for P(x, y), and compute P(x, y) - P(x, y-1) to get exactly (all but one of
> the terms in the expansions of the two sums cancel out):
>
> P(x, y) - P(x, y-1) =
> -[P(x/y, y-1) - P(y-1, sqrt(y-1)] * deltaP(y, y)
>
> Or perhaps he started from an exact expression for P(x, y+1) - P(x, y)
> instead.
>
> Regardless, he then tries computing the partial derivative wrt y, and
> various forms of magical thinking permit ignoring the chain rule in the part
> you identified, but also ignoring the chain rule in the P(x/y, y-1) part;
> and in the P(y-1, sqrt(y-1)) part (repeated in the expansion of the
> deltaP(y, y) part); ignoring the differences between y and y-1; and, for the
> "coup de grace", wishing away that he /knows/ deltaP(y, y) was torturously
> contrived to be an indicator function (0 if y composite, 1 if y prime), so
> that the notion he's going to get a lick of sense out of this by presuming
> to differentiate it is exceedingly dubious on the face of it.
>
Right. It makes no sense. Even the little pieces of it make no
sense. I was basically focussing on one little piece.
> > and, passing to the limit (replacing the "-1" by "dk" and
> > then replacing "k" by "y", in your last equation, you end
> > up with
> >
> > dP(y, sqrt(y))/dy.
> >
> > However, the derivative of P(k,sqrt(k)) should be
> >
> > (dP/dx1) + (dP/dx2)*(.5*y^(-.5)),
> >
> > where dP/dx1 and dP/dx2 represent partial derivatives of P with
> > respect to the first and second arguments respectively. and these
> > are evaluated at x1 = y, x2 = y.
> >
> > So your implied claim to have numerically integrated to find
> > a solution is obviously false.
>
> Hard to say because he hasn't shown any of his work.
P(x, y) is an unknown function of two variables. No matter
what he does, the derivative of P(k, k) or P(k, sqrt(k)) will
have to involve the partial derivatives of P with respect to
each of those variables.
> As David Ullrich has
> repeatedly pointed out to him over a span of years now, numerical
> integration can deliver all sorts of nonsense depending not just on the
> function, but also on the method and step size chosen. It's quite
> conceivable to me that James thrashed at random until he stumbled into some
> misunderstanding of some numeric method using an inappropriae step size
> that, e.g., got close to 4 when he crunched the numbers at x=10. Woo hoo!
> HUGE result.
>
One cannot know for sure what he means by "numerical integration".
However, I would guess he replaces the "-1" by "-0.5", "-0.2", "-0.1",
etc., and does his recursive computation. Of course with -1, he gets
the expected answer. I would guess that as dx --> 0, you get farther
and farther away from the prime distribution function.
David has shown that solutions to the differential equation
associated
with a given recursive function need not be the original function.
> What is certain is that he hasn't /proved/ a damn thing about it, so that
> his:
>
> and a numerical integration of that partial derivative then
> answers the final questions about the relation between the prime
> distribution, and continuous functions.
>
> is pure bluff, which he often does on Usenet.
B-b-b-b-but he has invariably been right in the past! How can
you be such a doubter?
> In the context of
> publication, "fraud" is a more accurate term.
>
> > For this reason alone the paper should be withdrawn.
>
> Well, let's not go overboard ;-)
Well, OK. Revised and resubmitted. Well, no.
Actually, I don't think he should be encouraged at all to
keep submitting papers for publication. Editors work hard and
have a great many serious papers to deal with. They are
not servants and their time should not be wasted. Wasting
people's time here is different: it's not anyone's job to respond
to people like Harris - it's strictly voluntary. He has been
given every chance and he has a track record of 100%
failure. But his "work" is very likely much more widely read
than that of Wiles. He of course thinks the editors of math
journals "owe" him a chance. They are probably nice polite
people who will tell him gently that his paper is not suitable
for the Annals. Saying any more than that invites an
interminable argument from him. What they SHOULD say is,
"Your work is not suitable for lining a bird cage."
Marcus.
jst...@msn.com
Why?
It's not even dx as the key variable but dy.
Extreme mathematics has a brainstorming phase followed by a critiquing
phase.
I brainstormed the derivation out on math newsgroups YEARS ago, and
opened up the floor to critiques, where it was noted that I did one
piece wrong as I looked for a partial derivative on the right side as I
had one on the left.
So I'd just kind of tried to fudge as my gut feeling was that I needed
all partial derivatives and members of the group jumped on that
mistake, and I fixed it.
That was YEARS ago.
Now you think you can walk in and make all kinds of wacky claims about
the derivation and have any weight against a dedicated and rigorous
process that has gone through several iterations over a period of
years?
Where mostly you just babble support for concepts other posters have
half-assed thrown out there like Ullrich's, duh, well, you can duh,
have a difference equation, uh, and that difference equation, can uh,
have a differential equation that duh, doesn't look any thing like it's
duh, sum, uh huh huh, duh...as if that matters?
You were brought up on out-dated ways of problem solving.
Your ways do not work with modern problems as they are too hard for the
old stuff.
The mathematical world has become too complex for those medieval ways.
It is past time for the math world to evolve to techniques that work.
And here you see how rigid extreme mathematics is in the later phases
with an idea that has gone through several iterations and several
brainstorming and critiquing phases in that your half-assed, b.s.
attack is just a way for you to look like an infant among adults in
mathematics.
Extreme mathematics works.
And it's damn hard to do right as well, but when it's done right, it's
almost impossible to find any kind of error with the end product.
James Harris
jshs...@yahoo.com
You had another typo there James. You meant that Extreme Mathematics
doesn't work.
No need to thank me.
marcus_b
Yes - I should have said "dy" to be consistent with what you
wrote.
> Extreme mathematics has a brainstorming phase followed by a critiquing
> phase.
>
> I brainstormed the derivation out on math newsgroups YEARS ago, and
> opened up the floor to critiques, where it was noted that I did one
> piece wrong as I looked for a partial derivative on the right side as I
> had one on the left.
>
You still have it wrong.
> So I'd just kind of tried to fudge as my gut feeling was that I needed
> all partial derivatives and members of the group jumped on that
> mistake, and I fixed it.
>
It's still wrong. Did anyone ever say you got it right?
> That was YEARS ago.
>
> Now you think you can walk in and make all kinds of wacky claims about
> the derivation and have any weight against a dedicated and rigorous
> process that has gone through several iterations over a period of
> years?
>
Yep. Prove me wrong.
> Where mostly you just babble support for concepts other posters have
> half-assed thrown out there like Ullrich's, duh, well, you can duh,
> have a difference equation, uh, and that difference equation, can uh,
> have a differential equation that duh, doesn't look any thing like it's
> duh, sum, uh huh huh, duh...as if that matters?
>
Of course it matters. Ullrich showed that the solution to the
differential equation corresponding to a difference equation is
not necessarily close to a solution to the original equation.
What you want is a nice function which is close to pi(n). You
have to prove that your process produces that. It doesn't happen
automatically - that is the point of Ullrich's example. You have n
ever given the slightest shred of a proof.
> You were brought up on out-dated ways of problem solving.
>
You have no idea how I was brought up.
> Your ways do not work with modern problems as they are too hard for the
> old stuff.
>
Some things don't change. Like the chain rule. Or are you
saying you have a proof that the chain rule is wrong?
> The mathematical world has become too
> complex for those medieval ways.
>
You wouldn't know. The only math you do, the quadratic
formula, is medieval. You have not yet caught up even
to Euler.
> It is past time for the math world to evolve to techniques that work.
>
> And here you see how rigid extreme mathematics is in the later phases
> with an idea that has gone through several iterations and several
> brainstorming and critiquing phases in that your half-assed, b.s.
> attack is just a way for you to look like an infant among adults in
> mathematics.
>
Could be. But your expression for the derivative is trivially
wrong.
> Extreme mathematics works.
>
Extreme or not, basic differential calculus still applies.
> And it's damn hard to do right as well, but when it's done right, it's
> almost impossible to find any kind of error with the end product.
>
Sure. But what you have is just plain wrong. Go back and
review the chain rule.
The editors at the Annals will take one look at what you wrote
and trash it instantly. You might as well save yourself some
embarrassment.
Marcus.
>
> James Harris
jst...@msn.com
Read the derivation. I don't give it in the paper as it's trivial
enough for expert mathematicians, but it is on my math blog.
James Harris
sg...@hotmail.co.uk
> Now you think you can walk in and make all kinds of wacky claims about
> the derivation and have any weight against a dedicated and rigorous
> process that has gone through several iterations over a period of
> years?
You simply have no idea what constitutes rigour, mathematical or
otherwise.
> Where mostly you just babble support for concepts other posters have
> half-assed thrown out there like Ullrich's, duh, well, you can duh,
> have a difference equation, uh, and that difference equation, can uh,
> have a differential equation that duh, doesn't look any thing like it's
> duh, sum, uh huh huh, duh...as if that matters?
Well, duh, uh huh huh, duh, yes it does matter. It matters a fucking
hell of a lot. If you are going to claim that your partial differential
equation answers some, duh, uh, unspecified questions about the prime
distribution, you have to show that the solution of that equation
*actually bears some resemblance to the prime distribution*. Prof.
Ullrich's example shows that there is no a priori reason to believe
that this is the case. Duh.
> You were brought up on out-dated ways of problem solving.
Yes, those outdated methods that require grandiose claims to have
actual proofs. How much easier it would be to make progress with
mathematics if we could all just pull equations out of our arses and
declare the Riemann Hypothesis false without actually proving anything
about those equations. We'd certainly have many exciting new results if
we used extreme mathematics. On the downside, they'd mostly be wrong.
> Your ways do not work with modern problems as they are too hard for the
> old stuff.
How the hell would you know what works with modern problems? You don't
have a clue about what has been going on in mathematics for the past
few centuries. Can you name a single branch of mathematics you have
learnt about that wasn't already known during the time of Gauss?
> The mathematical world has become too complex for those medieval ways.
The only one here who is stuck in medieval ways is you. You haven't
made the slightest attempt to find out about any even slightly recent
developments in mathematics. You would rather sit at home playing with
the quadratic formula than read up on results that aren't known to a
twelve-year-old. You would rather declare that ring theory is "excess
baggage" than actually learn a damn thing about ring theory, and how it
is used by mathematicians and physicists alike to prove useful results.
> It is past time for the math world to evolve to techniques that work.
I read recent mathematics on a regular basis, and to me it looks like
the techniques that mathematicians are using at present work just fine.
On what grounds do you claim that they don't work?
> And here you see how rigid extreme mathematics is in the later phases
> with an idea that has gone through several iterations and several
> brainstorming and critiquing phases
Doesn't look that rigid to me. You're still claiming that a PDE answers
questions about the prime distribution without having proved that the
solution has anything to do with the prime distribution, or even that
the solution exists. You're still claiming that the algebraic integers
have a "coverage problem" without being able to define what a coverage
problem actually is.
> in that your half-assed, b.s.
> attack is just a way for you to look like an infant among adults in
> mathematics.
Looks that way to you, huh?
> Extreme mathematics works.
No, it doesn't. You have produced nothing even vaguely interesting.
> And it's damn hard to do right as well, but when it's done right, it's
> almost impossible to find any kind of error with the end product.
In that case you have yet to do it right.
-Duh, uh huh, Rotwang
Tim Peters
> Read the derivation.
I suppose you mean this?
<http://mymath.blogspot.com/2005/06/partial-differential-prime-counting.html>
> I don't give it in the paper as it's trivial enough for expert
> mathematicians, but it is on my math blog.
As marcus_b said the first time, you neglected to apply the chain rule. As
I elaborated, you neglected to apply it in several places.
Since, as marcus_b also said, that's at the level of freshman calculus,
explaining that to you is likely hopeless.
Instead I'll just point out that your blog at least remembered to include
the necessary floor function, the lack of which in your paper renders the
latter wrong starting with its first equation.
I'd explain more, except my hands are shaking too badly feom fear ;-)
junoexpress
be extremely interesting to read. (Unfortunately, I don't think that it
will it even make it that far, as the reviewers are busy reading real
math papers, and I doubt that your "paper" will be deemed worthy of
their time). I don't know if you have ever had a paper rejected "with
prejudice" (from the looks of your writing, it seems to be limited to
math.sci and bathroom walls), but it is _not_ a pleasant experience. I
hope that you are as forthcoming with their comments as you have been
about the paper itself: I am sure that some of them will be pure gems.
The fun has just begun.
Matt B.
jst...@msn.com wrote:
> Larry Hammick wrote:
> > <jst...@msn.com>
> > >I wrote another paper and sent it to the Annals of Mathematics which
> > > verified receipt.
> > Conjecture: When they reject it, you will blame the sinister cadre of
> > sci-dot-math (again).
>
> No. I know and you know that Princeton doesn't give a damn what you
> think.
>
> There is only one case where sci.math definitely interfered and killed
> a journal and that is with SWJPAM.
>
> There is no way that Princeton would fall in the same way.
>
> You people can't kill the Annals of Mathematics.
>
> I dare you to try. Send your emails. Just try.
>
>
> James Harris
David Moran
How come a class in extreme mathematics is not offered in math departments?
Simple, there's no such thing, except in your pathetic little world.
Dave
jst...@msn.com
Tim Peters wrote:
> [jst...@msn.com]
> > Read the derivation.
>
> I suppose you mean this?
>
> <http://mymath.blogspot.com/2005/06/partial-differential-prime-counting.html>
>
> > I don't give it in the paper as it's trivial enough for expert
> > mathematicians, but it is on my math blog.
>
> As marcus_b said the first time, you neglected to apply the chain rule. As
> I elaborated, you neglected to apply it in several places.
>
Ok, I'll bite, then what do you claim is the correct equation from the
proper derivation?
> Since, as marcus_b also said, that's at the level of freshman calculus,
> explaining that to you is likely hopeless.
Why explain? Give what you think is the correct derivation.
> Instead I'll just point out that your blog at least remembered to include
> the necessary floor function, the lack of which in your paper renders the
> latter wrong starting with its first equation.
>
With x and y declared to be natural numbers that's the ring. All is
correct within the ring of natural numbers where floor() is redundant.
> I'd explain more, except my hands are shaking too badly feom fear ;-)
You're finally proven to be incompetent and a bad liar--or, more
brilliant than me.
If so, give the derivation I missed.
James Harris
jshs...@yahoo.com
James, it has been proven that he is more brilliant then you since
about his second reply to you.
It has also been proven that you are also incompetent and a bad liar.
sg...@hotmail.co.uk
> > Instead I'll just point out that your blog at least remembered to include
> > the necessary floor function, the lack of which in your paper renders the
> > latter wrong starting with its first equation.
>
> With x and y declared to be natural numbers that's the ring. All is
> correct within the ring of natural numbers where floor() is redundant.
Nice try, but it doesn't work like that. For a start, the natural
numbers are *not a ring*. Also there is no division in the natural
numbers. If your equations contain division, it will be assumed that
you are working in at least the rationals by anyone reading the paper;
there is simply no generally recognised convention that division of
integers means floor division. Also, later in your paper, you talk
about differential equations; this is meaningless in the naturals,
which suggests that you are working in the reals.
> You're finally proven to be incompetent and a bad liar--or, more
> brilliant than me.
Hint: it's the second one.
-Rotwang
Davie
<jst...@msn.com> wrote in message
news:1164402802....@l12g2000cwl.googlegroups.com...
> Rupert wrote:
*fear reeking* Thats ripe, but not for a paranoid.
>
> It is more than a great enough accomplishment to merit publication in
> one of the world's top math journals.
>
*Delusional* (JSH dosen't read "pure math" journals.)
>
> James Harris
>
Tim Peters
>>> Read the derivation.
[Tim Peters]
>> I suppose you mean this?
>>
>> <http://mymath.blogspot.com/2005/06/partial-differential-prime-counting.html>
>>> I don't give it in the paper as it's trivial enough for expert
>>> mathematicians, but it is on my math blog.
>> As marcus_b said the first time, you neglected to apply the chain
>> rule. As I elaborated, you neglected to apply it in several places.
> Ok, I'll bite, then what do you claim is the correct equation from
> the proper derivation?
Nope, first /you/ make an effort to understand what the chain rule is, and
why it might be that people say you've neglected it. If you haven't noticed
yet, I have: your "challenges" are easy to meet, but solutions go whizzing
over your head. Especially in this case, as I said a few msgs back:
and, for the "coup de grace", wishing away that he /knows/
deltaP(y, y) was torturously contrived to be an indicator
function (0 if y composite, 1 if y prime), so that the notion
he's going to get a lick of sense out of this by presuming
to differentiate it is exceedingly dubious on the face of it.
That is, this is a fool's errand from the word "go!".
>> Since, as marcus_b also said, that's at the level of freshman
>> calculus, explaining that to you is likely hopeless.
> Why explain? Give what you think is the correct derivation.
You gave your game away with that, James. If you were interested in truth
or understanding, the /explanation/ would have been everything to you.
>> Instead I'll just point out that your blog at least remembered to
>> include the necessary floor function, the lack of which in your
>> paper renders the latter wrong starting with its first equation.
> With x and y declared to be natural numbers that's the ring. All is
> correct within the ring of natural numbers where floor() is redundant.
In your fantasies, perhaps, but not in standard mathematics. You think
equation [1] at:
http://mathworld.wolfram.com/LegendresFormula.html
bothered slopping in all the "floor" symbols because the author enjoyed
tedious typesetting? This is mathematics, NOT A PROGRAMMING LANGUAGE with
implicit value-changing type conversions. If you want to lose fractional
parts in math, you need to do so /explicitly/. Else 17/2 is 8.5, period.
>> I'd explain more, except my hands are shaking too badly
>> [from] fear ;-)
> You're finally proven to be incompetent and a bad liar--
Consistency, James: I've personally lost track of how often you've
announced you've proven that. Suddenly it's "finally"? LOL.
> or, more brilliant than me.
I certainly have a deeper and broader understanding of math than you have.
"Who's more brilliant?" is one of those irrelevant whose-dick-is-bigger
games you should have noticed that: (1) I never /start/; and, (2) you never
win.
> If so, give the derivation I missed.
Try to figure it out yourself. Here's a practice problem: suppose
f(x, y) = g(x, y^2)
What's the partial derivative of f w.r.t. y? Do you understand that a
correct answer /must/ contain "2*y"? Do you understand the same must be
true if you slop in a "delta(y)" symbol and "take the limit" instead? If
you don't understand those, read a book. If you do understand those, why
did you ignore the same issues in your PDE derivation?
jst...@msn.com
Tim Peters wrote:
> [jst...@msn.com]
> >>> Read the derivation.
>
> [Tim Peters]
> >> I suppose you mean this?
> >>
> >> <http://mymath.blogspot.com/2005/06/partial-differential-prime-counting.html>
>
> >>> I don't give it in the paper as it's trivial enough for expert
> >>> mathematicians, but it is on my math blog.
>
> >> As marcus_b said the first time, you neglected to apply the chain
> >> rule. As I elaborated, you neglected to apply it in several places.
>
> [jst...@msn.com]
> > Ok, I'll bite, then what do you claim is the correct equation from
> > the proper derivation?
>
> Nope, first /you/ make an effort to understand what the chain rule is, and
> why it might be that people say you've neglected it. If you haven't noticed
> yet, I have: your "challenges" are easy to meet, but solutions go whizzing
> over your head. Especially in this case, as I said a few msgs back:
Then meet a single challenge.
I just don't like useless discussions with you all over the map never
saying anything so I gave you a direct way to prove your point, by
giving a correct derivation if mine were actually wrong.
>
> and, for the "coup de grace", wishing away that he /knows/
> deltaP(y, y) was torturously contrived to be an indicator
> function (0 if y composite, 1 if y prime), so that the notion
> he's going to get a lick of sense out of this by presuming
> to differentiate it is exceedingly dubious on the face of it.
>
> That is, this is a fool's errand from the word "go!".
How many times do I have to point out I already did my own numerical
integration which showed that integrating the partial differential
equations gives a better fit to the prime distribution than Li(x) but
was just a bit further out than R(x)?
Now I don't consider myself to be an expert on numerical integration,
so I've looked to others to verify--in keeping with my training as I
have a degree in physics.
To date though for reasons that still escape me to my knowledge math
people have avoided actually integrating that partial differential
equation, almost as if they fear looking too deeply into it as if it
opens the door into Hell itself.
James Harris
David Moran
<jst...@msn.com> wrote in message
news:1164512414....@f16g2000cwb.googlegroups.com...
For someone with a degree in physics, you sure have a lousy understanding
of, what I think, is basic math.
Dave
Davie
<jshs...@yahoo.com> wrote in message
news:1164498625....@j44g2000cwa.googlegroups.com...
By definition Extreme Mathematics must contain errors, as Pure Mathematics
does not.
So JSH has chosen the path of bogus math.
(is that a Proof? that JSH does bogus math?)
Davie
"David Moran" <dmor...@NOSPAMcox.net> wrote in message
news:w48ah.3278$hz3....@newsfe21.lga...
He does not have a degree in physics, he took some courses in physics, he
let that slip a few months ago.
probably high school physics
he is at a high school math level, doesn't know the chain rule, cannot
integrate.....
Troll Crackpot
Davie
"David Moran" <dmor...@NOSPAMcox.net> wrote in message
"extreme" needs to be replaced with "gumbie"
marcus_b
OK, here's a direct quote from the 'derivation' on your 'My Math'
blogsite:
S(x,y) = (P(x/y,y-y) - P(y - y,sqrt(y-y)))*
(P(y,sqrt(y)) - P(y-y,sqrt(y-y))).
Now, last time I checked, y-y=0. So this should reduce to
S(x,y) = (P(x/y,0) - P(0, 0)))(P(y,sqrt(y)) - P(0, 0)).
But that's not what you get. Instead, you divide by y to get
S(x,y)/y = (P(x/y,y-y) - P(y-y,sqrt(y-y)))*
(P(y,sqrt(y)) - P(y-y,sqrt(y-y)))/y,
which so far is not wrong ... though it could be simplified. Here's
what you actually INTENDED to write:
S(x,y)/dy = (P(x/y,y-dy) - P(y-dy,sqrt(y-dy)))*
(P(y, sqrt(y)) - P(y-dy,sqrt(y-dy)))/dy.
Now you want to let dy -->0.
On the left, you say you get S'_y(x, y).
This is wrong. The derivative of S(x, y) with respect
to y by definition is
dS(x,y)/dy = lim ((S(x, y) - S(x, y-dy))/dy.
dy->0
What you have is outright wrong because in general
lim S(x,y)/dy
dy->0
is infinity.
OK, now for the right side. Let's look in particular at
lim (P(y, sqrt(y)) - P(y-dy,sqrt(y-dy)))/dy.
dy->0
This is where the chain rule comes in. After some
manipulations, you are going to end up with
dP_1(y, sqrt(y)) + dP_2(y, sqrt(y))*(.5/sqrt(y)),
where dP_1 and dP_2 are the partial derivatives of the
function P with respect to its first and second
arguments respectively. This is really very far from
what you wrote down. If you can't figure this out,
ask for help.
Bottom line: your derivation is totally, hopelessly
screwed up. Neither side of your equation is even
close to correct, and the left side is infinity.
This is basic, first-year differential calculus, man!
Don't give me this crap about 'Extreme Mathematics'!
This is sheer, rank incompetence, stacked on top of
Extreme _Carelessness_! You think the editors of the Annals
are not going to see through this instantly? Do you
think they're idiots???
Marcus.
>
> James Harris
Proginoskes
Too difficult for him. Choose something simple like g(x, y) = 2 x + 3
y, and have him calculate the partial derivative of f with respect to y
by doing the full substitution, then comparing it with the partial
derivative of g with respect to y.
After all, he only has a B.S.[1] in physics. But any B.S. in physics
would certainly require Calculus III, which is where this problem comes
from. (I also teach Calc III, so I know what problems students should
be able to do to pass that class.)
--- Christopher Heckman
[1] JSH continually uses "B.Sc." in his posts. My guess is he's trying
to avoid mention of the abbreviation "B.S." in his posts where he can;
if he used "B.S." instead, that would subconsciously "taint" the rest
of the post.
Proginoskes
If you're not going to derive it in the paper, the *courteous* thing to
do is include a reference where it _is_ derived. For those of you who
don't know how to author math papers (JSH in particular), that means
put in a footnote, then a book, journal, or website (which would be a
last-ditch effort) where the derivation is given.
I'd bet my weekly salary against JSH's that if a mathematician looks at
that formula and knows immediately how it is derivated, he/she is
thinking, "Oh, this is a variation of Lagrange's (or Meissel's, etc)
Method."
--- Christopher Heckman
Proginoskes
Well, the corrected version of it _is_. I've been teaching such a class
for the last seven semesters or so; see
http://math.asu.edu/~checkman/x94/ for a description of this class.
--- Christopher Heckman
Tim Peters
[jst...@msn.com]
>>> Ok, I'll bite, then what do you claim is the correct equation from
>>> the proper derivation?
[Tim Peters]
>> Nope, first /you/ make an effort to understand what the chain rule
>> is, and why it might be that people say you've neglected it. If
>> you haven't noticed yet, I have: your "challenges" are easy to meet,
>> but solutions go whizzing over your head. Especially in this case,
>> as I said a few msgs back:
> Then meet a single challenge.
No. It's clear to me that you have no idea how to proceed here, and you
ignored my much simpler challenge to compute the partial derivative w.r.t. y
of f(x, y) = g(x, y^2). Proginoskes suggests that's probably too hard for
you, and I expect he's right, but if you can't do that correctly it's
impossible that you could even begin to follow a correct derivation starting
from your much messier formula.
> I just don't like useless discussions with you all over the map never
> saying anything so I gave you a direct way to prove your point, by
> giving a correct derivation if mine were actually wrong.
But you have no way to know whether it's correct or not, because you've:
(a) already failed to do it even close to correctly yourself; and, (b)
ignored every technical point raised in favor of mindlessly repeating your
"challenge". I've seen this act from you 1000 times before -- no matter
what /technical/ response I'd give, you'd either ignore it or start another
"liar! liar!" rant.
If you're capable of doing better than that, reply to marcus_b, who has
since /started/ you down a correct path. He's apparently willing so far to
expend more effort on your behalf than I am here.
...
>> and, for the "coup de grace", wishing away that he /knows/
>> deltaP(y, y) was torturously contrived to be an indicator
>> function (0 if y composite, 1 if y prime), so that the notion
>> he's going to get a lick of sense out of this by presuming
>> to differentiate it is exceedingly dubious on the face of it.
>>
>> That is, this is a fool's errand from the word "go!".
> How many times do I have to point out I already did my own numerical
> integration which showed that integrating the partial differential
> equations gives a better fit to the prime distribution than Li(x) but
> was just a bit further out than R(x)?
>
> Now I don't consider myself to be an expert on numerical integration,
> so I've looked to others to verify--in keeping with my training as I
> have a degree in physics.
>
> To date though for reasons that still escape me to my knowledge math
> people have avoided actually integrating that partial differential
> equation, almost as if they fear looking too deeply into it as if it
> opens the door into Hell itself.
Do you live on a diet of 100% melodramatic fantasy?
If you want help with numeric integration, show your work: post the code
you used to perform your numeric integration. There are many possible
methods you may have tried, and many ways you may have failed to implement
them correctly, and it's impossible to /guess/ what you did. David Ullrich
has explained to you some of the subtleties involved, but you didn't appear
to grasp them. It's quite possible for a naive approach to /appear/ to
produce just about any results imaginable, and especially if you "tuned"
what you were doing when initial results weren't close to what you hoped to
get. This can be true even if the PDE you arrived at is dead wrong (which
yours certainly is).
Tim Peters
> ...
> Right. It makes no sense. Even the little pieces of it make no
> sense. I was basically focussing on one little piece.
And, if you let him, James will keep you on that one little piece for years
;-)
...
[marcus_b]
>>> So your implied claim to have numerically integrated to find
>>> a solution is obviously false.
[Tim Peters]
>> Hard to say because he hasn't shown any of his work.
> P(x, y) is an unknown function of two variables. No matter
> what he does, the derivative of P(k, k) or P(k, sqrt(k)) will
> have to involve the partial derivatives of P with respect to
> each of those variables.
This was a different point: he hasn't shown any of his /numerical
integration/ work. It's possible that, with some effort, he came up with
bogus numeric results he liked /despite/ starting with a wrong PDE. I'm
unwilling to assume he did numeric integration in a sensible way. I can't
guess what he did, but if he did manage to get numeric results close to
pi(x), then it's essential to see his code in order to understand how that
came about.
...
> One cannot know for sure what he means by "numerical integration".
> However, I would guess he replaces the "-1" by "-0.5", "-0.2", "-0.1",
> etc., and does his recursive computation.
Sounds more like numeric differentiation than numeric solution of a PDE.
But can't rule that out either ;-)
...
> David has shown that solutions to the differential equation
> associated with a given recursive function need not be the original
> function.
Well, if the derivative of a function "makes sense", the function has to
satisfy whatever equation(s) you derive involving the derivative. In the:
a(0) = 1
a(n+1) = 2*a(n)
example, what /justifies/ going from the derived:
a(n+1) - a(n) = a(n)
to:
a'(n) = a(n)
to begin with? "Divide both sides by 1 then pretend 1 approaches 0 on the
left but not the right" isn't terribly satisfying ;-) Flesh it out and "the
problem" goes away; e.g.,
a(0) = 1
a(n+d) = 2^d * a(n)
is one way of "fleshing it out" that obviously says the same thing at d=1.
But then:
a(n+d) - a(n) = (2^d - 1)*a(n)
so:
(a(n+d) - a(n))/d = (2^d - 1)/d * a(n)
Then we can legitimately ask what happens as d -> 0, to get:
a'(n) = ln(2) * a(n)
and the general solution to that is the expected a(n) = C*2^n.
I think an obvious problem with James's formula w.r.t. similar fiddling is
that it's discontinuous: to get anywhere sane, he needs to "flesh it out"
too, at least so that derivatives /exist/. It's certainly true that, as
David's example illustrated, wishing that away and then merely /hoping/ that
a carelessly derived PDE will magically fill in the gaps is unlikely to work
worth beans.
...
>>> For this reason alone the paper should be withdrawn.
>> Well, let's not go overboard ;-)
> Well, OK. Revised and resubmitted. Well, no.
>
> Actually, I don't think he should be encouraged at all to
> keep submitting papers for publication. Editors work hard and
> have a great many serious papers to deal with. They are
> not servants and their time should not be wasted. Wasting
> people's time here is different: it's not anyone's job to respond
> to people like Harris - it's strictly voluntary. He has been
> given every chance and he has a track record of 100%
> failure. But his "work" is very likely much more widely read
> than that of Wiles. He of course thinks the editors of math
> journals "owe" him a chance. They are probably nice polite
> people who will tell him gently that his paper is not suitable
> for the Annals. Saying any more than that invites an
> interminable argument from him. What they SHOULD say is,
> "Your work is not suitable for lining a bird cage."
Alas, that only happens on Usenet. It's clear that James doesn't recognize
a polite brush-off when he gets one, but instead sees at as (a) "they didn't
find any problem in my work!"; and, (z) "so why are they silent -- ah,
conspiracy!".
However, because it's profoundly mistaken from A to Z, I have no hope that
any rational input can change any of it. If mathematicians telling him for
years and years and years that the same old stuff is still wrong in the same
old ways isn't sufficient discouragement from submitting papers, why
possibly /could/ be? Really. If an absued editor /did/ say "your work is
not suitable for lining a bird cage", I'm afraid the reaction would be
"excellent! the lying scum are finally showing their true colors in a way I
can prove to everyone!".
That is, same thing he says about everyone here. Hell, although not very
recently that I can recall, he's already made the connection that "post hoc
ergo propter hoc" Wiles is a major force at Princeton, and sucking him into
a conspiracy to keep James's work out of the Annals should be baby play for
JSH.
W. Dale Hall
... stuff deleted ...
>
> He does not have a degree in physics, he took some courses in physics, he
> let that slip a few months ago.
> probably high school physics
> he is at a high school math level, doesn't know the chain rule, cannot
> integrate.....
> Troll Crackpot
>
>
I don't recall any such article. Do you have a reference for this?
Dale.
David C. Ullrich
>
>marcus_b wrote:
>> [...]
>>
>> Sure. But what you have is just plain wrong. Go back and
>> review the chain rule.
>>
>> The editors at the Annals will take one look at what you wrote
>> and trash it instantly. You might as well save yourself some
>> embarrassment.
>>
>> Marcus.
>
>Read the derivation. I don't give it in the paper as it's trivial
>enough for expert mathematicians, but it is on my math blog.
Saying "read the derivation" doesn't change the fact that the
derivation was wrong.
More important: I never worried about the "derivation". Because
even if you _did_ give a proper "derivation", that's still no
reason whatever to think that the solution to the "pde" has
anything at all to do with pi.
There's no problem "deriving"
f' = f, f(0) = 1
from the recurrence
a[n+1] - a[n] = a[n], a[0] = 1.
But even though _that_ "derivation" is correct, the solution
of the de still has nothing to do with the solution to the
recurrence.
>
>James Harris
************************
David C. Ullrich
David C. Ullrich
wrote:
Of course nothing justifies that - that's been my point.
This is nonetheless the sort of thing that was done in
"deriving" the famous Harris so-called "pde". Or that's
my impression - _did_ he in fact do something more
sensible, as below? [Turns out that's a rhetorical
question, answered below.]
> "Divide both sides by 1 then pretend 1 approaches 0 on the
>left but not the right" isn't terribly satisfying ;-) Flesh it out and "the
>problem" goes away; e.g.,
>
> a(0) = 1
> a(n+d) = 2^d * a(n)
>
>is one way of "fleshing it out" that obviously says the same thing at d=1.
>But then:
>
> a(n+d) - a(n) = (2^d - 1)*a(n)
>
>so:
>
> (a(n+d) - a(n))/d = (2^d - 1)/d * a(n)
>
>Then we can legitimately ask what happens as d -> 0, to get:
>
> a'(n) = ln(2) * a(n)
>
>and the general solution to that is the expected a(n) = C*2^n.
>
>I think an obvious problem with James's formula w.r.t. similar fiddling is
>that it's discontinuous: to get anywhere sane, he needs to "flesh it out"
>too, at least so that derivatives /exist/. It's certainly true that, as
>David's example illustrated, wishing that away and then merely /hoping/ that
>a carelessly derived PDE will magically fill in the gaps is unlikely to work
>worth beans.
I guess he didn't, then. Didn't think so.
************************
David C. Ullrich
marcus_b
> [marcus_b]
> > ...
> > Right. It makes no sense. Even the little pieces of it make no
> > sense. I was basically focussing on one little piece.
>
> And, if you let him, James will keep you on that one little piece for years
> ;-)
We both know that if you can reduce his argument to the level
of about 9th grade math, Harris will abandon the field. In this
case however there is a lot at stake (for him). In the past, shown
errors in a paper he has submitted, he refused to withdraw it.
Here this particular error is too big, too central to the paper. He
cannot brush it aside. This puts him in an untenable position:
he issued a challenge to criticize it and we showed it is no good
in a way that he cannot ignore; but he cannot lose face to the
extent of withdrawing the paper. Best to leave it in the hands of
the editors in the hope that they won't notice. The immediate
effect is: he is going to squeal like the proverbial stuck pig.
>
> ...
>
> [marcus_b]
> >>> So your implied claim to have numerically integrated to find
> >>> a solution is obviously false.
>
> [Tim Peters]
> >> Hard to say because he hasn't shown any of his work.
>
> > P(x, y) is an unknown function of two variables. No matter
> > what he does, the derivative of P(k, k) or P(k, sqrt(k)) will
> > have to involve the partial derivatives of P with respect to
> > each of those variables.
>
> This was a different point: he hasn't shown any of his /numerical
> integration/ work. It's possible that, with some effort, he came up with
> bogus numeric results he liked /despite/ starting with a wrong PDE. I'm
> unwilling to assume he did numeric integration in a sensible way. I can't
> guess what he did, but if he did manage to get numeric results close to
> pi(x), then it's essential to see his code in order to understand how that
> came about.
>
I completely agree.
> ...
>
> > One cannot know for sure what he means by "numerical integration".
> > However, I would guess he replaces the "-1" by "-0.5", "-0.2", "-0.1",
> > etc., and does his recursive computation.
>
> Sounds more like numeric differentiation than numeric solution of a PDE.
Doing the computation with dy = -1 produces pi(n). It is natural -
more or less - to simply modify his program, which correctly
computes pi(n), by substituting smaller and smaller values of "dy".
It's numeric at least, though not exactly integration.
> But can't rule that out either ;-)
>
> ...
>
> > David has shown that solutions to the differential equation
> > associated with a given recursive function need not be the original
> > function.
>
> Well, if the derivative of a function "makes sense", the function has to
> satisfy whatever equation(s) you derive involving the derivative. In the:
>
> a(0) = 1
> a(n+1) = 2*a(n)
>
> example, what /justifies/ going from the derived:
>
> a(n+1) - a(n) = a(n)
>
> to:
>
> a'(n) = a(n)
>
> to begin with? "Divide both sides by 1 then pretend 1 approaches 0 on the
> left but not the right" isn't terribly satisfying ;-)
That was the point I made in my later post about his dS_y(x, y)
computation: the denominator is zero, the numerator in general
is not.
> Flesh it out and "the
> problem" goes away; e.g.,
>
> a(0) = 1
> a(n+d) = 2^d * a(n)
>
> is one way of "fleshing it out" that obviously says the same thing at d=1.
> But then:
>
> a(n+d) - a(n) = (2^d - 1)*a(n)
>
> so:
>
> (a(n+d) - a(n))/d = (2^d - 1)/d * a(n)
>
> Then we can legitimately ask what happens as d -> 0, to get:
>
> a'(n) = ln(2) * a(n)
>
> and the general solution to that is the expected a(n) = C*2^n.
>
> I think an obvious problem with James's formula w.r.t. similar fiddling is
> that it's discontinuous: to get anywhere sane, he needs to "flesh it out"
> too, at least so that derivatives /exist/. It's certainly true that, as
> David's example illustrated, wishing that away and then merely /hoping/ that
> a carelessly derived PDE will magically fill in the gaps is unlikely to work
> worth beans.
>
That's the central problem. But _hoping for a magical solution_ is
a well-worn Harrisian pattern.
They can be a lot less namby-pamby without actually using the
bird-cage phrase - e.g.,
"The editorial board has examined your paper. It includes no
statements of theorems and no proofs or coherent arguments.
Further, because it is based on a number of elementary and
conceptual errors, it is completely lacking in mathematical content
and merit. We will not consider a resubmission of this paper."
The polite brush-off just sustains his delusions, and that is what
has happened in his previous encounters with editors and, e.g.,
Granville, Lagarias, Odlyzko, Mazur, the prof at Vanderbilt, etc.,
etc.. Someone up there needs to get tough with him.
At one time I thought it would be useful to tell him to stop his
endless whining on sci.math and submit a paper to a journal.
So you know what has happened there. Now he is wasting the
valuable time of editors and reviewers for no good reason whatsoever.
Marcus.
Jesse F. Hughes
There is no such article.
In <1157308158.7...@i42g2000cwa.googlegroups.com>, James
wrote, "Oddly enough to me, as when I was a physics student years
ago..."
The brilliant Dr. Moria (same guy as Davie) saw through the ruse,
writing, "HUGE slip-up! you were only a student, and you never got a
degree!"
It was a stupid claim then, but this idiot has shown he likes really
stupid claims. Like James is a female librarian/English teacher and
James hates Poles (since he used the word "polish" as in "refinement"
in a post).
--
"Now, once [James's research] is accepted, number theory is the wild,
wild, west of the intellectual field and the hottest field on the
planet in terms of potential for new entries. [...] The future in
number theory belongs to the kids." -- James S. Harris corrupts youth
Chip Eastham
David C. Ullrich wrote:
[snip -- my apologies for preserving only a vague hint of context]
> More important: I never worried about the "derivation". Because
> even if you _did_ give a proper "derivation", that's still no
> reason whatever to think that the solution to the "pde" has
> anything at all to do with pi.
>
> There's no problem "deriving"
>
> f' = f, f(0) = 1
>
> from the recurrence
>
> a[n+1] - a[n] = a[n], a[0] = 1.
>
> But even though _that_ "derivation" is correct, the solution
> of the de still has nothing to do with the solution to the
> recurrence.
A broad brush account of what, conversely, the solution
of the discrete recurrence relation has to do with the
continuous solution of the differential equation.
The mantra for finite difference schemers is that "stability
plus consistency equals convergence". "Derivation" as
discussed here is loosely the consistency part of this
analysis.
Noting that the ODE has solution f(x) = e^x while the
recurrence relation has a[n] = 2^n, one might well ask
if we'd get a better approximation to e^x (convergence)
if we "refine the discretization", e.g. take differences at
intervals h --> 0 rather than simply at the integers. We
subscript by h to denote the uniform length of intervals:
a_h[0] = 1
a_h[n+1] = a_h[n] + h * a_h[n]
This simple forward difference scheme is consistent,
meaning that if the ODE solution e^x is plugged into
the difference scheme, it is "satisfied" up to suitably
small error of discretization:
A_h[n] = e^(n*h)
A_h[n+1] - (1+h) A_h[n]
= e^((n+1)h) - (1+h) e^(nh)
= e^(nh) * [ e^h - (1 + h) ]
= A_h[n] * O(h^2)
The above is called the "local" error of discretization,
and is to be contrasted with the "global" error (which
if small, would imply convergence).
On a finite interval [0,x] we do get convergence of the
above scheme. The explicit solution:
a_h[n] = (1+h)^n,
where h = x/N divides [0,x] into N equal subintervals,
increases monotonically to a_h[N] = (1+h)^N, which
is well known to converge to e^x:
limit (1+h)^(x/h) = e^x
h-->0
In rough terms it happens, as we add more intervals
for smaller h, that the local (relative) error of O(h^2)
accumulates N = x/h times to give a global (absolute)
error of O(h), where "stability", meaning boundedness
of the discrete solution, allows us to translate the
relative error into an absolute error.
regards, chip
marcus_b
I see that part of what I said here was not justified. I had
printed the blog text, but my printer did not show 'deltas' -
for example, there was no delta in front of S(x, y) and no
delta in the "y - y" expressions.
But the main point, that the chain rule was used incorrecty,
still stands.
Marcus.
jst...@msn.com
> ...
>
> [jst...@msn.com]
> >>> Ok, I'll bite, then what do you claim is the correct equation from
> >>> the proper derivation?
>
> [Tim Peters]
> >> Nope, first /you/ make an effort to understand what the chain rule
> >> is, and why it might be that people say you've neglected it. If
> >> you haven't noticed yet, I have: your "challenges" are easy to meet,
> >> but solutions go whizzing over your head. Especially in this case,
> >> as I said a few msgs back:
>
> [jst...@msn.com]
> > Then meet a single challenge.
>
> No. It's clear to me that you have no idea how to proceed here, and you
> ignored my much simpler challenge to compute the partial derivative w.r.t. y
> of f(x, y) = g(x, y^2). Proginoskes suggests that's probably too hard for
> you, and I expect he's right, but if you can't do that correctly it's
> impossible that you could even begin to follow a correct derivation starting
> from your much messier formula.
>
Let
g(x,y) = x+y
to keep things simple as then
g(x,y^2) = x + y^2 = f(x,y)
now take the partial derivative of g(x,y^2) with respect to y and do
the same for f(x,y).
Have you actually taken calculus at a university or college?
Readers should remember that being able to post on Usenet does not mean
a person actually has expertise in a particular area or even knows
ANYTHING significant in that area.
James Harris
gjed...@gmail.com
Classic behaviour by someone suffering from your illness. This annals
nonsense is a trick played by the delusional part of your mind on the
sane part. It's saying "hold on in there james, don't do anything
radical like getting help, at least wait to see if annals publish us,
go on james, give me us more chance, we are a genius, I promise". Time
for your rational side to fight back. Think man, think. This trick is
so transparent - how could annals publish you if, as your bad side
claims, mathematicians are protecting reimann, etc.? The paper
reviewers will all be extremely accomplished mathematicians - defending
tenure, etc.Don't put up with this nonsense, and get some help.
José Carlos Santos
> Readers should remember that being able to post on Usenet does not mean
> a person actually has expertise in a particular area or even knows
> ANYTHING significant in that area.
Anyone who reads you posts is fully aware of that.
Best regards,
Jose Carlos Santos
jst...@msn.com
Except I have a degree in in physics and as a kid was doing calculus
when I was twelve years old.
I learned calculus before I learned trigonometry so I had to backtrack
a bit.
I was playing with partial differentials in high school, and for a
while was fixated on the calculus of variations though I will admit I
never got really good at it.
It was when I went to college that I shifted more from the tools of
mathematics to what those tools were used on, and only after college
that I fond myself playing with number theory, which to my surprise
with my prime counting research, lead me back to the calculus.
James Harris
xun...@gmail.com
You say that you were doing calculus when you were 12, but where is
your proof. It is more important to demonstrate ability than to talk
about it. You have mentioned time and time again how you were "gifted"
when you were young. That may be true, but somewhere along the line you
have lost it. You have proven yourself inept at mathematics time and
time again and that trumps your repeated statements to the contrary.
William Hughes
jst...@msn.com wrote:
> Tim Peters wrote:
> > ...
> >
> > [jst...@msn.com]
> > >>> Ok, I'll bite, then what do you claim is the correct equation from
> > >>> the proper derivation?
> >
> > [Tim Peters]
> > >> Nope, first /you/ make an effort to understand what the chain rule
> > >> is, and why it might be that people say you've neglected it. If
> > >> you haven't noticed yet, I have: your "challenges" are easy to meet,
> > >> but solutions go whizzing over your head. Especially in this case,
> > >> as I said a few msgs back:
> >
> > [jst...@msn.com]
> > > Then meet a single challenge.
> >
> > No. It's clear to me that you have no idea how to proceed here, and you
> > ignored my much simpler challenge to compute the partial derivative w.r.t. y
> > of f(x, y) = g(x, y^2). Proginoskes suggests that's probably too hard for
> > you, and I expect he's right, but if you can't do that correctly it's
> > impossible that you could even begin to follow a correct derivation starting
> > from your much messier formula.
> >
>
> Let
>
> g(x,y) = x+y
>
> to keep things simple as then
Nope, this is pretty simple as it stands. You do
not get to make it simpler.
Take f(x, y) = g(x, y^2).
let h_1(x,y) be the partial derivative with respect to x of h(x,y)
Let h_2(x,y) be the partial derivative with respect to y of h(x,y)
Write f_2(x,y) in terms of y and g_2(x,y).
(Hint, let r(y) = s(y^2). then r'(y) = 2 y s'(y^2))
-William Hughes
jst...@msn.com
You have an equality.
Necessarily f'_y(x,y) = g'_y(x,y^2) as that's what "equal" means.
If f(x,y) = g(x,y^2) then they are EQUAL.
Also you deleted out a great example to show what is happening:
g(x,y) = x + y
then
g(x,y^2) = x + y^2
and if you let f(x,y) = x + y^2, notice that the partial derivative
with respect to y, of f(x,y) equals that of g(x,y^2) because they are
EQUAL.
James Harris
Tim Peters
[marcus_b]
>>> David has shown that solutions to the differential equation
>>> associated with a given recursive function need not be the original
>>> function.
[Tim Peters]
>> Well, if the derivative of a function "makes sense", the function
>> has to satisfy whatever equation(s) you derive involving the
>> derivative. In the:
>>
>> a(0) = 1
>> a(n+1) = 2*a(n)
>>
>> example, what /justifies/ going from the derived:
>>
>> a(n+1) - a(n) = a(n)
>>
>> to:
>>
>> a'(n) = a(n)
>>
>> to begin with?
>>
>> ... [nothing, but if more care is taken then "the mysteries"
>> disappear] ...
[David C. Ullrich]
> Of course nothing justifies that - that's been my point.
Right, but we're so far removed from your original here that I thought it
helpful to spell out what the "associated with" (from "the differential
equation associated with a given recursive function") actually means in this
context: sloppily / wishfully / by-naive-eyeball-without-care "associated
with".
> This is nonetheless the sort of thing that was done in
> "deriving" the famous Harris so-called "pde". Or that's
> my impression - _did_ he in fact do something more
> sensible, as below? [Turns out that's a rhetorical
> question, answered below.]
And it's worse than /just/ this.
>> "Divide both sides by 1 then pretend 1 approaches 0 on the
>> left but not the right" isn't terribly satisfying ;-) Flesh it
>> out and "the problem" goes away; e.g.,
>>
>> ...
>> I think an obvious problem with James's formula w.r.t. similar
>> fiddling is that it's discontinuous: to get anywhere sane, he
>> needs to "flesh it out" too, at least so that derivatives /exist/.
>> It's certainly true that, as David's example illustrated, wishing
>> that away and then merely /hoping/ that a carelessly derived PDE
>> will magically fill in the gaps is unlikely to work worth beans.
> I guess he didn't, then. Didn't think so.
The "derivation" is so messed up it's hard to fathom. The /only/ use of y
in his equation for P(x, y) is to bound a summation over integers, a sum
from k = 2 to y. In addition, P(x, y) yields only integer values. So P(x,
y) is constant across y in [i, i+1) for all natural i; and P(x, i+1) is
either the same integer as P(x, i), or leaps to some other integer.
So, more accurately, the derivative of P w.r.t. y is, in reality, 0 almost
everywhere, while at the only interesting points (integers) it has a right
derivative of 0 but whether the left derivative /exists/ depends on the
specific value of y.
This gets most extreme in the subexpression:
deltaP(k, k) = P(k, sqrt(k)) - P(k-1, sqrt(k-1))
which is 1 when k is prime and 0 when k is composite. WTF is "the
derivative" of that wrt k? Well, depending on k, and on whether you're
looking at the left or right derivative, 0, +inf, or -inf.
Mostly in this particular thread marcus_b & I have been harping on an
entirely different class of "WTF?!"s: ignore all that, and James also
missed the necessity of applying the chain rule to function compositions,
and even seemed to miss the multiplication rule (seemingly treating (f*g)'
as f*g', forgetting the f'*g half entirely).
Nevertheless, "numeric integration" works great ;-)
marcus_b
So let's take a look at how you compute derivatives in your
blog on this topic. You start with:
dS(x,y)/dy = (P(x/y,y-dy) - P(y-dy,sqrt(y-dy)))*
(P(y, sqrt(y)) - P(y-dy,sqrt(y-dy)))/dy.
Now you want to let dy -->0.
On the left, you get S'_y(x, y)
OK, now for the right side. Let's look in particular at
lim (P(y, sqrt(y)) - P(y-dy,sqrt(y-dy)))/dy.
dy->0
This is where the chain rule comes in. After some
manipulations, you SHOULD end up with
dP_1(y, sqrt(y)) + dP_2(y, sqrt(y))*.5/sqrt(y),
where dP_1 and dP_2 are the partial derivatives of the
function P with respect to its first and second
arguments respectively. This is really very far from
what you wrote down, namely,
P'(y, sqrt(y)),
which is not even well-defined.
This is basic differential calculus. You say you
learned this at age 12 and that you "played with"
partial differentials in high school.
That Duke summer camp - pretty much a joke, eh?
Marcus.
sg...@hotmail.co.uk
> > [Tim Peters]
> > No. It's clear to me that you have no idea how to proceed here, and you
> > ignored my much simpler challenge to compute the partial derivative w.r.t. y
> > of f(x, y) = g(x, y^2). Proginoskes suggests that's probably too hard for
> > you, and I expect he's right, but if you can't do that correctly it's
> > impossible that you could even begin to follow a correct derivation starting
> > from your much messier formula.
>
> Let
>
> g(x,y) = x+y
>
> to keep things simple as then
>
> g(x,y^2) = x + y^2 = f(x,y)
>
> now take the partial derivative of g(x,y^2) with respect to y and do
> the same for f(x,y).
>
> Have you actually taken calculus at a university or college?
Wow, I see what you did there: rather than actually answer Tim's
challenge, thereby demonstrating an ability (or otherwise) to do
school-level calculus problems, you have simply replaced his problem
with a simpler problem, and then cast aspersions on /his/ ability to do
that problem. Brilliant. You really turned the tables on him. I bet
nobody noticed that you haven't actually shown that you can answer the
question.
> Readers should remember that being able to post on Usenet does not mean
> a person actually has expertise in a particular area or even knows
> ANYTHING significant in that area.
You remind readers of this every time you post.
-Rotwang
William Hughes
No, if g(x,y^2) = x + y^2, then g_2(x,y) is not equal to the partial
derivative of (x+y^2) with respect to y, but the partial derivative of
(x+y^2) with respect to y^2.
Let f(x,y) = x + y^2. Then f_2(x,y) is the partial derivative of
(x+y^2) with resprect to y, 2y.
Let g(x,y) = x + y^2. Then g_2(x,y^2) is the partial derivative
of (x + y^2) wih respect to y^2, 1.
2y is not equal to 1.
- William Hughes
>
>
> James Harris
Tim Peters
[Tim Peters]
>>>> No. It's clear to me that you have no idea how to proceed here,
>>>> and you ignored my much simpler challenge to compute the partial
>>>> derivative w.r.t. y of f(x, y) = g(x, y^2). Proginoskes suggests
>>>> that's probably too hard for you, and I expect he's right, but if
>>>> you can't do that correctly it's impossible that you could even
>>>> begin to follow a correct derivation starting from your much
>>>> messier formula.
>>> Let
>>>
>>> g(x,y) = x+y
>>>
>>> to keep things simple as then
[William Hughes]
>> Nope, this is pretty simple as it stands. You do
>> not get to make it simpler.
>>
>> Take f(x, y) = g(x, y^2).
>>
>> let h_1(x,y) be the partial derivative with respect to x of h(x,y)
>> Let h_2(x,y) be the partial derivative with respect to y of h(x,y)
>>
>> Write f_2(x,y) in terms of y and g_2(x,y).
>>
>> (Hint, let r(y) = s(y^2). then r'(y) = 2 y- s'(y^2))
You meant "*" or a blank (implied "*") instead of "-", right?
[jst...@msn.com]
> You have an equality.
Stop, James. It's clear that you don't even understand what the /question/
here is. William does, so listen to him (I'm not bothering with you on this
anymore). A correct answer would indeed, in his terminology, express f_2(x,
y) as a function of y and g_2(x, y).
> Necessarily f'_y(x,y) = g'_y(x,y^2) as that's what "equal" means.
>
> If f(x,y) = g(x,y^2) then they are EQUAL.
Trivially true, yes, but mostly irrelevant.
> Also you deleted out a great example to show what is happening:
Irrelevant to the /answer/. If you need to "plug in" some specific function
for g() to help you arrive at the answer, that's fine, but the correct
answer makes /no assumption/ about g beyond that it has a derivative w.r.t.
y (*). A correct answer "works" regardless of whether g(x, y) = x+y or g(x,
y) = x^(4*y/pi) * sin(sqrt(x^2+y^2)), etc.
(*) Which is a larger point about your much messier "derivation":
the derivative of P(x, y) w.r.t. y is mostly 0, but at the
"interesting" values of y has no left derivative. That's what
makes this is a fool's errand from the start: even if you did
know how to compute messy partial derivatives symbolically, your
specific P(x, y) function doesn't /have/ the necessary
derivatives w.r.t. y for this exercise to make /sense/.
> ...
jst...@msn.com
Clearly you're getting lost in hypotheticals so let's make it simpler
for you:
Give the partial derivatives of g(x,y^2) and f(x,y) with respect to y,
when
g(x,y^2) = x + y^2, and f(x,y) = g(x,y^2).
James Harris
William Hughes
Note that the partial derivative of g(x,y^2) with respect to y
is not g_2(x,y).
The partial derivative with respect to y of f(x,y) = x + y^2
is f_2(x,y) = 2y
The partial derivative with respect to y of g(x,y^2) is
2y g_2(x,y) = 2y
So g_2(x,y) = 1 which is not equal to f_2(x,y) = 2y.
- William Hughes
marcus_b
There is a disconnect here between you and William Hughes,
and oddly, you are both right. That's because you are talking about
different things.
The partial derivative of g(x, y) = x + y, with respect to its second
argument, is 1. That could be written as
dg(x, y)/dy = 1.
Now if f(x, y) = g(x, y^2), the partial derivative of f with respect
to its second argument is (by the chain rule)
df(x, y)/dy = (dg(x, y/dy)*(dh/dy), where h(y) = y^2. Therefore
df(x, y)/dy = 2y. You and Hughes get the same answer.
A better example might be the following. Assume
f(x, y) = g(r(x,y), s(x,y)),
and assume you know all of the following:
1. dg_1 and dg_2, the partial derivatives of g with respect
to its first and second arguments, respectively
2. The partial derivatives dr/dx, dr/dy, ds/dx, ds/dy.
Now: what is the partial dervative of f(x, y) with respect to y,
in terms of all of the above?
Do you know?
Evidently not, since you got it wrong in your blog.
Marcus.
>
>
> James Harris
William Hughes
Tim Peters wrote:
> ...
>
> [Tim Peters]
> >>>> No. It's clear to me that you have no idea how to proceed here,
> >>>> and you ignored my much simpler challenge to compute the partial
> >>>> derivative w.r.t. y of f(x, y) = g(x, y^2). Proginoskes suggests
> >>>> that's probably too hard for you, and I expect he's right, but if
> >>>> you can't do that correctly it's impossible that you could even
> >>>> begin to follow a correct derivation starting from your much
> >>>> messier formula.
>
> [jst...@msn.com]
> >>> Let
> >>>
> >>> g(x,y) = x+y
> >>>
> >>> to keep things simple as then
>
> [William Hughes]
> >> Nope, this is pretty simple as it stands. You do
> >> not get to make it simpler.
> >>
> >> Take f(x, y) = g(x, y^2).
> >>
> >> let h_1(x,y) be the partial derivative with respect to x of h(x,y)
> >> Let h_2(x,y) be the partial derivative with respect to y of h(x,y)
> >>
> >> Write f_2(x,y) in terms of y and g_2(x,y).
> >>
> >> (Hint, let r(y) = s(y^2). then r'(y) = 2 y- s'(y^2))
>
> You meant "*" or a blank (implied "*") instead of "-", right?
I could claim that I am using an out of date browser
and the google editor, a combination that has a couple of quirks
so that updates are not always done, and sometimes
I see a space where there is actually a character.
But this would be a stupid exuse. The real reason
I put that mistake in is to see if anyone was paying
attention.
- William (There is three deliberate mistales in this
sentence. ) Hughes
willo_...@hotmail.com
jst...@msn.com wrote:
>
> Necessarily f'_y(x,y) = g'_y(x,y^2) as that's what "equal" means.
>
> If f(x,y) = g(x,y^2) then they are EQUAL.
>
Oh, EXCELLENT! No matter how stupid you think James might
be, he always manages to say something even stupider.
Michael Press
<p4uem2p8n59aundv2...@4ax.com>,
Patrick Hamlyn <path@multipro.N_OcomSP_AM.au> wrote:
> "Tim Peters" <tim...@comcast.net> wrote:
>
> >[jst...@msn.com]
> >>> I wrote another paper and sent it to the Annals of Mathematics which
> >>> verified receipt.
> >
> >[fishfry]
> >> Only one 'n' needed.
> >
> >Possibly, but ragging on James for using Princeton's spelling is of dubious
> >value when there's so much of substance you /could/ object to instead ;-)
> >
> > http://annals.princeton.edu/
>
> Possibly you are operating under the humour-impaired model commonly used by
> denizens of the US and A.
>
> 'Fishfry' is not seriously challenging the spelling of annals, he is implying
> that a different spelling would have been a more appropriate address for James'
> paper.
>
> In other countries we call this humour and appreciate it 'in and of itself'.
Spare us the jingoist cant. Do you expect everyone who
got the joke to chime in? Then how could we "appreciate
it 'in and of itself'."
--
Michael Press
Davie
<jst...@msn.com> wrote in message
news:1164565411....@j72g2000cwa.googlegroups.com...
José Carlos Santos wrote:
> jst...@msn.com wrote:
>
> > Readers should remember that being able to post on Usenet does not mean
> > a person actually has expertise in a particular area or even knows
> > ANYTHING significant in that area.
>
> Anyone who reads you posts is fully aware of that.
>
> Best regards,
>
> Jose Carlos Santos
Except I have a degree in in physics
NO YOU DO NOT, LIER.
and as a kid was doing calculus
when I was twelve years old.
YOU FORGOT IT ALL.
I learned calculus before I learned trigonometry so I had to backtrack
a bit.
YOU FORGOT IT ALL
I was playing with partial differentials in high school, and for a
while was fixated on the calculus of variations though I will admit I
never got really good at it.
YOU ARE NOT GOOD AT MATH.
It was when I went to college that I shifted more from the tools of
mathematics to what those tools were used on, and only after college
that I fond myself playing with number theory, which to my surprise
with my prime counting research, lead me back to the calculus.
BUT YOU NEVER GOT A DEGREE IN PHYSICS. IF SO, POST SOMETHING IN PHYSICS
James Harris
Patrick Hamlyn
>Spare us the jingoist cant. Do you expect everyone who
>got the joke to chime in? Then how could we "appreciate
>it 'in and of itself'."
Yes I can see why you might have trouble appreciating humour under any but the
most nurturing conditions.
Now just why would I expect anyone to chime in? You're not making sense. In fact
I didn't expect anyone to chime in, whereas in fact you did, completely
pointlessly as it turns out.
--
Patrick Hamlyn posting from Perth, Western Australia
Windsurfing capital of the Southern Hemisphere
Moderator: polyforms group (polyforms...@egroups.com)
jmor...@idirect.com
Proginoskes wrote:
> Tim Peters wrote:
> > [jst...@msn.com]
> > >>> Read the derivation.
> >
> > [Tim Peters]
> > >> I suppose you mean this?
> > >>
> > >> <http://mymath.blogspot.com/2005/06/partial-differential-prime-counting.html>
> >
> > >>> I don't give it in the paper as it's trivial enough for expert
> > >>> mathematicians, but it is on my math blog.
> >
> > >> rule. As I elaborated, you neglected to apply it in several places.
> >
> > [jst...@msn.com]
> > > Ok, I'll bite, then what do you claim is the correct equation from
> > > the proper derivation?
> >
> > why it might be that people say you've neglected it. If you haven't noticed
> > yet, I have: your "challenges" are easy to meet, but solutions go whizzing
> > over your head. Especially in this case, as I said a few msgs back:
> >
> > deltaP(y, y) was torturously contrived to be an indicator
> > function (0 if y composite, 1 if y prime), so that the notion
> > he's going to get a lick of sense out of this by presuming
> > to differentiate it is exceedingly dubious on the face of it.
> >
> >
> > >> Since, as marcus_b also said, that's at the level of freshman
> > >> calculus, explaining that to you is likely hopeless.
> >
> > > Why explain? Give what you think is the correct derivation.
> >
> > You gave your game away with that, James. If you were interested in truth
> > or understanding, the /explanation/ would have been everything to you.
> >
> > >> Instead I'll just point out that your blog at least remembered to
> > >> include the necessary floor function, the lack of which in your
> > >> paper renders the latter wrong starting with its first equation.
> >
> > > With x and y declared to be natural numbers that's the ring. All is
> > > correct within the ring of natural numbers where floor() is redundant.
> >
> > In your fantasies, perhaps, but not in standard mathematics. You think
> > equation [1] at:
> >
> > http://mathworld.wolfram.com/LegendresFormula.html
> >
> > bothered slopping in all the "floor" symbols because the author enjoyed
> > tedious typesetting? This is mathematics, NOT A PROGRAMMING LANGUAGE with
> > implicit value-changing type conversions. If you want to lose fractional
> > parts in math, you need to do so /explicitly/. Else 17/2 is 8.5, period.
> >
> > >> I'd explain more, except my hands are shaking too badly
> > >> [from] fear ;-)
> >
> > > You're finally proven to be incompetent and a bad liar--
> >
> > Consistency, James: I've personally lost track of how often you've
> > announced you've proven that. Suddenly it's "finally"? LOL.
> >
> > > or, more brilliant than me.
> >
> > I certainly have a deeper and broader understanding of math than you have.
> > "Who's more brilliant?" is one of those irrelevant whose-dick-is-bigger
> > games you should have noticed that: (1) I never /start/; and, (2) you never
> > win.
> >
> > > If so, give the derivation I missed.
> >
> > Try to figure it out yourself. Here's a practice problem: suppose
> >
> > f(x, y) = g(x, y^2)
> >
> > What's the partial derivative of f w.r.t. y?
>
> Too difficult for him. Choose something simple like g(x, y) = 2 x + 3
> y, and have him calculate the partial derivative of f with respect to y
> by doing the full substitution, then comparing it with the partial
> derivative of g with respect to y.
>
> After all, he only has a B.S.[1] in physics. But any B.S. in physics
> would certainly require Calculus III, which is where this problem comes
> from. (I also teach Calc III, so I know what problems students should
> be able to do to pass that class.)
>
> --- Christopher Heckman
>
> [1] JSH continually uses "B.Sc." in his posts. My guess is he's trying
> to avoid mention of the abbreviation "B.S." in his posts where he can;
> if he used "B.S." instead, that would subconsciously "taint" the rest
> of the post.
Could be worse... I spent four years of university tuition so my
soparano daughter could write B.M after her name.. :)
Proginoskes
Tim Peters wrote:
> [...]
That reminds me of the "logic" of a paranoid personality: Anyone who
tries to convince a paranoid individual that his delusions are wrong
becomes "part of the cover-up."
--- Christopher Heckman
Proginoskes
I don't recognize that as a proper definition of g, since y^2 can be
the square of two different real numbers. (You made a big fuss about
this a few months ago, remember?)
You actually got lucky here, since this difference does not show up on
the right-hand side of
g(x,y^2) = ...; if you had said g(x,y^2) = x + y, then we could have
had
1 = 0 + 1 = g(0,1^2) = g(0, 1) = g(0, (-1)^2) = 0 + (-1) = -1.
Now, once g is defined properly the confusion vanishes.
--- Christopher Heckman
David Moran
José Carlos Santos wrote:
> jst...@msn.com wrote:
>
> > Readers should remember that being able to post on Usenet does not mean
> > a person actually has expertise in a particular area or even knows
> > ANYTHING significant in that area.
>
> Anyone who reads you posts is fully aware of that.
>
> Best regards,
>
> Jose Carlos Santos
Except I have a degree in in physics and as a kid was doing calculus
when I was twelve years old.
So what?
I learned calculus before I learned trigonometry so I had to backtrack
a bit.
Sure....how do you learn calculus without knowing trig?
I was playing with partial differentials in high school, and for a
while was fixated on the calculus of variations though I will admit I
never got really good at it.
It was when I went to college that I shifted more from the tools of
mathematics to what those tools were used on, and only after college
that I fond myself playing with number theory, which to my surprise
with my prime counting research, lead me back to the calculus.
James Harris
Dave