Bob, I'm calling you out! - Page 2

amateur · 2004-10-25, 17:24

Quote:

Originally Posted by Bob Silverman

see above.
(10) "uniqueness". More unqualified hyperbolee.

You would be much better off if you READ ABOUT THIS SUBJECT.

Until you do, I simply can't be bothered looking at your work. It isn't worth
my time.

With all the respect, why are you writing books about subjects you are not interested in? :surprised

If I was an expert in ornithology, and someone told me he has seen a bird he doesn't know what it is, but it doesn't look like anything he has seen or read about so far, I would jump at the opportunity to see it. Even if it would turn out to be a sparrow fallen into red dye.

But I couldn't hold my curiosity back.

You make a big logical error. It is not necessary to have previous knowledge to have original ideas. If you were right and nobody could come up with anything useful without first studying it, we would be still sitting on the trees.

Thanks anyways.

amateur

synergy · 2004-10-25, 18:32

Bob said
(10) "uniqueness". More unqualified hyperbolee.
You would be much better off if you READ ABOUT THIS SUBJECT.
Until you do, I simply can't be bothered looking at your work. It isn't worth
my time.

[Answer]Now, I've admitted to the shortcomings in my own theory, and swallowed my pride and all, and my opinion may not be worth much here anymore, but I don't see why you judge somebody else's work as "not worth your time" without even viewing it. It's a test like any test program, not a generator like mine. I've programmed it on my calculator, and it is slow. But it also is unique (it uses a characteristic that all primes have {I can't say how, I'm sworn to secrecy}, and doesn't require knowledge of any other prime), and it
DOESN'T SEEM TO GET MUCH SLOWER AS THE PRIMES GROW IN SIZE,
which may (again, I'm saying maybe, as before, so don't get upset, I'm not making claims, I'm just saying maybe) make it faster than other methods as the primes get large. Probably not, but isn't every idea worth checking out? Remember, Einstein had to go and modify Relativity after experimental results only partly agreed with his predictions. Science (and math) sometimes moves forward in large quantum jumps, but more often it moves forward by a bunch of barely-useful ideas coming together in just the right way to make a giant leap forward. If you turn your back on every little idea that comes along, it makes it more difficult to find that "rare pebble on the beach of knowledge" as Einstein would say.

Bob said:

Synergy suggested, in a sci.math post that (2^x 3^y - 5^w 7^z) was
prime if it was less than 121 because it is not divisible by 2,3,5,7.

[Answer]In just a minute (read down about 3 paragraphs) you agree.

I asked synergy to demonstrate how to determine x,y,w,z so that
N = (2^x 3^y - 5^w 7^z) for a given N < 121. This request was ignored.
I asked synergy if he could demonstrate that every prime p < 121 was in
fact in the range of this exponential expression. This too was ignored.

[answer] I don't remember this request, specifically, but the help I was asking for didn't require the details be presented for every prime. I'll reproduce my results for p<121 in a later post.

Synergy proposed that if p = A - B or A + B, where A and B were taken
as the product of primes less than some bound, and if p were bounded, then
p was prime. [Here's where you agreed]I agreed with this. But I pointed out that this was handwaving,
not an algorithm, because it failed to specify a method for constructing A
and B. Indeed, synergy even failed to show that every p HAD such a
representation.

[answer] I always stated that every disjoint partition (subset, as you say below, but disjoint partition is more specific, it means everything from the set is in either A or B, but nothing from the set is in BOTH, hence disjoint) that every disjoint partition be used, otherwise you can't find ALL of the primes. I always said use all of the set of primes from 2 to p(n) in this way.

Now, in later posts, synergy suggests finding A & B by searching all the
possible subsets. I suspected this is what he had in mind originally.
If synergy had bothered to do even the minimal amount of background
research, he would have discovered what I told him in the first place:
It requires too much arithmetic. Yet, synergy continued to cling to
his idea that because his "method" did not involve checking composites,
that it would be faster than existing methods. I also suggested that synergy
look at the SIZE of A and B and estimate the amount of arithmetic needed
to calculate them. This too was ignored.

[Answer] No, I didn't cling to my idea despite what you said, I clung to it because I actually was too close to the problem to see the big picture, and I was asking for someone, anyone who could see the big picture to EXPLAIN to me why it would or would not work. Respectfully, I DIDN'T KNOW if you were an accomplished mathematician or if you yourself were a bit of a crackpot who liked shooting down other peoples ideas, I ONLY KNEW that you didn't like my idea for some reason, and I didn't know if it was a good reason or not, you didn't care to tell me any details of your reasoning. I was asking for an explanation, not "go look it up" because, frankly, I am taking more time on this than I should already, and maybe I just wanted an answer now rather than wait until I could take more time with it. Sorry for being over enthusiastic. I searched for this for about 12 years, and after finding it I've had it bouncing around in my head for about 9 more years, I guess I should have learned patience by now.

Let S(n) = {p < n | p prime} and let S = S1 + S2. To prove p prime, we
must find A = product(p in S1) and B = product(p in S2) such that p = A+B,
or p = A-B and p < square of the largest element of S.

[Answer]Not to put you down or anything, but I like the way I said it better, I thought it was a bit more clear. Sorry you didn't like it.

This does indeed give a method for proving p prime. However, we have not
proved that every prime p has such a representation. Furthermore there
are 2^n such subsets to be examined. And the arithmetic to compute A and
B is extensive. We have #S(n) ~ n/log n by the PNT. max(A, B)
is approximately exp(n). Thus, the arithmetic to find a represention in
the desired form is O(2^(n/log n) exp(n)). This is DOUBLY exponential in the
size of the problem.

[Answer] I said in earlier posts that I hadn't been able to prove that every prime has such a representation, and in fact, That is what I would most like help in doing. The efficiency thing was only because I began thinking it might be efficient, but even if it isn't, I would still like help proving this. I recognize the problem is huge, that is why I asked you guys for input on how fast it would be, I don't yet know how to compute such things. Computer efficiency is not my main thing. Also, I have only 2 semesters to go, I have a B average, and I must maintain that B average to stay in the Master's program, and I have point-set topology with a professor who gives a Massive amount of homework, so I don't have as much time as I would like to give to this problem.

I hope this helps you see why my focus has been so narrow, I just want to round out my understanding of this monster, bring it to a close, and go on with my life. It isn't efficient? - Okay. Does every prime have such a representation? - nobody seems to have any clear idea of how to prove it does or doesn't, least of all myself. When I come up with an answer, I'll let you know.
Aaron

synergy · 2004-10-25, 20:39

By the way, after this and the above post referring to certain "inconsistencies in communication on the part of both parties" to try to phrase things in a non-personal fashion, after these two posts I will refrain from posting on this thread so that it will drop out of sight. When I get the representation of primes < 121 done, I'll start a new thread for it. I didn't mean for the "calling you out" title to continue being perpetuated. Sorry.
Aaron

Xyzzy · 2004-10-25, 21:17

If Bob has to deal with even 1% of the stuff I get by email from people reading this forum, I can see how he might get annoyed...

I get weird emails all the time from people claiming to have solved all sorts of math related stuff... I'm not sure why people would email me, because I don't have a clue at all...

Usually I just reply and tell them to post in the forum... To date, I can't remember any of them actually doing this...

All these letters are similar... After a while they start to sound like Nigerian scam emails... I can almost predict what they will say...

I'm not saying your idea is wacko, but I can certainly see how someone who knows what they are doing might view anything like this...

I know very little, but the few things I do know I know forwards and backwards... It would seems strange to me if someone with no training or experience came up to me and told me they had a breakthrough, even though I had spent years researching and working in said field...

I'm not saying it couldn't happen... But I doubt it would happen a lot...

IMO, the best thing to do is lay everything out on the table... No secrets... Just post... If it is legit you will find out soon... If not, you will find that out too... If someone gave me pointers, even if they were grouchy, I'd go look them up...

The older I get the more I realize how little I know... Maybe that attitude is holding me back, but that's how I feel...

amateur · 2004-10-25, 21:38

[QUOTE=Xyzzy]
IMO, the best thing to do is lay everything out on the table... No secrets... Just post... If it is legit you will find out soon... If not, you will find that out too... If someone gave me pointers, even if they were grouchy, I'd go look them up...
[QUOTE]

Thanks, Xyzzy. The reasons for secrecy are twofold.

First, if what I have here is old news or a nonsense, I would like as few people as possible to know about it. I guess I don't have to explain why...
People just don't like to be embarrassed in front of thousand others.

Secondly, if the idea is worthwhile, I would like to work it out as well as I can; maybe with the help of an expert. You know, someone who works with this might see something in it that I don't know. With other words, it could be the solution to something I have never thought of.

The reason why I would like to talk to an expert is the same why we see a doctor when we have a sore throat instead of showing it to the whole world. Only the doctor can tell me if what I have is just a cold or a throat cancer.....

As a matter of fact I have expected this reaction. It is not the first time I try to ask an expert and all I get is haughtiness and lecturing.
Where are the people who are truly enthusiastic about mathematics? The ones who can hardly wait to see some new ideas? Has that kind died out?

Or it never existed, only in books and movies?

Mystwalker · 2004-10-26, 07:31

Quote:

Where are the people who are truly enthusiastic about mathematics? The ones who can hardly wait to see some new ideas?

I guess Bob is literally overwhelmed with interesting approaches, it's just the time that's missing to follow all trails. I know this feeling, it's like

...
The ugly thing is: In this situation, you have to priorize - What looks most promising relative to the effort it takes to accomblish? (Although, if the effort is big, but a great result likely, why not?)

If you switch do his view, it's rather unlikely (though nevertheless possible) that this situation is going to end in a big breakthrough. The thing you (as Bob) want is to quickly see if the approach has potential (that's why all scientific publications have an "abstract", which shortly describes what it is all about - a reader can decide after 1 minute if the paper is worth (for him/her!) to be read).
That's why he asked you to make sure that this technique is sound and hasn't been already thought of (and dismissed) and to present it in a way that takes Bob minimal time to comprehend. All of this requires to get used to the mathematical basics. Although I see that your writing improved somewhat (possibly beyond my knowledge, but I'm an ongoing software engineer, no blooded mathematician

), it's still way below scientific publication quality.
IMHO, it would be a complaisant gesture to just get familiar with the books/publications Bob suggested you. After all, it optimizes one of your requirements:

Quote:

First, if what I have here is old news or a nonsense, I would like as few people as possible to know about it.

With the help of the provided material, I think it's well possible to find that out for yourself, thus no one gets to know it. You can just

and move on, being the only one knowing of it. Besides, I don't think that a stupid mistake in your thoughts should embarass you - everyone has made those, attendees included.

I'm not sure about "books and movies", but I think to remember in some of them, you see (resp. read) the chief character sitting in a library studying a pile of books that nearly touches the ceiling. Afterwards, he talkes with some friends/colleagues and after that gets in contact with the head scientist / president.
And I think that's what you really should do. Study the books (those also give you an idea how to present your thoughts in a way that a mathematician can quickly comprehend them), talk with some colleagues / people which are willing to debate woth you about this subject (the better your idea is worked out, presented and sounds promising, the more people will be interested in it) and then get in touch with the head scientist / president (a.k.a. Bob Silverman - he may decide which one he prefers

).

If your idea is really brilliant, this is the way to do it right. I've learned that getting accepted by a scientific group means to comply with certain rules they have set. For newcomers, it's sometimes

, but that's the way it is. When you are accepted, you are free to bend the rules here and there...

And even if your idea doesn't directly lead to a breakthrough, it could bring up other ideas. And anyway, you have learned a lot, and IMHO that's reason enough to undertake this journey.

amateur · 2004-10-26, 15:26

Thanks, Mystwalker,

I have written a long reply, but I erased it all.
This just doesn't make sense. This is not what I wanted.
I wanted help and I am caught in a debate with no end.

It is just not worth MY TIME.

Thanks anyways everybody.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Calling airsquirrels	Prime95	GPU Computing	16	2015-09-29 18:06
Calling all 64-bit Linux sievers!	frmky	NFS@Home	25	2013-10-16 15:58
Calling a spade a "spade"	davieddy	Soap Box	5	2009-01-23 05:17
Calling hotshot coders: NFS code challenge	R.D. Silverman	Programming	63	2006-10-09 05:48

2004-10-25, 18:32	#13
synergy Aug 2004 42₁₀ Posts	Bob said (10) "uniqueness". More unqualified hyperbolee. You would be much better off if you READ ABOUT THIS SUBJECT. Until you do, I simply can't be bothered looking at your work. It isn't worth my time. [Answer]Now, I've admitted to the shortcomings in my own theory, and swallowed my pride and all, and my opinion may not be worth much here anymore, but I don't see why you judge somebody else's work as "not worth your time" without even viewing it. It's a test like any test program, not a generator like mine. I've programmed it on my calculator, and it is slow. But it also is unique (it uses a characteristic that all primes have {I can't say how, I'm sworn to secrecy}, and doesn't require knowledge of any other prime), and it DOESN'T SEEM TO GET MUCH SLOWER AS THE PRIMES GROW IN SIZE, which may (again, I'm saying maybe, as before, so don't get upset, I'm not making claims, I'm just saying maybe) make it faster than other methods as the primes get large. Probably not, but isn't every idea worth checking out? Remember, Einstein had to go and modify Relativity after experimental results only partly agreed with his predictions. Science (and math) sometimes moves forward in large quantum jumps, but more often it moves forward by a bunch of barely-useful ideas coming together in just the right way to make a giant leap forward. If you turn your back on every little idea that comes along, it makes it more difficult to find that "rare pebble on the beach of knowledge" as Einstein would say. Bob said: Synergy suggested, in a sci.math post that (2^x 3^y - 5^w 7^z) was prime if it was less than 121 because it is not divisible by 2,3,5,7. [Answer]In just a minute (read down about 3 paragraphs) you agree. I asked synergy to demonstrate how to determine x,y,w,z so that N = (2^x 3^y - 5^w 7^z) for a given N < 121. This request was ignored. I asked synergy if he could demonstrate that every prime p < 121 was in fact in the range of this exponential expression. This too was ignored. [answer] I don't remember this request, specifically, but the help I was asking for didn't require the details be presented for every prime. I'll reproduce my results for p<121 in a later post. Synergy proposed that if p = A - B or A + B, where A and B were taken as the product of primes less than some bound, and if p were bounded, then p was prime. [Here's where you agreed]I agreed with this. But I pointed out that this was handwaving, not an algorithm, because it failed to specify a method for constructing A and B. Indeed, synergy even failed to show that every p HAD such a representation. [answer] I always stated that every disjoint partition (subset, as you say below, but disjoint partition is more specific, it means everything from the set is in either A or B, but nothing from the set is in BOTH, hence disjoint) that every disjoint partition be used, otherwise you can't find ALL of the primes. I always said use all of the set of primes from 2 to p(n) in this way. Now, in later posts, synergy suggests finding A & B by searching all the possible subsets. I suspected this is what he had in mind originally. If synergy had bothered to do even the minimal amount of background research, he would have discovered what I told him in the first place: It requires too much arithmetic. Yet, synergy continued to cling to his idea that because his "method" did not involve checking composites, that it would be faster than existing methods. I also suggested that synergy look at the SIZE of A and B and estimate the amount of arithmetic needed to calculate them. This too was ignored. [Answer] No, I didn't cling to my idea despite what you said, I clung to it because I actually was too close to the problem to see the big picture, and I was asking for someone, anyone who could see the big picture to EXPLAIN to me why it would or would not work. Respectfully, I DIDN'T KNOW if you were an accomplished mathematician or if you yourself were a bit of a crackpot who liked shooting down other peoples ideas, I ONLY KNEW that you didn't like my idea for some reason, and I didn't know if it was a good reason or not, you didn't care to tell me any details of your reasoning. I was asking for an explanation, not "go look it up" because, frankly, I am taking more time on this than I should already, and maybe I just wanted an answer now rather than wait until I could take more time with it. Sorry for being over enthusiastic. I searched for this for about 12 years, and after finding it I've had it bouncing around in my head for about 9 more years, I guess I should have learned patience by now. Let S(n) = {p < n \| p prime} and let S = S1 + S2. To prove p prime, we must find A = product(p in S1) and B = product(p in S2) such that p = A+B, or p = A-B and p < square of the largest element of S. [Answer]Not to put you down or anything, but I like the way I said it better, I thought it was a bit more clear. Sorry you didn't like it. This does indeed give a method for proving p prime. However, we have not proved that every prime p has such a representation. Furthermore there are 2^n such subsets to be examined. And the arithmetic to compute A and B is extensive. We have #S(n) ~ n/log n by the PNT. max(A, B) is approximately exp(n). Thus, the arithmetic to find a represention in the desired form is O(2^(n/log n) exp(n)). This is DOUBLY exponential in the size of the problem. [Answer] I said in earlier posts that I hadn't been able to prove that every prime has such a representation, and in fact, That is what I would most like help in doing. The efficiency thing was only because I began thinking it might be efficient, but even if it isn't, I would still like help proving this. I recognize the problem is huge, that is why I asked you guys for input on how fast it would be, I don't yet know how to compute such things. Computer efficiency is not my main thing. Also, I have only 2 semesters to go, I have a B average, and I must maintain that B average to stay in the Master's program, and I have point-set topology with a professor who gives a Massive amount of homework, so I don't have as much time as I would like to give to this problem. I hope this helps you see why my focus has been so narrow, I just want to round out my understanding of this monster, bring it to a close, and go on with my life. It isn't efficient? - Okay. Does every prime have such a representation? - nobody seems to have any clear idea of how to prove it does or doesn't, least of all myself. When I come up with an answer, I'll let you know. Aaron

2004-10-25, 20:39	#14
synergy Aug 2004 101010₂ Posts	By the way, after this and the above post referring to certain "inconsistencies in communication on the part of both parties" to try to phrase things in a non-personal fashion, after these two posts I will refrain from posting on this thread so that it will drop out of sight. When I get the representation of primes < 121 done, I'll start a new thread for it. I didn't mean for the "calling you out" title to continue being perpetuated. Sorry. Aaron

2004-10-25, 21:17	#15
Xyzzy Aug 2002 2×3²×13×37 Posts	If Bob has to deal with even 1% of the stuff I get by email from people reading this forum, I can see how he might get annoyed... I get weird emails all the time from people claiming to have solved all sorts of math related stuff... I'm not sure why people would email me, because I don't have a clue at all... Usually I just reply and tell them to post in the forum... To date, I can't remember any of them actually doing this... All these letters are similar... After a while they start to sound like Nigerian scam emails... I can almost predict what they will say... I'm not saying your idea is wacko, but I can certainly see how someone who knows what they are doing might view anything like this... I know very little, but the few things I do know I know forwards and backwards... It would seems strange to me if someone with no training or experience came up to me and told me they had a breakthrough, even though I had spent years researching and working in said field... I'm not saying it couldn't happen... But I doubt it would happen a lot... IMO, the best thing to do is lay everything out on the table... No secrets... Just post... If it is legit you will find out soon... If not, you will find that out too... If someone gave me pointers, even if they were grouchy, I'd go look them up... The older I get the more I realize how little I know... Maybe that attitude is holding me back, but that's how I feel...

2004-10-25, 21:38	#16
amateur Aug 2004 24₈ Posts	[QUOTE=Xyzzy] IMO, the best thing to do is lay everything out on the table... No secrets... Just post... If it is legit you will find out soon... If not, you will find that out too... If someone gave me pointers, even if they were grouchy, I'd go look them up... [QUOTE] Thanks, Xyzzy. The reasons for secrecy are twofold. First, if what I have here is old news or a nonsense, I would like as few people as possible to know about it. I guess I don't have to explain why... People just don't like to be embarrassed in front of thousand others. Secondly, if the idea is worthwhile, I would like to work it out as well as I can; maybe with the help of an expert. You know, someone who works with this might see something in it that I don't know. With other words, it could be the solution to something I have never thought of. The reason why I would like to talk to an expert is the same why we see a doctor when we have a sore throat instead of showing it to the whole world. Only the doctor can tell me if what I have is just a cold or a throat cancer..... As a matter of fact I have expected this reaction. It is not the first time I try to ask an expert and all I get is haughtiness and lecturing. Where are the people who are truly enthusiastic about mathematics? The ones who can hardly wait to see some new ideas? Has that kind died out? Or it never existed, only in books and movies?

2004-10-26, 15:26	#18
amateur Aug 2004 14₁₆ Posts	Thanks, Mystwalker, I have written a long reply, but I erased it all. This just doesn't make sense. This is not what I wanted. I wanted help and I am caught in a debate with no end. It is just not worth MY TIME. Thanks anyways everybody.