20051221, 20:50  #1 
Sep 2002
Austin, TX
3×11×17 Posts 
Should GIMPS immediately disclose alleged new primes?
Suppose GIMPS had a different philosophy on the discloser of an unverified prime. For example, the exponent was candidly (less candid than it is now) disclosed on the mersenneforum.org before verification. Would the press be more likely to prematurely report the finding? Is this the logic that keeps the unverified primes secret from active GIMPS participants, is it a formality in the scientific process, ect...? I would think no news agency would report on the finding until GIMPS released its official press release.

20051221, 22:38  #2  
Mar 2004
Belgium
29^{2} Posts 
Quote:
Yeah, but somebody did reported it already on SlashDot result: www.mersenne.org & on distributedcomputing.info forum...... I thought of deleting that thread... My $0.02: I think in the future, the new Mersenne Prime should not be announced announced, only to avoid these kind of problems.... Please don't flame me please 

20051222, 00:24  #3 
Jul 2004
Potsdam, Germany
3×277 Posts 
It indeed has been published at a lot of corresponding sites. But I think it's nevertheless an adequate procedure to only tell the particular number if one is certain it's prime.
In addition, you get two news items instead of only one. 
20051222, 00:29  #4  
Aug 2002
140_{16} Posts 
Quote:


20051222, 00:58  #5 
P90 years forever!
Aug 2002
Yeehaw, FL
7·1,069 Posts 
Several comments in no particular order of importance:
Prematurely disclosing the exponent would tempt some smaller operators to publish the result before verification completes. Yes, they might say it is pending verification, but they might not as reporters are of uneven caliber. At times it has been easier to "sell" the story to a reporter if he is told he will be the first to publish the number. If we predisclose the number and discoverer that increases the incentive to falsify a prime find for the notoriety of it all. If we predisclose the number and not the discoverer then the discoverer might not get the credit he is due if the news is published early. Prematurely announcing the exponent is not particulary harmful to GIMPS. The last several have been deduced by people in this forum with no ill effect. GIMPS reputation is firmly established such that reporting a false or faked prime find is unlikely to have much negative impact. If there were no press involved (my preference as it would make life much simpler) then I agree we could preannounce the exponent and discoverer. It is easier to "erase" a false find from a web page than a newspaper! Be thankful for multithreaded Glucas. Verification used to take well over a month!! Last fiddled with by Prime95 on 20051222 at 01:00 
20051222, 01:16  #6 
May 2004
28_{10} Posts 
Very good
I am very happy that Prime95 realizes that it is not harmful for GIMPS if someone guesses the number in advance. Still, there is only one blog in the world that informs its readers about the right number.
At any rate, everyone, including Curtis Cooper, seems to know that the number is around thirty million. How many people may learn the number before the Xmas? If they start to believe the single blog that has broken the news, there may be as many as four hundred thousand people who will get it  half of Boston. The propagation of information can be very fast; we are living after the year two thousand, after all. There are about four hundred truly active people involved in GIMPS. There is about a fifty percent probability that the discoverer would be a part of one of the two leading teams. We will still have to find seven Mersenne prime before the total number is nicely rounded. I hope that my encryption of the discoverer and the new Mersenne exponent has been sufficiently obscure for a casual reader. 
Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Small FFTs immediately crashes my computer, help please!  scrawlings  Information & Answers  39  20140802 21:48 
Alleged MSM doublestandards on Obama  cheesehead  Soap Box  56  20130619 15:25 
Alleged crimes and transgressions by recent U.S. Presidents  cheesehead  Soap Box  27  20111021 03:29 
Alleged Humongous Prime Number Thread #117  CR24  Miscellaneous Math  70  20100715 00:02 
mprime finished factoring almost immediately, possible HW problem?  optim  Software  4  20040621 22:19 