mersenneforum.org so who got the $25,000 cut of the$100k EFF prize?
 Register FAQ Search Today's Posts Mark Forums Read

 2012-09-18, 05:41 #1 ixfd64 Bemusing Prompter     "Danny" Dec 2002 California 2·32·137 Posts so who got the $25,000 cut of the$100k EFF prize? According to the GIMPS press release, George donated $25,000 of the$100k EFF prize to a "math-related charity" of his choice as outlined in the GIMPS prize distribution terms. However, the GIMPS press release gives no indication of the recipient. I've searched all the news from October 2009 (when the prize was announced) but could find no mention of any math-related group receiving a $25,000 gift. This gives rise to four possibilities: The money was not donated. George kept both$25,000 portions for GIMPS. George donated the money but asked to remain anonymous. George split it among several charities. To start off, the first scenario seems the least likely. I highly doubt George would go back on his word; it's dishonest and just doesn't fit his character. Similarly, the second scenario also seems extremely unlikely. While GIMPS could qualify as a math-related charity, this also doesn't seem like anything George would do. However,many people who make donations do ask to remain anonymous, and I have a feeling this is what may have happened. George said he donated the money to "a" charity, but I think it would be acceptable to donate to more than one in this case. Speaking of which, I believe the following could be considered math-related charities: Another math-related DC project. A school's math department. An organization that awards scholarships. A research institution. A science museum. A math teachers' union. Out of the six, I think a scholarship organization would be the most likely recipient as they usually receive the least funding. So my guess is that George split the money among several of such organizations and asked not to be named. This could explain why I couldn't find an instance of any group receiving a $25,000 donation (even an anonymous one). Any other guesses? Last fiddled with by ixfd64 on 2012-09-18 at 05:43  2012-09-18, 05:51 #2 Xyzzy Aug 2002 33×313 Posts Maybe the "Derek Zoolander Center For Kids Who Can't Read Good And Wanna Learn To Do Other Stuff Good Too"? 2012-09-18, 07:16 #3 jasong "Jason Goatcher" Mar 2005 3·7·167 Posts Quote:  Originally Posted by Xyzzy Maybe the "Derek Zoolander Center For Kids Who Can't Read Good And Wanna Learn To Do Other Stuff Good Too"? I watched the DVD of that movie with a friend and he just DID NOT get the humor. We ended up returning it and doing something else. I never have seen the end of that movie. 2012-09-21, 03:16 #4 Prime95 P90 years forever! Aug 2002 Yeehaw, FL 3·11·239 Posts Quote:  Originally Posted by ixfd64 According to the GIMPS press release, George donated$25,000 of the $100k EFF prize to a "math-related charity" of his choice as outlined in the GIMPS prize distribution terms. However, the GIMPS press release gives no indication of the recipient. I've searched all the news from October 2009 (when the prize was announced) but could find no mention of any math-related group receiving a$25,000 gift.
Some background:

At the time GIMPS received the $100K EFF award, we had committed to the following expenses:$50K to UCLA math department for the 10M digit prime discovery.
$20K ($3333 each) to the 6 previous prime discoveries.
$3K to M46 discoverer (this prime was discovered before we had a chance to post new award rules and I strongly felt it should be covered by our new award rules - whatever they might be)$3K to M47 discoverer.
~$1K to cover incorporation expenses and 501c3 filing fees. So we had already "spent"$77K of the $100K, leaving$23K.

We were facing the following potential expenses:

$5K -$7K for a new server.
$1K -$2K for software on a new server.
$300 -$500 / month for an ISP in case ours got tired of giving us free service (they were grumbling mightily).
$3K to fund the next Mersenne prime award. Given these potential expenses, I reluctantly concluded that the only prudent strategy going forward was to designate GIMPS as the math-related charity for the$25K. The fact that $56.7K of the$100K went to math related purposes ($50K to UCLA and$6.7K to Curtis Cooper's university) eased my conscious somewhat. Also note that should GIMPS ever shutdown, by law (501c3 law as I understand it) any remaining bank balances must go to math related charities.

Fast forward to today. I'm a tightwad. We've procrastinated on investing in a new server - though we can't hold out much longer. Our ISP that was grumbling about charging us going rates, has only recently started charging us a nominal amount. Unfortunately, no one has claimed the next $3K prime award. So essentially we are in the same postion we were 3 years ago. We have approximately$23K in the bank, we need a new server, and we have reserves to cover monthly ISP expenses and prime discovery awards.

 2012-09-21, 18:18 #5 ixfd64 Bemusing Prompter     "Danny" Dec 2002 California 2×32×137 Posts George, that is completely understandable. Running a large DC project can definitely cost a lot of money. Speaking of which, it's a little disheartening to see that nobody has donated to GIMPS for a long time. I'd certainly pitch in if it wasn't for my own financial situation (I'm the only one in my family with an income right now). Hopefully the discovery of M#48 will get the ball rolling. So keep finding those primes, people!
2012-09-21, 18:50   #6
chalsall
If I May

"Chris Halsall"
Sep 2002

35×43 Posts

Quote:
 Originally Posted by Prime95 We were facing the following potential expenses: $5K -$7K for a new server. $1K -$2K for software on a new server. $300 -$500 / month for an ISP in case ours got tired of giving us free service (they were grumbling mightily).
With respect, I would argue strongly you should consider renting a server, rather than buying one. And $300 to$500 a month in ISP charges alone is ridiculous.

Personally, I use and have had great success with 1and1 and Rackspace, and there are many other options available which would satisfy your requirements for much less money than you've itemized above.

And, to publicly put on the table again, both James and I would be happy to assist implementing PrimeNet V6 on a LAMP stack.

Last fiddled with by chalsall on 2012-09-21 at 20:52 Reason: s/happen/happy/ Sigh....

2012-09-21, 21:31   #7
c10ck3r

Aug 2010
Kansas

10438 Posts

Quote:
 Originally Posted by Prime95 Some background: At the time GIMPS received the $100K EFF award, we had committed to the following expenses:$50K to UCLA math department for the 10M digit prime discovery. $20K ($3333 each) to the 6 previous prime discoveries. $3K to M46 discoverer (this prime was discovered before we had a chance to post new award rules and I strongly felt it should be covered by our new award rules - whatever they might be)$3K to M47 discoverer.
I find it odd that M46 & M47's discoverers got less than M40-M45 this time...but then again, I'm just a third party with no vested interest...

2012-09-21, 21:59   #8
Prime95
P90 years forever!

Aug 2002
Yeehaw, FL

173178 Posts

Quote:
 Originally Posted by c10ck3r I find it odd that M46 & M47's discoverers got less than M40-M45 this time...but then again, I'm just a third party with no vested interest...
The rules before the 10M discovery were $50K to 10M discoverer,$20K split among discoverers of primes between 1M and 10M.

The rules after 10M were based upon what we felt we could safely fund. This is guesswork as expenses can change, donation amounts unknown, number of primes unknown, etc. We guessed $3K would be fairly safe. (At our present prime finding rate it could have been more ) 2012-09-22, 00:49 #9 davieddy "Lucan" Dec 2006 England 2×3×13×83 Posts Quote:  Originally Posted by Prime95 We guessed$3K would be fairly safe. (At our present prime finding rate it could have been more )
Yep.
\$3K per six years is a bit mean, even without taking inflation into account.

Still, some of us suckers keep trying :)

D

 2012-09-22, 08:17 #10 NBtarheel_33     "Nathan" Jul 2008 Maryland, USA 5·223 Posts What about scrapping the physical server, and running in the cloud? Perhaps one of the major cloud providers would be willing to offer GIMPS reduced rates. This would eliminate the need to maintain hardware and deal with its lifecycle. George, I read your comment about what happens to the money if GIMPS "shuts down". I truly hope that day, if it must come, is far off in the future. I was introduced to GIMPS in 2002 by one of my math professors (she had us visit the site and we had to find out what the largest-known prime was; at that time it was M13466917). I installed Prime95 on our old PIII 1 GHz and started crunching away (my first exponent was M14941313). Since then I have enjoyed working on the project, the thrill of finishing tests and waiting to see the result, and learning some good mathematics from this forum. This project has also increased my interest in computer performance and has given me enough encouragement to start upgrading my own systems and someday soon, attempt a build from scratch. Even studying the meta-data of the project (e.g. how fast we get to milestones, percentage of candidates with a factor, etc.) is interesting, on its own! So I certainly hope that GIMPS is able to keep advancing for a long time to come. I would truly feel bereft if there was no longer a reason to visit this forum, or check the milestones page, or if I didn't get to hear the fan on my Prescott do its impression of a jet engine. Thank you for all of the time and, undoubtedly, all of the money that you have invested in the project. It has brought us years of fun, excitement, and great results. I do believe that Father Mersenne would be proud.
 2012-09-22, 15:55 #11 ixfd64 Bemusing Prompter     "Danny" Dec 2002 California 246610 Posts A little OT: If GIMPS is ever to be shut down, then I hope the only reason for it is that someone finds a formula to easily generate Mersenne primes. Last fiddled with by ixfd64 on 2012-09-22 at 15:55

 Similar Threads Thread Thread Starter Forum Replies Last Post Gordon GPU Computing 1 2016-02-24 04:52 10metreh Aliquot Sequences 203 2010-11-14 15:00 10metreh Aliquot Sequences 0 2010-03-11 18:24 axn Sierpinski/Riesel Base 5 1 2006-08-01 20:07 ixfd64 Lounge 26 2006-01-20 02:33

All times are UTC. The time now is 19:48.

Sat May 21 19:48:17 UTC 2022 up 37 days, 17:49, 0 users, load averages: 1.37, 1.36, 1.42