Prime95 License/Untrusted Versions?

[Dubslow]

I have a few questions which were initially unrelated, but the more I thought about it the more I realized the answers relate to each other, so I thought I'd start a new thread to ask them together. If the title seems a bit confusing, well, I wasn't quite sure what to put

Initially, it started with this question:

Originally Posted by Dubslow

If I were to succeed in building my own "untrused" version, would PrimeNet still accept/credit work like usual, and, more importantly, will it clear DC expos?

That is, what are the rules about modifying Prime95 for production GIMPS work?

Secondly, I've had this other, seemingly unrelated question bouncing around in my head. Wikipedia:

Although most of the GIMPS software's source code is publicly available, it is technically not free software as users must abide by the project's distribution terms if the software is used to discover a prime number with at least 100,000,000 decimal digits and wins the $150,000 bounty offered by the EFF.[3] As such, a user who uses Prime95 to discover a qualifying prime number would not be able to claim the prize directly. A free software package would not have this restriction.

Now I'm not sure who wrote that or what their definition of free software is, but somehow that sentence hasn't been sitting right with me. So:
Is Prime95 free software?



So, of course, I went digging through the GIMPS legalese as well as the FSF's definition of free software:

Originally Posted by FSF

A program is free software if the program's users have the four essential freedoms:

The freedom to run the program, for any purpose (freedom 0).
The freedom to study how the program works, and change it so it does your computing as you wish (freedom 1). Access to the source code is a precondition for this.
The freedom to redistribute copies so you can help your neighbor (freedom 2).
The freedom to distribute copies of your modified versions to others (freedom 3). By doing this you can give the whole community a chance to benefit from your changes. Access to the source code is a precondition for this.

And Wikipedia's for good measure:

Free software, software libre or libre software is software that can be used, studied, and modified without restriction, and which can be copied and redistributed in modified or unmodified form either without restriction, or with restrictions that only ensure that further recipients have the same rights under which it was obtained and that manufacturers of consumer products incorporating free software provide the software as source code.

Now, in one case it's pretty clear the answers to these questions (in fact, the first question hardly applies). "non-Participants" meaning (loosely) people not crunching for GIMPS do not have to abide by the Terms and Conditions of Use, which means that to use Prime95 without GIMPS, you must only satisfy the 5 clause EULA, which reads:

This EULA applies to all versions of GIMPS Prime95 and MPrime software and source code ("Software").
Software is free to download and use indefinitely on any computer(s) you own or for which you have permission and authority to install and run Software. Software is not export-restricted.
To use the Software you agree to be bound by this EULA and the Terms and Conditions of Use.
GIMPS reserves the right to change this EULA without notice and with reasonable retroactive effect. Last updated 15 October 2008.
GIMPS not responsible for any damages or losses arising from use of Software. SOFTWARE AND INFORMATION IS PROVIDED "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE IMPLIED WARRANTIES OF MERCHANTABILITY AND/OR FITNESS FOR A PARTICULAR PURPOSE.

It's not specific, but that's pretty clearly a "free license" by any definition.

Now, the two hard parts are what GIMPSters are allowed to do. Nowhere in the TCU/Rules are GIMPSters denied any of the FSF-defined freedoms -- however, as with my original question, I have no clue how PrimeNet would respond to modification of Prime95 (which would mean it's built without the security code). The second, and probably stickier problem, is the one that the Wiki article brings up, the EFF award. Here are the relevant parts of the Wiki/FSF pages:

Free software, software libre or libre software is software that can be used... without restriction, and which can be copied and redistributed in modified or unmodified form either without restriction, or with restrictions that only ensure that further recipients have the same rights under which it was obtained and that manufacturers of consumer products incorporating free software provide the software as source code.

Originally Posted by FSF

The freedom to run the program, for any purpose (freedom 0).
The freedom to study how the program works, and change it so it does your computing as you wish (freedom 1).

What counts as "using" Prime95? Does a restriction on how to use the results of the program count as "restricting freedom to" "run the program" or "do my computing as I wish"? Hell, while I'm at it, if I'm a GIMPSter, does running it count as "my computing"?



Even if you don't feel like getting into a philosophical/legal debate, how does PrimeNet respond to "untrusted" Prime95 versions?

[dabaichi]

Originally Posted by Dubslow

Even if you don't feel like getting into a philosophical/legal debate, how does PrimeNet respond to "untrusted" Prime95 versions?

I have the same question as well. I think if you manually report the results to PrimeNet, it will accept them.

[bcp19]

P95 would probably fall under shareware. A program like MLucas, CUDALucas, LLR, etc, would probably be considered freeware. If you had the programming knowledge to examine the source code of P95 and create your own software package, I believe that would also fall under free software, but that may be a grey area.

GIMPS is a distributed computing network. So technically, to my way of thinking at least, any work from Primenet is 'contracted' work, and any prime found from this contracted work would result in GIMPS taking overall credit and splitting the prize money as described, whether you use P95 or some other software to find it. If you wanted to risk duplicating work, you could use CUDALucas for example and randomly pick mersenne primes to test on your own and if you got lucky, you could then turn it in for the prize on your own. I suppose you could use P95 to work on 332M+ exponents and turn off reporting in the last few weeks and if a prime were found, manually report a bad residue, but you then have a conundrum... You have to a) rerun the test with a free software, that would take at least another year before you could submit your prime, b) worry that entire time someone might beat you to it and your chance at the money vanishes and c) wait another year+ for it to be verified prime before you even got your award. Personally, knowing that even a top notch system would take over a year to run a single test, I wouldn't worry about it, esp since I know of no place other than GIMPS that offers a reward for a new prime under 100M digits.

[Dubslow]

Originally Posted by bcp19

GIMPS is a distributed computing network. So technically, to my way of thinking at least, any work from Primenet is 'contracted' work, and any prime found from this contracted work would result in GIMPS taking overall credit and splitting the prize money as described, whether you use P95 or some other software to find it.

Hmm... That's why I'm wondering about "results" of a program, because the distribution rules apply to all participants, not just Prime95 users. In other words, it's not a part of the Prime95 license per se, but is part of GIMPS' rules. But then, in license.txt:

At the time of this software release, the Electronic Frontier Foundation is offering a $150,000 award to the first person or group to discover a hundred million digit prime number. If you find such a prime with the software provided, GIMPS will claim the award and distribute the award according to the rules published at http://mersenne.org/legal/#awards

But then, if you didn't use PrimeNet to either get work or find expos without a factor, then technically you're not a participant so those Rules don't apply; am I missing something here?

[chalsall]

Originally Posted by Dubslow

Hmm... That's why I'm wondering about "results" of a program, because the distribution rules apply to all participants, not just Prime95 users.

Read the terms carefully.

If you use any of the code developed by George to discover a Mersenne Prime, then you agree to his terms.

The "security" encryption is simply a way to prevent casual users from causing problems. This can be bypassed easily.

At the end of the day, it is possible for someone to use George's code to discover an unknown MP, and then not report it and write their own code and claim to have discovered it independently.

It would be more profitable to invest one's life savings in lottery tickets....

[Dubslow]

Originally Posted by chalsall

If you use any of the code developed by George to discover a Mersenne Prime, then you agree to his terms.

I think you missed my edit, because the only thing you could be referring to here that I can think of is the quote I edited into my previous post, where I mention that it's slightly contradictory with the legalese on PrimeNet.

(Incidentally, to justify this post a bit more, Google is fast: This thread is already second when searching for "gimps legalese" .)

[LaurV]

I did not read all the posts yet (I will do as soon as my coffee is ready!) and I don't want to transform this thread in a discussion about free versus not free software, but FSF definition is far in the weeds. They confuse the free software with open source software, which is a totally different dish.

The "precondition" of freedom 1 and the whole freedom 3 have no way to fit here. The whore freedom 1 is arguable. There are also one or two missing freedoms, let me see how can I formulate them. Of course, I am not talking about the "destructive" side, like the freedom to destroy the copy or to print it on a paper and use the paper when you go to the toilet :D

[Dubslow]

Originally Posted by LaurV

I did not read all the posts yet (I will do as soon as my coffee is ready!) and I don't want to transform this thread in a discussion about free versus not free software, but FSF definition is far in the weeds. They confuse the free software with open source software, which is a totally different dish.

The thread was mostly intended for discussion of that topic, and to be fair, I think the open source movement confuse themselves with FSF. (I guess I say that because "free software" came before "open source", not because one is better than the other.)

[chalsall]

Originally Posted by LaurV

I did not read all the posts yet (I will do as soon as my coffee is ready!) and I don't want to transform this thread in a discussion about free versus not free software, but FSF definition is far in the weeds. They confuse the free software with open source software, which is a totally different dish.

Have you ever tried to have a conversation with Richard Stallman?

While a brilliant man, he's got the attention span of a crane fly.

[Dubslow]

Originally Posted by chalsall

Have you ever tried to have a conversation with Richard Stallman?

Have you? I think it'd be fun (perhaps for precisely that reason. I'm sure xkcd's influence hasn't been that great for me )

[chalsall]

Originally Posted by Dubslow

Have you?

Yes.

Originally Posted by Dubslow

I think it'd be fun

It wasn't.