Join PrimeGrid
Returning Participants
Community
Leader Boards
Results
Other
drummers-lowrise
|
Message boards :
General discussion :
PrimeGrid Ethics
| Author |
Message |
|---|
 compositeVolunteer tester
Send message
Joined: 16 Feb 10
Posts: 1022
ID: 55391
Credit: 888,003,992
RAC: 140,912
                       
|
|
In this thread discussing "Number of Tasks" Michael Goetz wrote: If by "stub" you mean when the work unit page hides certain information,
we do that because some people were using that information to game the system
and unfairly make themselves more likely to be the prime finder. We hide the
information to make it fairer for everyone.
When it's safe to show that information, we show it.
There's a clear justification for hiding certain information for a period of time.
The fairness aspect being promoted here prevents a Ponzi-like scheme where
knowledgeable PrimeGrid participants would have an advantage over newcomers.
It's in PrimeGrid's interest to encourage newcomers to join, so a policy of
fairness aligns with PrimeGrid's interests. But it's also nice to hold forth PrimeGrid
as an example where social justice (fairness) directly supports societal benefits
(scientific advancement). NB This observation would probably earn me a good
grade in sociology, but I've never studied sociology, and never cared to.
Some PrimeGrid participants may have long secretly exploited the task information
to gain the advantage noted above. However it was only through a public
discussion on a PrimeGrid message board that this possibility even came to light. An
actual demonstration resulted in the swift implementation of hidden task information.
Full disclosure: I was involved in that discussion and proposed the theoretical "exploit".
That discussion topic doubtless rankled the PrimeGrid admins, but fell within the
moderation guidelines, so I'm thankful that PrimeGrid at least unofficially tolerates a
level of related research in the spirit of academic freedom, although it's not the kind
of research that PrimeGrid necessarily would like to see results for.
Or does it? Some clarity is needed. Academic freedom is just an assumed ethic here,
because PrimeGrid is engaged in scientific research and its association with education
is indirect through BOINC. Non-academic organizations (corporations, national security
agencies) also engage in scientific research but their ethics do not align with academic
freedom because they have different reasons for their existence.
PrimeGrid's rules and policies don't address "gaming" and other atypical conduct that
conflicts with PrimeGrid's interests because these are far from the typical users' experience.
Fairness is mentioned explicitly only in relation to moderation of posts, but obviously there
is an unstated ethic at work - fairness was mentioned twice in the quoted post.
Comments, Michael?
| |
|
|
|
|
can you please clarify what du you want to say?
____________
 | |
|
Michael Goetz Volunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13804
ID: 53948
Credit: 345,369,032
RAC: 4,797
                      
|
Comments, Michael?
I'm not sure what you want me to comment on.
____________
My lucky number is 75898524288+1 | |
|
compositeVolunteer tester
Send message
Joined: 16 Feb 10
Posts: 1022
ID: 55391
Credit: 888,003,992
RAC: 140,912
|
Comments, Michael?
I'm not sure what you want me to comment on.
Once we start tossing around flexible terms like "fairness",
it's better to come to a common understanding what it
means, rather than allow people to infer whatever they want.
So far, all we know is about fairness is that clever people
shouldn't be able to get better results than a concrete slab.
Is fairness about creating a level playing field for PrimeGrid
participants? Is it about ensuring that expected outcomes
related to incomes are achieved? Is it about maximizing
PrimeGrid's throughput? It's your call. | |
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13804
ID: 53948
Credit: 345,369,032
RAC: 4,797
|
Comments, Michael?
I'm not sure what you want me to comment on.
Once we start tossing around flexible terms like "fairness",
it's better to come to a common understanding what it
means, rather than allow people to infer whatever they want.
So far, all we know is about fairness is that clever people
shouldn't be able to get better results than a concrete slab.
Is fairness about creating a level playing field for PrimeGrid
participants? Is it about ensuring that expected outcomes
related to incomes are achieved? Is it about maximizing
PrimeGrid's throughput? It's your call.
Fairness is whatever I say it is, since I make the rules. In this case, selectively aborting tasks, especially by using a script, to make it more likely to finish first is unfair.
That is to say, there's no predefined, all-encompassing definition of "fairness". If there's some behavior that we deem to be unfair, we'll let you know.
Faking results (really cheating), needless to say, is way beyond "unfair" and will get you banned from PrimeGrid as well as having all your credit rescinded.
____________
My lucky number is 75898524288+1 | |
|
mikey
Send message
Joined: 17 Mar 09
Posts: 1398
ID: 37043
Credit: 592,082,665
RAC: 39,536
               
|
Comments, Michael?
I'm not sure what you want me to comment on.
Once we start tossing around flexible terms like "fairness",
it's better to come to a common understanding what it
means, rather than allow people to infer whatever they want.
So far, all we know is about fairness is that clever people
shouldn't be able to get better results than a concrete slab.
Is fairness about creating a level playing field for PrimeGrid
participants? Is it about ensuring that expected outcomes
related to incomes are achieved? Is it about maximizing
PrimeGrid's throughput? It's your call.
Fairness is whatever I say it is, since I make the rules. In this case, selectively aborting tasks, especially by using a script, to make it more likely to finish first is unfair.
That is to say, there's no predefined, all-encompassing definition of "fairness". If there's some behavior that we deem to be unfair, we'll let you know.
Faking results (really cheating), needless to say, is way beyond "unfair" and will get you banned from PrimeGrid as well as having all your credit rescinded.
To me a hard and fast definition is really hard because it's alot like some Judge in the US said about porn "I know it when I see it'. With more users comes more risk of people gaming the system for their own purposes. At Collatz it was someone trying to make more money, that was stopped here so that now makes crunching 'fair' here, but it doesn't stop the next person who thinks they 'have a better idea' to do their own thing that then needs stopping too. To me 'fairness' is the same chance for everyone to complete a workunit and be the 1st person to do it. Now all that depends on ALOT of variables but that's the basic premise to me. A friend once told me 'the measure of a persons character isn't what he does in public but instead what he does when no one is looking', to me that is also related to 'fairness'. | |
|
|
|
To me a hard and fast definition is really hard because...
In the context of this thread simply apply the "clever" action to everyone in Prime Grid.
Example 1:
Set your preferences to store zero work and app_config to use 1 ave_cpu.
Outcome for individual - possibility of a slight time savings vs wingman if workunits received at the same time.
Outcome for everyone - advantage eliminated.
Example 2:
Abort any tasks that would be completed as a double-checker.
Outcome for individual - more time spent as possible prime finder (positive), no half-points for Top Prime Finders Leader Board (negative).
Outcome for everyone - task results never double-checked, lack of verification slows or stops Prime Grid from reporting any primes, Prime Grid shuts down. | |
|
Message boards :
General discussion :
PrimeGrid Ethics |
