There are two points of view here. They seem to be more or less 
irreconcilable. This argument repeats on a regular basis.

I have to say that:

(a) I agree completely that having an assignment "poached" is 
annoying to extremely offputting, especially for a relatively new 
user;

(b) I don't see any harm in the occasional triple-check. In fact 
completing these assignments does give some information of 
limited statistical value relating to the error rate.

Personally I feel it will never be possible to eliminate the problem 
entirely. After all, we're all volunteers; as such, firm application of 
discipline is not a realistic option. Probably the best way to rectify 
the situation is to make sure that when an unexpired assignment 
completion is notified by someone other than the person/team who 
owned the assignment;

(a) PrimeNet credits the person/team who owned the assignment, 
rather than the person/team reporting completion;

(b) in the case of a double-check assignment, or when a factor is 
reported, PrimeNet notifies the person/team who owned the 
assignment so that unneccessary waste of effort can be avoided. 
This is one way in which the email address associated with each 
user could usefully be employed - waiting for a routine check-in 
leaves effort being wasted in the interim.

If (unlike Mary) you're running a _first_ test which appears to be 
"poached", the probable reason is that the person who had the 
assignment before you has reported their result after allowing it to 
expire. You are _not_ wasting time by completing the assignment 
as a double check run will eventually be required - _unless_ a 
factor has been found!

As for Mary's current predicament, with that block of small 
exponents you have, my best suggestion is to reverse the order of 
the lines in worktodo.ini so that the _largest_ ones are done first. 
Intermediate checkins to PrimeNet will then cause unstarted 
assignments which are no longer neccessary to be automatically 
removed.


Regards
Brian Beesley
_________________________________________________________________________
Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm
Mersenne Prime FAQ      -- http://www.tasam.com/~lrwiman/FAQ-mers

Reply via email to