How to get LL assignment back

[RickC]

I lost this assignment because it didn't check in for 60 days.

I have a 65% complete partial file with no errors so I want the assignment back.

Is there a way to get it back?


[Mon Mar 07 02:47:59 2011 - ver 25.11]
Got assignment 98xxxxxxxxxxxxxxxxxxxxxxxxxxxx2C: LL M48843239


[Thu Aug 11 02:05:59 2011 - ver 25.11]
Updating computer information on the server
Sending expected completion date for M48843239: Aug 25 2011
PrimeNet error 43: Invalid assignment key
ap: no such assignment key, GUID: 6fxxxxxxxxxxxxxxxxxxxxxxxxxxxx4e, key: 98xxxxxxxxxxxxxxxxxxxxxxxxxxxx2C

Rick

[ATH]

It's been reassigned today:
http://www.mersenne.org/report_expon...&B1=Get+status

You can still finish it and the one who finishes first gets first time LL credit and the other get double check credits. Just add this line to worktodo.txt:
Test=48843239,69,1

It will probably give an error that it's already assigned, but it will still finish and give you credit.

It's always wise to make a backup your partial files before doing stuff like this, just in case.

[Brian-E]

Although...
You might of course also decide that it's polite now to take on another assignment and to wait with this one until the new assignee has completed it as a first-time test, monitoring it with the link provided by ATH, before continuing it yourself as a double-check. This takes into consideration that the exponent may now have been assigned to someone, possibly a new participant, who particularly wants to be the first to test a Mersenne number.
Life is so full of difficult moral choices...

[Christenson]

I'm told that a large fraction of assignments in this range are returned uncompleted, so I think the OP should go ahead and finish the assignment. But he should also PM Prime95, who would be the final arbiter and be able to ask the server just who it is that got the assignment and might be able to discern whether they would be upset by this kind of thing.

Me, a first-time LL test is a lottery ticket, and if I end up with a double-check instead, oh well, I've got more of those than first-time checks already.

[Uncwilly]

I agree with ATH, finish it up and take the credit which ever way you get it.

[cheesehead]

Originally Posted by Brian-E

You might of course also decide that it's polite now to take on another assignment and to wait with this one until the new assignee has completed it as a first-time test, monitoring it with the link provided by ATH, before continuing it yourself as a double-check. This takes into consideration that the exponent may now have been assigned to someone, possibly a new participant, who particularly wants to be the first to test a Mersenne number.

While I see merit in this argument, in view of the OP's failure (for whatever reason) to report status for 60 days, so that the OP may be considered to have forfeited first-priority claim to the assignment, there are also these considerations:

1. The probability that the current assignee is a new participant who particularly wants to be the first to test a Mersenne number is low. It could well be, instead, that the current assignee is one of the teams or individuals having oodles of systems running assignments, someone well past the newbie stage.

2. Even in the case of a new participant, the new participant's expectation should have been tempered by this cautionary note in license.txt: "We are not responsible for lost prize money, fame, credit, etc. should someone accidentally or maliciously test the number you are working on and find it to be prime." (I.e., "You _did_ read license.txt, didn't you?")

OTOH, the preceding sentence in license.txt, "Every effort has been made to ensure that you will be assigned an exponent that no one else has tested or is testing", does not adequately warn the new (or old) participant of the possibility that a reassigned-after-60-days exponent may still be in the midst of being tested by the previous assignee -- i.e., that "every effort" does not really mean every effort.

Also. "accidentally or maliciously" does not really cover the reassigned-after-60-days-but-still-tested-by-previous-assignee case, does it?

Life is so full of difficult moral choices...

Life's also, as has often been said, unfair. :-)

- - -

I agree with Uncwilly.

[ATH]

The new assignee has forfeited claim by not using username and password.

The chance that an ANONYMOUS LL test will finish is lower than an LL test assigned with username, since many anonymous assignments are new people testing it out and not completing it, or people not knowing what they are doing.

[cheesehead]

Originally Posted by ATH

The new assignee has forfeited claim by not using username and password.

Oops... I seem to have overlooked your previous post with the link to exponent status. My preceding comments were mistakenly based on our not knowing anything about the new assignee. My bad.

The chance that an ANONYMOUS LL test will finish is lower than an LL test assigned with username, since many anonymous assignments are new people testing it out and not completing it, or people not knowing what they are doing.

... which is compatible with the recommendation that RickC should go ahead and finish his in-progress test of 48843239.

[Brian-E]

Originally Posted by cheesehead

... which is compatible with the recommendation that RickC should go ahead and finish his in-progress test of 48843239.

Note that no-one has so far suggested otherwise.
I raised an issue which RickC might also consider in relation to the point when he allows his result to be reported to PrimeNet, without particularly recommending any course of action.

On the subject of anonymous participation: after version 5 of PrimeNet went live in 2008, all existing participants were encouraged to manually register their machines with the new server but numerous people did not do this. So results were still reported in the old way as if to the old v4 server, and these were registered as "anonymous" on v5. Is this situation still in force so that some "anonymous" participants are in fact simply users who joined before the change-over and never updated their settings?

[markr]

Originally Posted by ATH

The new assignee has forfeited claim by not using username and password.

The chance that an ANONYMOUS LL test will finish is lower than an LL test assigned with username, since many anonymous assignments are new people testing it out and not completing it, or people not knowing what they are doing.

It appears many people make a deliberate choice* to appear as "ANONYMOUS" - from the register new account / update account details page http://www.mersenne.org/update/:

Public Name for website report display (blank, or at least 2 characters and up to 32, A-Z, 0-9, dashes or underscores).
If not set, default public name display will be '(anonymous)'.

There are many ANONYMOUS entries well up in the top producers lists, some not far behind the "big one" in second place.

* Or maybe they just couldn't decide on a display name.

[ATH]

Originally Posted by markr

There are many ANONYMOUS entries well up in the top producers lists, some not far behind the "big one" in second place.

All ANONYMOUS stats are piled in 1 "user", and when primenet switched to v5 all users who didn't actively change their settings went on as anonymous.