Newer milestone thread

[cheesehead]

Originally Posted by Brian-E

However, directing your post at aketilander in this case may possibly be inappropriate. I know his post apparently identifies himself as one of those who want milestones to be achieved as soon as possible and takes part in poaching to do so. But my hunch/guess is that he is using this form of writing merely in order to show empathy with the thought-processes behind milestone-chasing. And my feeling is that the real poaching culprits are less likely to identify themselves so publicly.

Thank you.

I see that I could have improved by using the third-person "one" more in place of second-person "you". I've noted that for next time.

[NBtarheel_33]

Originally Posted by Dubslow

Consider that he may have a large assignment queue, and that he's only been working on the assignment the last month or however long.

IIRC, it is bad GIMPSiquette to "hoard" large quantities (i.e. more than might be feasibly completed in three to six months) of assignments, especially those assignments in the "Preferred" realm that tend to hold up milestones. I know that Scott had mentioned at one point that worktodo files were not to be abused to queue large numbers of assignments.

[Dubslow]

Originally Posted by NBtarheel_33

IIRC, it is bad GIMPSiquette to "hoard" large quantities (i.e. more than might be feasibly completed in three to six months) of assignments, especially those assignments in the "Preferred" realm that tend to hold up milestones. I know that Scott had mentioned at one point that worktodo files were not to be abused to queue large numbers of assignments.

Well obviously, but that doesn't mean that everybody follows such policy.

[chalsall]

Originally Posted by cheesehead

I'm polishing it a bit more through each repetition.

Would you at least agree that an Anonymous account should be considered less trust-worthy than someone who has actually registered?

Or, put another way, if an Anonymous worker was "poached" that it probably wouldn't have an overall negative impact on the project?

[Dubslow]

Also keep in mind that many who do register appear as ANON anyways, one of the nice features of PrimeNet.

Originally Posted by Dubslow

Consider that he may have a large assignment queue, and that he's only been working on the assignment the last month or however long.

The Days to Go metric supports this. (I only now had the time to check the assignments page.) We should check back in two weeks and see what it says.


(As of now: 23 DtG, 2012/03/05 ETA.)

[aketilander]

Originally Posted by cheesehead

It already is done in an official and organized form! (But one could argue that this is not as clearly documented as it should be, so that newcomers learn of it only through random forum experience.)

Perhaps you're not yet fully familiar with the "procedures in PrimeNet and GIMPS" to which I referred. One of the less-prominently-documented is that the administrators (usually George) do keep an eye on such assignments, and will intervene when they think it's justified.

Well if there already is a procedure, which I was not aware of, to avoid this problem by manually reassigning exponents such as M24077267 to others, then I am satisfied. It will be the same thing I am asking for done in a manual way instead of an automated way.

Just for the sake of clarity it would be interesting to have a few examples of interventions in the past if that is possible.

[chalsall]

Originally Posted by Dubslow

Also keep in mind that many who do register appear as ANON anyways, one of the nice features of PrimeNet.

I disagree.

Please provide an example to support your claim.

[cheesehead]

Originally Posted by aketilander

To tell the whole truth I am working on a few exponents in the 50M region where I lost the original assignment and they were reassigned to others. I picked up old backup copies of the half way through exponents and finish them. It won't harm the project since its no loss of capacity on the contrary its a gain. A person actually doing the first time LL-tests could though consider this as poaching if I finish first but I consider it better doing it like this, as a premature double-check work, instead of waste all that work.

Yes, completing old unfinished/lost assignments as you describe is a gain.

The drawback is that if you finish and report first, the person who actually has the assignment now gets credited with a double-check instead of the first-time LL that s/he is expecting. This is not trivial, and needs to be avoided.

= = =

I can tell you, from personal experience in that second person's position, that for some people a mysterious "demotion" from first-time LL to double-check can be quite disturbing and discouraging.

In my early GIMPS experience, I actually had several of my first-time LL assignments poached. Because I did not then understand much about poaching, I was disturbed by this mysterious-to-me "malfunction" of the system (I started being informed that I was working on a DC instead of a first-time LL), and was actually too discouraged to continue doing LL. Only later did I understand what had happened.

GIMPS lost my potential contributions for that period. The poacher actually did cause GIMPS harm and actually did slow down GIMPS's progress because of that effect. The poacher didn't "speed up" anything. I had been making proper progress on each assignment when it was poached, and would have completed and reported it on time if I hadn't been poached.

GIMPS would today be somewhat further advanced, by the amount of work I could have done while I was discouraged, if no one had poached my assignments. GIMPS can never make up what it loses because of poaching.

= = =

Another consideration: Mersenne prime discovery

Some of us who've had a lifelong fascination with math are motivated by the possibility of becoming a Mersenne prime discoverer. That's a very rare and precious event, for us. Having our first-time LLs poached destroys that possibility.

Example:

Suppose that poacher XYZ (even if XYZ is just resuming an LL that used to be assigned but was then lost) proceeds with the LL and happens to be informed that s/he has found a Mersenne prime.

Think about the situation of the current legitimate assignee -- s/he will soon also (assuming neither LL run has had an error in the calculation) be informed that s/he has discovered a Mersenne prime! The incredible joy that might bring to him/her will be considerably tarnished when s/he learns that XYZ, after having previously abandoned the assignment, had resumed it with a faster system and "scooped" him/her in that discovery!

= = =

I think the ethical way for the LL-resumer to handle the situation is to voluntarily refrain from completing the LL until after the current assignee has reported his/her result. That way, the current assignee gets all the proper credit (first-time LL, plus the remote possibility of being a Mersenne prime discoverer). The LL-resumer can then finish up and report to get his/her proper credit (DC, plus the remote possibility of confirming a Mersenne prime discovery).

Furthermore, by waiting for the current assignee to finish, the LL-resumer can then register with PrimeNet to get an official DC assignment, thus avoiding the wasteful overlap that could occur if someone else started an unnecessary DC/triple-check on that exponent.

Using the established PrimeNet reservation system optimizes GIMPS progress by preventing unnecessary overlap. If all would-be poachers curbed their impatience and refrained from working outside the PrimeNet reservation system, they would help GIMPS in at least two ways:

(1) avoid performing unnecessary work, and

(2) avoid discouraging other users who trust the reservation system.

[chalsall]

Originally Posted by cheesehead

The drawback is that if you finish and report first, the person who actually has the assignment gets credited with a double-check instead of the first-time LL that s/he is expecting. This is not trivial, and needs to be avoided.

I can tell you, from personal experience in that second person's position, that for some people this mysterious "demotion" from first-time LL to double-check can be quite disturbing and discouraging.

But... Those who are credited with DC work get the same GDs credit as LL work, although in a different column.

Where is the loss?

(Except, perhaps, for those who take this non-serious game too seriously.)

[Dubslow]

Originally Posted by chalsall

I disagree.

Please provide an example to support your claim.

http://www.mersenne.org/update/

Originally Posted by PrimeNet

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)'.

@cheesehead: I interpreted his post to mean that PrimeNet/Prime95 conspired to unreserve them while he was in the middle of doing them. This happened to me a few months ago; I was about 45% through when suddenly my AID was invalid (my computer talks to PrimeNet more than once a day, and the AID was less than two weeks old). This was in no way my fault, and waiting with finishing the LL until whoever else got it is a waste of my time, if I even remember to check three months later if it's done. (If akeitlander voluntarily unreserved his expos, then ignore this post.)

[chalsall]

Originally Posted by cheesehead

It's not "automated", but this safeguard does indeed already exist and has already been used in the past.

Here is an interesting example demonstrating that perhaps PrimeNet is not sane: 25195613.

Hmmmm....