I don't see why we need to send out messages and expect replies. Either we have an automated system for unreserving old (unupdated?) assignments or we don't. You could, as part of that system, include e-mailed notices like "your assignment will expire in x days [unless you extend the reservation]", but there's no need for admin intervention or emails from users explaining why they're taking so long.

[QUOTE=GARYP166;243997][FONT=Bookman Old Style][SIZE=3][/SIZE][/FONT][FONT=Bookman Old Style][SIZE=3]Dear member[/SIZE][/FONT]

[FONT=Bookman Old Style][SIZE=3]You were assigned exponent 123456 on 1st January 2009 but we have had no progress reports from you for 180 days. If there is a problem with your testing I should be grateful if you would contact me at the address at the head of this email; otherwise we look forward to receiving reports from you.[/SIZE][/FONT]

[FONT=Bookman Old Style][SIZE=3]YZ[/SIZE][/FONT]

[FONT=Bookman Old Style][SIZE=3]Gimps Administrator[/SIZE][/FONT]
[FONT=Bookman Old Style][SIZE=3][/SIZE][/FONT][/QUOTE]
You are aware, aren't you, that this situation is not the same as what was being discussed in this thread? If I understand correctly, a large proportion of contributors abandon their assignments in this way and these jobs are [I]routinely[/I] reassigned when they have not been updated for a certain period. Sometimes this goes wrong because the lapsed user/machine suddenly starts up again (or it never stopped but for some reason wasn't updating its progress) and then there will be duplicated effort and possibly an unintended triple check done, or a first time LL test converted to a double check. But usually the original assignee had indeed abandoned the job and the reassignment was exactly what was required.
The special situation occurs when someone is making steady but extremely slow progress and updating normally. That's when the temptation arises to steal that contributor's work.

[FONT=Bookman Old Style][SIZE=3]Yes I am aware and that’s why I introduced member Y. My concern was not with members who make slow progress – I believe we should let them alone – but with members who report no progress at all. You say that [/SIZE][/FONT]

[FONT=Arial Unicode MS][SIZE=3][COLOR=black]these jobs are [I]routinely[/I] reassigned when they have not been updated for a certain period[/COLOR][/SIZE][/FONT]

[FONT=Bookman Old Style][SIZE=3]but concede that [/SIZE][/FONT]

[COLOR=black][FONT=Arial Unicode MS][SIZE=3]Sometimes this goes wrong because the lapsed user/machine suddenly starts up again (or it never stopped but for some reason wasn't updating its progress[/SIZE][/FONT][/COLOR]

[FONT=Bookman Old Style][SIZE=3]My concern is that exponents are reassigned without the members being informed. Even if the emails are ignored – and I am sure that many will not be – I think they should be sent.[/SIZE][/FONT]


[FONT=Bookman Old Style][SIZE=3]To Mini Geek: maybe it’s not admin intervention if an exponent is reassigned – I need to get my terminology right – but it’s intervention of some sort. [/SIZE][/FONT]

On Nov 29:
[QUOTE=Uncwilly;239093][CODE]Countdown to testing all exponents below M(32582657) once: 1
Countdown to testing all exponents below M(37156667) once: 2
Countdown to proving M(24036583) is the 41st Mersenne Prime: 698
Countdown to proving M(25964951) is the 42nd Mersenne Prime: 25,341[/CODE][/QUOTE]

Now:
[CODE]Countdown to testing all exponents below M(42643801) once: 1,621
Countdown to testing all exponents below M(43112609) once: 2,417
Countdown to proving M(24036583) is the 41st Mersenne Prime: 497
Countdown to proving M(25964951) is the 42nd Mersenne Prime: 21,148[/CODE]

[QUOTE=Uncwilly;244750]On Nov 29:


Now:
[CODE]Countdown to testing all exponents below M(42643801) once: 1,621
Countdown to testing all exponents below M(43112609) once: 2,417
Countdown to proving M(24036583) is the 41st Mersenne Prime: 497
Countdown to proving M(25964951) is the 42nd Mersenne Prime: 21,148[/CODE][/QUOTE]

:goodposting:

Fascinating info that couldn't possibly be found elsewhere.

A Cynic

[QUOTE=A Cynic]Fascinating info that couldn't possibly be found elsewhere.[/QUOTE]Part of the idea is to record a snapshot of the data. If it is not captured, it can't be reconstructed. It can't be plotted, etc. once it is gone.

The reason that i reported it was the lowest "to prove number" dropped below 500.

Where are people able to access this countdown information? I'd like to follow the progress without depending on the forum.

[url]http://mersenne.org/report_milestones/[/url]

Another snapshot from the same source:

[quote]Major GIMPS Milestones:
[SIZE=2]December 25, 2010: All exponents below M(37156667) tested at least once.
December 25, 2010: All exponents below M(32582657) tested at least once.
July 11, 2010: Double-checking proves M(20996011) is the 40th Mersenne prime.
[/SIZE][/quote][SIZE=2]

Now I see what our biggest 2010 milestone was, and
where we're soon headed. Pointers ALWAYS appreciated.


[/SIZE]

[quote]:goodposting:

Fascinating info that couldn't possibly be found elsewhere.

A Cynic[/quote]I think some of us have caught on to your personalized
brand of irony/cynicism. But it's always possible to change.

[QUOTE=davar55;245163]I think some of us have caught on to your personalized
brand of irony/cynicism. But it's always possible to change.[/QUOTE]

Nicely put davar, but I try to choose my targets judiciously.
You and UncWilly post at about the same frequency as I do,
we are getting to know our respective styles well, and proved
ourselves capable of giving as good as we get.
We can all dish it out on occasions, and (I hope) can all take it!

To be serious(?) for a change, I agree with UncWilly's defence
of keeping a record of milestones, which is precisely why I suggested
that this "Newer Milestone" thread missed the point about backtracking
through the countdowns. This excercise now sprawls over several different
threads.
The first one was Petrw's (now in disuse) "Only 10,000 left",
referring to untested exponents with < 10M digits.
Since that "milestone" has just been achieved, perhaps someone could
bump that thread again.

BTW these "countdowns" are a fairly poor way of monitoring
GIMPS' progress. The "wavefronts" and rate of completion of LL and DC
are much more clear-cut indicators.
Re monitoring completion of LLs, a count of "status unknown" below say
60M would be a great help.

David