Just to stick my oar in a bit;

The Wu count does have some merit, and I am sure if there is enough demand for it, it will be added to the "tracking" sites so you will still be able to brag about your numbers.

The validation gives the results a better figure of merit because of the redundent processing. Now, there is some assurance that the results are valid results and now we even have tracibility to the class of machines that processsed the results. This allows, particularly for those projects (unlike SETI) where life is involved. This allows tracking back in case of an issue like the "Pentium Bug" possibly contaminating or biasing the results.

The only way that you know for sure how long a WU will take can only be measured after you are done processing. Central to issues like this are the processor architecture, effectiveness and size of cache, etc. So, there is no way that you can predict the time it is going to take, until you are done. With SETI@Home Classic I knew for most of my machines that the WU would take about 3 hours, but, there was no way to know before hand what the time was going to be.

There was a table that allowed you to predict only that a WU might take longer than normal because of its intrinsic capture (slew rate) but that could also be false because the Wu is so noisy that its processing is halted very early because of that. I have some WU that were abandoned in Beta in just a few minutes.

Anyway, I beat this to death in the FAQ's (with able assistance of John & John) and the glossary.

>> If only life were so simple. If UCB are the only people who count in
>> this
>> project - why do they need us? Oh yes - becuase we donate millions $$$
>> of
>> computer time free of charge!!!
>
> Surely the key word is "donate" as in "give". If I "give" UCB some computer
> time, I don't expect anything for it. If I didn't get one e-mail from an
> automated database anually saying "thanks for your continued support",
> personally I wouldn't really care. I CHOOSE to give. I wasn't promised
> anything in return. And that to me is the heart of the matter - it's up to me
> to take part or not under whatever rules and environment the project sets.

Your symantics are correct, but I doubt your arguement would hold true in the real world.

A donation is not necessarily given without expectation. For example, if I give money to the Railway Children of India (a superb charity I migth add), then I expect that money to go toward help those children and not some other worthy charity. Once given, I feel a sense of satisfaction and responsibility - my reward as it were. Now to attract my attention, the RCI have to advertise and explain why they believe in their cause, and hopefully to get you to believe too.

In the case of SETI, they need our CPU's, and because of that, a social culture has developed around this based on competition and achievement. Now there is no personal benefit for myself, and anomymous user, whether I achieve a top 1% or top 5%, but I feel a sense of achievement, just as I have a sense of satisfaction when I help the RCI.

SETI has to keep these donors motivated, just as the RCI have to make sure you know the problems in India still exist. My motivation for SETI was, of course, doing the Science, but it was also about my skills as an IT person, about learning and about achieving certain personal goals. Without this motivation, I would find it hard to justify why I have over 10 machines here, gigabit-ethernet, a dedicated bomb-proof server room (ummm - this is a legal requirement for all buildings in Switzerland - I didn't build it for SETI ;o), Unintruptable power supply and a flat whose rent includes electricity! ;o)

If you remove the ability for me to achieve goals or indulge in a bit of friendly competition, then you remove some of my motivation for SETI.

Extrapolated, this means the more changes BOINC make away from users ideals, the more users it will loose.

Now if the changes it is making warrant the small percentage loss, which it could redeem by attracting new users later - then there is no problem. But as always in life, I see other ways of achieving things without hurting or disappointing those around you.

That is what this discussion is all about - trying to suggest ideas which would make both the donors happy and BOINC run at full potential.


------------------------------------------------
Once you have ruled out the impossible, everything else, however improbable, is possible!
A.C. Doyle.

>
> > If only life were so simple. If UCB are the only people who count in
> this
> > project - why do they need us? Oh yes - becuase we donate millions $$$
> of
> > computer time free of charge!!!
>
> Surely the key word is "donate" as in "give". If I "give" UCB some computer
> time, I don't expect anything for it. If I didn't get one e-mail from an
> automated database anually saying "thanks for your continued support",
> personally I wouldn't really care. I CHOOSE to give. I wasn't promised
> anything in return. And that to me is the heart of the matter - it's up to me
> to take part or not under whatever rules and environment the project sets.
>
> >Therefore, if they want to keep us sweet, I guess respecting our part in
> this
> >project, and our viewpoints might help!!!
>
> The work so far shows that the team IS listening to us, but they will always
> have their own "red lines" - immovable rules which apply to BOINC and its
> sub-projects. Surely we can't expect to have our own way in all things?
> Especially as a lot of the posts on these forums at the moment are, I think, a
> release of frustration at not being able to up & download units rather
> than reasoned response or logical argument.

I agree with you both, and I have enjoyed reading this discussion whenever it pops up, it's such a volatile issue in which there is no clear winner. I myself think that the team has done a good job in how they have decided to organize the calculation of credits, it seems to me to be about the fairest method of all that have been discusssed.

But I have to say that as far as how Berkeley is handling the participants in BOINC goes, I have to agree with Nicholas that it leaves much to be desired. Granted, the object of prime importance here is the science and the results we obtain. But we are in the real world out here, not academia, and despite how much we may wish that everyone would just tow the line and not grumble - it's not going to happen, people have their own interests and their own goals.

In order for a project like this to be a success (and one of the reasons that SETI@Home was a success) it has to be able to appeal to people, to offer them something they find interesting enough to compensate them for the time, money, and at times like this, the grief that they are putting in to it.

I'm getting too far off topic, but basically I beleive the directors of the BOINC project (which I'm sure could desperately use more bodies to help spread the workload) need to pay more attention to the members, including just simply keeping us updated. There has been some marked improvement of late, but still most people are feeling left in the dark - sooner or later they are going to leave - and without them there is no project regardless of how hard they work on the technical end. I'm sure they are short of resources and are doing their best with a difficult situation, but I hope that when the present crisis is over and things cool down a bit that priorities will shift a bit to make the site more interesting and the communication with users more comprehensive.

> Your symantics are correct, but I doubt your arguement would hold true in the
> real world.

This is the real world. I've been doing SETI since April 2000 and amassed 63000 "points" from a spend of approximately £30000. I still don't care about said points. I am in a team. I enjoy the points competition both within my team and outside with other teams. But it isn't worth that sort of money to me, the science IS. And if you took the "points" away I'd keep crunching.

I'm not saying I'm right and anyone who takes part for the competition is wrong, or that my view is superior. But it is my view, and I am not alone in feeling this way. My motivation does not depend solely on points and never has.

> A donation is not necessarily given without expectation. For example, if I
> give money to the Railway Children of India (a superb charity I migth add),
> then I expect that money to go toward help those children and not some other
> worthy charity. Once given, I feel a sense of satisfaction and responsibility
> - my reward as it were. Now to attract my attention, the RCI have to
> advertise and explain why they believe in their cause, and hopefully to get
> you to believe too.

I do believe in the "cause" of SETI. They don't need to advertise it to me.

> In the case of SETI, they need our CPU's, and because of that, a social
> culture has developed around this based on competition and achievement. Now
> there is no personal benefit for myself, and anomymous user, whether I achieve
> a top 1% or top 5%, but I feel a sense of achievement, just as I have a sense
> of satisfaction when I help the RCI.

Yes you are correct in that people feel a motivation and sense of achievement. Whether the competitive atmosphere has developed solely because of the need for free CPU time as you state is debateable, but relatively unimportant. I agree that some participants require a measure of "reward" in the form of points/leagues/teams/certificates or whatever, I would only argue that it is some not all. I have a "sense of satisfaction" every time I crunch a unit, whether it has spikes, gaussians or is just a bunch of white noise. Because I need UCB as much as they need me, I would have no way of searching for ETI without them.

> If you remove the ability for me to achieve goals or indulge in a bit of
> friendly competition, then you remove some of my motivation for SETI.

I can accept that, it doesn't make you a bad person ;), and I know you are not in a minority. But the same does not apply to all of us.

> Extrapolated, this means the more changes BOINC make away from users ideals,
> the more users it will loose.

On balance I think you are correct. It will not, however, lose the people interested in the science, and I think you underestimate how many there are.

SETI and BOINC will also gain users tired of seeing cheats prosper under the old bean-counting methodology where the science is secondary as long as that user can be "first".

The question is, when you have added and subtracted these users are you left with a pool sufficient to perform the task? I can't answer that, but I know SETI classic was over-subscribed and the same units were crunched repeatedly to give machines something to do, which does flat zero for science.

> That is what this discussion is all about - trying to suggest ideas which
> would make both the donors happy and BOINC run at full potential.

I agree and I have enjoyed it. I don't think we are being fair on UCB however by stating at a very early stage without a proper work-stream and some time to bed in that BOINC as it stands will not fly as a project. Surely we have to let it come to life before we try to kill it off?

From Bill Farrant:

I'm getting too far off topic, but basically I beleive the directors of the BOINC project (which I'm sure could desperately use more bodies to help spread the workload) need to pay more attention to the members, including just simply keeping us updated. There has been some marked improvement of late, but still most people are feeling left in the dark - sooner or later they are going to leave - and without them there is no project regardless of how hard they work on the technical end. I'm sure they are short of resources and are doing their best with a difficult situation, but I hope that when the present crisis is over and things cool down a bit that priorities will shift a bit to make the site more interesting and the communication with users more comprehensive.

I agree communucation has been woefully inadequate and has probably cost the project dear. I wouldn't even try and justify this as acceptable.

bump

---

Pizza@Home Rays Place Rays place Forums

perhaps a fresh approach to the credit question is needed...how about this:

- if the computer is located in a Metric country, each completed work unit would receive 39.37 credits (the Metric cobblestone)

- if the computer is located in a non-Metric country, each completed work unit would receive 36.00 credits (the Customary cobblestone)

- if the computer is located in a country where people drive on the left-hand side of the road, each completed work unit would receive 40.00 credits (the Imperial cobblestone)

in case there is a dispute about Location, determination will be made by the NSA...

Sir Frank of the Wood
KWSN - the Knights Who Say Ni!

Citizens of the Peoples Republic of California demand 100 CS's, mainly because we are in danger of falling into the ocean after a major earthquake, if this demand is not met we will sue and sue and sue........And we are leaning to the far left err right depending on which direction you are looking.....;0)

---

And the beat goes on
Sonny and Cher

BOINC Wiki

I guess I'm wondering why this thread, from mid-2004, needed to resurface.

There is no perfect scoring system, and there is no "reform" that won't alienate some while making others happy.

---

But some one did.
Oh well.

It would have been simple to make everyone happy with the credit system.
Just keep four credit scores.

1) Give one Classic WU credit the instant a WU is uploaded. (This will keep all of us old Seti crunchers happy. Even allow this credit to be combine with the Seti@home credits.)

2) If any results fails validation, give one Cheater WU credit. (This will give all the cheaters and overclockers their own competition total and give those who dislike cheaters and overclockers with broken machines a way to know who they are.)

3) For each result finished, have an option for BOINC Manager to pause and await entry of the Feel Good credit the user feels that WU is worth, else just used the claimed credit for that WU. (This would give users either what their system claims or what they feel its worth. This would have to be limited to a ten digit input field.)

4) The current Cobblestone credit system. (So people who actually like the Cobblestone scores can compare actual work between projects. And so Dr. A. can write publishable papers.)

They could have this up and running with back credits in place by April.

At first glance, I was about to say: Oh no, not that One WU = One Credit nonsense again....

But, when I saw the bit about 'Cheater WU', I suddenly liked it!

It really would be nice to have a series of WU counters - not for the real 'official' statistics/ranking figures, but it would be facinating to see at least:
1) # WUs Issued
2) # WUs Returned
3) # WUs Good
4) # WUs Rejected/failed validation

I expect that the term 'Cheater WUs' might possibly upset someone along the way and could be avoided ;)

---

But some one did.
Oh well.

It would have been simple to make everyone happy with the credit system.
Just keep four credit scores.

2) If any results fails validation, give one Cheater WU credit. (This will give all the cheaters and overclockers their own competition total and give those who dislike cheaters and overclockers with broken machines a way to know who they are.)

That sounds good, with the email address of the top ten listed?