Credit adjustment

Bikeman (Heinz-Bernd Eggenstein)
Bikeman (Heinz-...
Moderator
Joined: 28 Aug 06
Posts: 3522
Credit: 686138689
RAC: 562245

You'll get my vote ... Mike

You'll get my vote ... Mike floated the idea earlier as well, as quoted here.

We are already enough people now to form a picket line in front of the hotel of the upcoming pan-galactic BOINC workshop :-) !

CU
Bikeman
EDIT: I would not go so far as to stop exporting the stats. It's nice to have the statistics on how many credits you've collected in each project, what your rank is there, etc... The one thing that I think does not make sense is adding the credits together. It's like having bank accounts in multiple currencies and adding Yens and Euros and dollars to a total, IMHO.

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6534
Credit: 284727526
RAC: 105878

RE: Not my original idea

Message 83888 in response to message 83883

Quote:
Not my original idea ....


It's been floating around for as long as I've been involved ( 12/05 onwards ) and so probably has an even longer pedigree. This wheel, plus/minus personal variants, gets re-invented every time there is a credit spat. You'd have to do some deep mining to accredit the first mention of it, but in any case it's an obvious solution ( no disrespect intended ). I think it will come, maybe not this year or the next, largely because it will avoid the exhaustion induced by the perpetual re-negotiations currently endured. If exhaustion/endured are the wrong words then will anyone who truly enjoys the ( cross project ) credit issues please speak up and tell us what their trick is. :-)

Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter ...

... and my other CPU is a Ryzen 5950X :-) Blaise Pascal

Alinator
Alinator
Joined: 8 May 05
Posts: 927
Credit: 9352143
RAC: 0

LOL... Agreed, but you

LOL...

Agreed, but you guys' have always chosen to express the idea from the philosophical higher ground. That often leaves much room for interpretation. ;-)

I chose to finally stop tiptoeing on the eggshells around the dead and rotting elephant in the middle of the living room and state some of the practical ramifications of taking the red pill! :-)

My guess is the vast majority of the population have realized, that even for NEZ, the sum total of all your Cobblestones from every project would not even get you a 'Great One' from Dunkin' Donuts. What will happen is they will take the blue pill and just continue knocking tasks off as they come in. :-D

FWIW, I don't believe for a instant that it is impossible to do, anymore than it's impossible to get a widely disparate set of timekeeping apparatus to give accurate representations of the time of day based on the definition of the second.

You just can't do it given the current paradigm and constraints we have to work with.

I suppose it would be theoretically possible under the current paradigm if all projects would agree to use the same procedures to 'calibrate' their work to the reference standard. One would hope that since the goal here is science compliance with that wouldn't be an issue. Unfortunately that doesn't seem to be the case, and says something (fairly ugly) in and of itself. :-(

Alinator

archae86
archae86
Joined: 6 Dec 05
Posts: 3145
Credit: 7024364931
RAC: 1806575

RE: If exhaustion/endured

Message 83890 in response to message 83888

Quote:
If exhaustion/endured are the wrong words then will anyone who truly enjoys the ( cross project ) credit issues please speak up and tell us what their trick is. :-)

I don't enjoy the cross project credit discussions at all, and have not participated before this.

However, I do think that (approximate) cross project credit parity is useful, and that the current level of effort expended to this end at SETI and Einstein may be about right. Destroying the system because some vocal people endlessly complain about it seems not the right response. I'm apparently in a small minority of those posting here (including moderators for whom I have great respect) on think so, so just mention it to tidy the balance of opinion accounting a little.

I think Prof. Anderson's great invention was not technical but social. The original SETI arrived with a package of bait that successfully engaged a great many people to donate computer time (arguably free) and therefore electric power (never free, though sometimes paid by a third party) at considerable cost. The screen saver, the credits, the participation certificates, the image of scientific usefulness, the exotic nature of the quest, and (yes) these forums were all part of that package of inducements. My favorite of the lot happens to be the forums, but they probably play a much smaller role in attracting and keeping computation contribution than does the credit system.

Brian Silvers
Brian Silvers
Joined: 26 Aug 05
Posts: 772
Credit: 282700
RAC: 0

RE: Destroying the system

Message 83892 in response to message 83890

Quote:
Destroying the system because some vocal people endlessly complain about it seems not the right response.

In your opinion, are those "vocal people" in favor of credits as high as the sky, or are they the people who support the current iterations (and slight variations) of the "Cross Project Parity" ideology ("Cobblestone"-based)?

Ananas
Ananas
Joined: 22 Jan 05
Posts: 272
Credit: 2500681
RAC: 0

RE: ... 2.) All projects

Message 83893 in response to message 83885

Quote:

...

2.) All projects should reset the running count of credit granted for all entities every time the basis for granting credit changes. The reason here is that a running total implies that the basis is constant over time, when in fact it is not.

Alinator

This already made me leave a (non-BOINC) project and I will not stay at any project that does this.

Especially old work done on old computers consuming more energy and a lot more time for each workunit has a value for me. It is not monetary but it still counts, probably more than a monetary value could.

I'm not a credit hunter and I do crunch for projects that are known to give low credits because I like the projects - but resetting the credits would make me really angry.

MAGIC Quantum Mechanic
MAGIC Quantum M...
Joined: 18 Jan 05
Posts: 1695
Credit: 1042974310
RAC: 1373830

Do you mean that after my 8

Do you mean that after my 8 years of running many different computers until they go to pc heaven I will not be allowed to partake in the prize Dunkin' Doughnuts?

No Nobel Prize?

No autographed pictures of Ben Franklin?

Never getting the chance to drive my car among the gravitational waves or even among the protons in the LHC?

Not to mention the SETI@home classic workunits 14,117

And SETI@home classic CPU time of 89,585 hours along with all the other credits in the three projects I have been spending all this time searching for a Higgs boson?

No rotating neutron stars named after me?

Say it ain't so!


Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6534
Credit: 284727526
RAC: 105878

RE: Agreed, but you guys'

Message 83895 in response to message 83889

Quote:
Agreed, but you guys' have always chosen to express the idea from the philosophical higher ground. That often leaves much room for interpretation. ;-)


As I've said before I personally don't care for credits ( beyond as a technical marker/surrogate for computation progress ), but I care that others do. :-)

I wouldn't say my view was 'higher' in any regard, but broader perhaps. For instance I am actually fascinated that many seem to quite emotionally invest in this credit thing. Mind you others probably wonder why I seem, because I am, passionate about the whole LIGO science thing. Neither are 'bad' views, nor need scaling against dichotomies - higher/lower, better/worse, right/wrong. But given that all types of people engage here ( and all are furthering both science value and credit scores regardless ) then it is probably more a left-hand/right-hand thing. It is good to have two complementary hands. Imagine having two right hands, one on the right as usual and the other one the left .... :-)

So the science view can be pleased by the WU's being done regardless of motivation, and the competitive credit view can be satisfied by having a playground constructed for it to thrive within. Win/win? Overall yes, but like most human systems requires some vigilance to prevent decay or subversion. That's philosophical for sure, but I think reasonably reflects some outcomes already witnessed. I guess what I really would like is not an abolition of CPP as a technical construct, but a supression of the negative behaviours that have emerged in it's context. Apart from that I'm disappointed, like MAGIC, that donuts ( of any variety ) are not given their true and proper place in the cosmos. :-)

Cheers, Mike.

( edit ) What are these strange Dunkin' Doughnuts of which you speak? Are we missing out on some valuable treasure here DownUnda? :-)

I have made this letter longer than usual because I lack the time to make it shorter ...

... and my other CPU is a Ryzen 5950X :-) Blaise Pascal

Arion
Arion
Joined: 20 Mar 05
Posts: 147
Credit: 1626747
RAC: 0

I don't know if my situation

Message 83896 in response to message 83890

I don't know if my situation is unique or is more normal across the spectum but here's kind of how I look at things.

I started with Seti in 1999. It was new, it was exciting, and it was interesting. Back during the "Classic" years there were a lot of ups and downs with the project. The majority of us "old timers" stuck with the project with all the trials and tribulations just because.... Sure there were a number of people that flitted in and out, but to those of us that were dedicated to the project it didn't matter really. Then there came a serious PR problem with the project when it was announced that "cheating" on credits was rampant and causing a lot of problems within the community. (I believe I still have the article somewhere) A lady did a major study (I think from Canada) for a university and published it on how this was destroying moral, etc etc etc. I'd have to give David Anderson some consideration for his now wanting cross project parity based on that. (pet peev as some would put it)

To eliminate some of the problems with the "Classic" version the idea of boinc was introduced. The great line given everyone was that with Boinc as a platform it would allow other science projects to band together with a common platform where #1 If Seti was down for any reason, participants could use a backup project to get them through the downtime. #2 It gave easy access to everyone to know what other projects were out there. #3 Based on #1 and #2 you could chose a project that you were interested in for any number of reasons (scientific, benefit to humanity, climate modeling, etc) You could chose whether or not you devoted all your cpu cycles to that project or you could work on any number at the same time. It's only natural that with all these different projects and their purposes that a segment of the participants with a competitive spirit would look for projects that gave more credits than others.

In 2005 I came to Einstein because one of my sons in high school was extremely interested in phyics. I split my CPU cycles 50/50 between Seti and Einstein and life was good. When Seti was down Einstein just kept on going. I'm a bit ashamed to admit it but eventually all the problems at Seti took it's toll on me and I devoted all my time to Einstein. When Einstein had problems (infrequently) I just allowed more work at Seti until Einstein was back up. The whole premise of a backup project was still working as far as I was concerned. In late 2006 I heard of a project called Climate Prediction that was doing climate modeling and after seeing that it was going to take 6 months to complete a work unit I aborted. One of my Pet Peeves is the time it takes for credit to be awared. I'm a reasonable guy but sometimes seeing people jump in and grab work and then just disappear, leaving you waiting for someone else to pick it up after it timed out, (sometimes for months) started to get to me. I remembered CPDN and a work unit lasting for several months and I found out that every day it would send up a trickle and you were credited with what you'd done daily sounded really nice. (this problem with long response times for credit was/is still a problem with both Seti and Einstein) I think it was June 2007 when we were doing a crossover from S5R2/3? and there was a number of problems with the AMD processor vs Intel. I waited patiently for a couple of months still chugging along with Einstein in the hope that it would be resolved soon. I figured while I was waiting I'd just give CPDN some work and I eventually just devoted all my time there for about 6 or 8 months. It was NEVER about credits. The whole time I was crunching for CPDN I kept tabs on the progress here and then I found out about the optimized client and lo and behold it seemed to level the playing field as far as I was concerned about the AMD/Intel mess. When I finished the last of the work I had at CPDN I jumped back over here 100% and things were good again.

Enter summer 2008 and here we are once again with a change over and sure as snuff we got problems AGAIN!!!!!!!! We have a Windows/Linux disparity and on top of that "CREDITS" problems. All these years I was never that worried about how many credits I had. I wasn't #1 and in 4 lifetimes I'd still wouldn't be since I just can't beat out the big guys. It's always been a project that interested me, had few problems, was stable and reliable.

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6534
Credit: 284727526
RAC: 105878

RE: Sorry for the long

Message 83897 in response to message 83896

Quote:
Sorry for the long post. I couldn't think of any better way to express my opinion in a shorter version that would make any sense.


Don't be sorry! That's a valuable and well thought out response, and congratulations to you for your perseverance and concern. What a good lad you clearly have, to be taking a greater interest in his lady friend than arcane computations .... :-)

Maybe we should invite others to outline their personal DC/BOINC experiences, long or short, if only to at least reveal what assumptions underlie our respective involvements - here or at any other project. As a mod I can often see friction developing in discussions due to unstated/hidden differing assumptions. Mostly credit is the focus of that. Any takers? :-)

Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter ...

... and my other CPU is a Ryzen 5950X :-) Blaise Pascal

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.