Letter to Professor Stuart Bale, Director SSL

Message boards : Number crunching : Letter to Professor Stuart Bale, Director SSL
Message board moderation

To post messages, you must log in.

1 · 2 · Next

AuthorMessage
Lionel

Send message
Joined: 25 Mar 00
Posts: 680
Credit: 563,640,304
RAC: 597
Australia
Message 1498013 - Posted: 31 Mar 2014, 23:58:58 UTC
Last modified: 1 Apr 2014, 0:03:34 UTC

Given the issue with the credit system (since implementation of v7) and the apparent lack of focus by Dr Anderson towards this issue, I sent the following letter to Professor Stuart Bale, Director Space Sciences Laboratories yesterday.

I know some will disagree with this, whilst others will agree. This is the nature of people. All I can say is that if you agree that the credit system is in need of revamp, then please send comment to Professor Stuart Bale. You may use my email as a base, or in full if you wish.

Email to Professor Stuart Bale, Director SSL follows:
-----------------------------------------------------

Professor Stuart Bale

I am writing to you to outline an issue that affects many, if not all, of the volunteers of Seti@home. Whilst it is fair to say that some are not concerned, that some are concerned but will accept the status quo, there are others that feel that this issue needs to be addressed and resolved appropriately and that it is not getting the attention that it deserves from the project’s scientists. I am one of those that fall in the last category. In July of 2013 (and again later that year) I wrote to Dr. David Anderson and Dr.Eric Korpela about this issue and to date have not had a reply from any of the parties mentioned.

On 1 June 2013, version 7 (v7) of seti@home was released. Many, including myself, migrated to v7 either on the day or over the next few days (in my case I migrated all computers to v7 on 1 June). The following week, optimised application became available and I installed those across all computers on the day they became available.

What I noticed after this has become the bane of many of the volunteers.

The system used to determine credit for v7 work units is not well bench marked against Seti@home Enhanced (v6) work units, nor against Astropulse (AP) work units. Whereas prior to 1 June, v6 and AP were well benchmarked against each other, v7 work units attract far less credit/acknowledgement than either v6 or AP work units, and take longer to complete. There is in effect a double side effect in here. As another volunteer recently put it, “if I do 1 hour of work on AP and get X credit, then I should reasonably expect to get X credit if I do 1 hour of v7 work”. This is not the case, at present anyone doing v7 work units can expect circa 60% of that compared to AP. It has not always been like this, in June and July, v7 work units attracted circa 50% of that compared to AP.

Dr. Korpela has appeared to have conducted some tweaking to try and resolve the issue but it did not work. As I have said before, I do not think tweaking is the answer. They need to look at the design (conceptual and logical) and determine the root cause of the issue. Once they understand the issue, then they can re-frame the solution and test, with volunteers if necessary. However at the moment they are not doing this which will only lead to more angst and frustration at both ends.

I would also like to offer the following comment on credit.

Whilst there is discussion between the volunteers on credit vs science, and some volunteers are focusing just on credit, the issue is not about credit as such. It’s about recognition.

There are many distributed computing projects to which people contribute resources. The manner and means in which those projects recognise individual contribution is through a system that is based on and allocates credits. Some projects choose to recognise a person’s contribution more than other projects, thus they grant a higher credit rate per contribution for that project. In short, credits are effectively an indication of a person’s contribution to a project.

In the case of "the New Credit System" implementation of version 7 work units by Berkeley, recognition of personal contribution has been significantly reduced. At present, the indication is that recognition for effort is notionally just over half that of what it was prior to the new recognition system being employed.

Lastly, can you please discuss this issue with Dr. David Anderson and Dr.Eric Korpela with the view towards framing a timetable for resolution of this issue which I believe most volunteers would heartily welcome.

regards
ID: 1498013 · Report as offensive
Profile Mr. Kevvy Crowdfunding Project Donor*Special Project $250 donor
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 15 May 99
Posts: 3776
Credit: 1,114,826,392
RAC: 3,319
Canada
Message 1498024 - Posted: 1 Apr 2014, 0:26:01 UTC - in response to Message 1498013.  
Last modified: 1 Apr 2014, 0:37:39 UTC

One thing to add to this is that the current system is encouraging AP hoarders who dump MB work units. This wastes project bandwidth and other resources, and further irritates the top contributors who do their workload fairly. Many of them have left SETI@Home or devote only a small fraction of their computing resources to the project because of the credit system.

Hope it can get fixed eventually in-house without needing any pressure from elsewhere.
ID: 1498024 · Report as offensive
Thomas
Volunteer tester

Send message
Joined: 9 Dec 11
Posts: 1499
Credit: 1,345,576
RAC: 0
France
Message 1498111 - Posted: 1 Apr 2014, 6:47:33 UTC - in response to Message 1498013.  

I'm not agree with this kind of methods, Lionel :(
You should not have addressed for one person who is not directly in charge of the SETI@home project.
It gives a bad image of our community and harms us.
You would have been able to make an internal petition to the community SETI@home, for example, that Eric and David would have seen if you are so much attached to your credits.
As Founder of team, I'm also attached to the credits and it's true which CreditNew do not pay us proportionally in our investment but it must be adjusted between us.
My two cents.
ID: 1498111 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 1498114 - Posted: 1 Apr 2014, 7:13:05 UTC
Last modified: 1 Apr 2014, 7:16:59 UTC

In strict sense DA is not in charge of the SETI@home project as well (though yes, he's listed on corresponding page).
SETI@home just _one_ of projects that using BOINC at this CreditScrew issue not SETI@home project-based one but BOINC issue. I'm quite confident that this issue just can't be solved by forces of SETI@home project per se, they bound to use CreditScrew paradigm and paradigm should be changed to achieve positive changes IMO.
So, no sense to write to peoples who in charge of SETI@home project, it's not where issue lies.
How to influence on decisions on BOINC-level - it's another BIG question. And I would not pretend I have answer cause failed to do anything on that level too many times...
SETI apps news
We're not gonna fight them. We're gonna transcend them.
ID: 1498114 · Report as offensive
Thomas
Volunteer tester

Send message
Joined: 9 Dec 11
Posts: 1499
Credit: 1,345,576
RAC: 0
France
Message 1498115 - Posted: 1 Apr 2014, 7:17:50 UTC - in response to Message 1498114.  

In strict sense DA is not in charge of the SETI@home project as well (though yes, he's listed on corresponding page).
SETI@home just _one_ of projects that using BOINC at this CreditScrew issue not SETI@home project-based one but BOINC issue. I'm quite confident that this issue just can't be solved by forces of SETI@home project per se, they bound to use CreditScrew paradigm and paradigm should be chamged to achieve positive changes IMO.
So, no sense to write to peoples who in charge of SETI@home project, it's not where issue lies.
How to influence on decisions on BOINC-level - it's another BIG question. And I would not pretend I have answer cause failed to do anything on that level too many times...

About SETI@home
ID: 1498115 · Report as offensive
Profile Link
Avatar

Send message
Joined: 18 Sep 03
Posts: 834
Credit: 1,807,369
RAC: 0
Germany
Message 1498116 - Posted: 1 Apr 2014, 7:19:33 UTC - in response to Message 1498111.  

You should not have addressed for one person who is not directly in charge of the SETI@home project.

People in charge of SETI@home have been addressed often enough, Eric has try few tweeks, David doesn't seem to care and as far as I have read in the forums still thinks that there's nothing wrong with CreditNew.



It gives a bad image of our community and harms us.

It gives an image of community, which is slowly falling apart because of this issue. So if anything, than it is CN that harms the community (and maybe even science if we also consider all those aborted MB WUs) and not this letter.
ID: 1498116 · Report as offensive
Profile Jord
Volunteer tester
Avatar

Send message
Joined: 9 Jun 99
Posts: 15184
Credit: 4,362,181
RAC: 3
Netherlands
Message 1498117 - Posted: 1 Apr 2014, 7:20:20 UTC - in response to Message 1498114.  

How to influence on decisions on BOINC-level - it's another BIG question. And I would not pretend I have answer cause failed to do anything on that level too many times...

There's a very simple way to do so... just do not try to influence, or at least, do not make it one email using big words. Instead keep on sending evidence upon evidence over the duration of weeks/months, show him how what you see is wrong, or other than you perceived it to be.

And else, eventually, David will come to his own senses and retract stuff he added. It happens.
ID: 1498117 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 1498118 - Posted: 1 Apr 2014, 7:20:46 UTC - in response to Message 1498115.  

In strict sense DA is not in charge of the SETI@home project as well (though yes, he's listed on corresponding page).
SETI@home just _one_ of projects that using BOINC at this CreditScrew issue not SETI@home project-based one but BOINC issue. I'm quite confident that this issue just can't be solved by forces of SETI@home project per se, they bound to use CreditScrew paradigm and paradigm should be chamged to achieve positive changes IMO.
So, no sense to write to peoples who in charge of SETI@home project, it's not where issue lies.
How to influence on decisions on BOINC-level - it's another BIG question. And I would not pretend I have answer cause failed to do anything on that level too many times...

About SETI@home

I know that he is listed there. But as I said, he is for BOINC, for for SETI...
SETI apps news
We're not gonna fight them. We're gonna transcend them.
ID: 1498118 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 1498119 - Posted: 1 Apr 2014, 7:23:18 UTC - in response to Message 1498117.  

How to influence on decisions on BOINC-level - it's another BIG question. And I would not pretend I have answer cause failed to do anything on that level too many times...

There's a very simple way to do so... just do not try to influence, or at least, do not make it one email using big words. Instead keep on sending evidence upon evidence over the duration of weeks/months, show him how what you see is wrong, or other than you perceived it to be.

And else, eventually, David will come to his own senses and retract stuff he added. It happens.

Sorry, sometimes I have time just to one single letter, and I not consider to fix BOINC as my daily job, I have another agenda. To me BOINC is just framework that helps SETI, nothing more. Some here for BOINC some for SETI, some for both. I'm for SETI.
SETI apps news
We're not gonna fight them. We're gonna transcend them.
ID: 1498119 · Report as offensive
Thomas
Volunteer tester

Send message
Joined: 9 Dec 11
Posts: 1499
Credit: 1,345,576
RAC: 0
France
Message 1498121 - Posted: 1 Apr 2014, 7:27:21 UTC - in response to Message 1498116.  

It gives a bad image of our community and harms us.

It gives an image of community, which is slowly falling apart because of this issue. So if anything, than it is CN that harms the community (and maybe even science if we also consider all those aborted MB WUs) and not this letter.

But Science is more important than credits.
And although I agree that the credits are very important.
I will not leave the project, and my teammates either, for a history of credits...
ID: 1498121 · Report as offensive
Profile Jord
Volunteer tester
Avatar

Send message
Joined: 9 Jun 99
Posts: 15184
Credit: 4,362,181
RAC: 3
Netherlands
Message 1498122 - Posted: 1 Apr 2014, 7:28:11 UTC - in response to Message 1498119.  

Sorry, sometimes I have time just to one single letter, and I not consider to fix BOINC as my daily job, I have another agenda. To me BOINC is just framework that helps SETI, nothing more. Some here for BOINC some for SETI, some for both. I'm for SETI.

And that's fine. I was just trying to say, that there are options available to get him to sway one way or another, when sufficient evidence has been brought to him. Although on the other hand, I have to agree that given the evidence he may just as well harden in his standpoint and blow you off for eternity.
ID: 1498122 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 1498124 - Posted: 1 Apr 2014, 7:28:32 UTC - in response to Message 1498118.  
Last modified: 1 Apr 2014, 7:47:48 UTC

In strict sense DA is not in charge of the SETI@home project as well (though yes, he's listed on corresponding page).
SETI@home just _one_ of projects that using BOINC at this CreditScrew issue not SETI@home project-based one but BOINC issue. I'm quite confident that this issue just can't be solved by forces of SETI@home project per se, they bound to use CreditScrew paradigm and paradigm should be chamged to achieve positive changes IMO.
So, no sense to write to peoples who in charge of SETI@home project, it's not where issue lies.
How to influence on decisions on BOINC-level - it's another BIG question. And I would not pretend I have answer cause failed to do anything on that level too many times...

About SETI@home

I know that he is listed there. But as I said, he is for BOINC, for for SETI...


to make it more clear some historical perspective: SETI was the _first_ BOINC project. One can say it gave birth to BOINC. That's where collaboration between BOINC devs and SETI devs started. But over time BOINC much increased in size and SETI's benefits from being The First are diminished. I would say now it's not benefits at all :/ Some new project can just leave CreditScrew for others and use own rewarding system. SETI@home seems can't do that (maybe just have no energy to do that, maybe the reason more complex).
SETI apps news
We're not gonna fight them. We're gonna transcend them.
ID: 1498124 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 1498127 - Posted: 1 Apr 2014, 7:36:37 UTC - in response to Message 1498122.  
Last modified: 1 Apr 2014, 7:45:39 UTC

Sorry, sometimes I have time just to one single letter, and I not consider to fix BOINC as my daily job, I have another agenda. To me BOINC is just framework that helps SETI, nothing more. Some here for BOINC some for SETI, some for both. I'm for SETI.

And that's fine. I was just trying to say, that there are options available to get him to sway one way or another, when sufficient evidence has been brought to him. Although on the other hand, I have to agree that given the evidence he may just as well harden in his standpoint and blow you off for eternity.


Yes, that's why I said I have no answer "how". Sometimes one way work sometimas not :) EDIT: I'm not sure of course that "letter to boss" could be right solution in any way, but it's apparent that something should be done...

And regarding CreditScrew: http://setiathome.berkeley.edu/forum_thread.php?id=74262&postid=1497795

So, IMHO, to fix that not some bugfix required indeed. Quite thorough rethinking of the goals and means of credit rewarding system is required.

It can't be FLOPs based, also, it can't be "stock CPU app as standart" based too. Variability of devices are too big.
Years ago I pointed on fundamental BOINC issue, it lacks of abstraction level that required. Many things could be improved with abstraction of DEVICE as main entity.
What we saw? BOINC was CPU oriented from beginning. Then GPU was added ad-hoc, all decoupling was done as numerous "bug-fixes" (separate scheduling and so on). All this could not be if BOINC would use DEVICE abstration with single type of CPU in initial stage then other devices could be added as well.
SETI apps news
We're not gonna fight them. We're gonna transcend them.
ID: 1498127 · Report as offensive
Profile Link
Avatar

Send message
Joined: 18 Sep 03
Posts: 834
Credit: 1,807,369
RAC: 0
Germany
Message 1498129 - Posted: 1 Apr 2014, 7:39:12 UTC - in response to Message 1498121.  

It gives a bad image of our community and harms us.

It gives an image of community, which is slowly falling apart because of this issue. So if anything, than it is CN that harms the community (and maybe even science if we also consider all those aborted MB WUs) and not this letter.

But Science is more important than credits.
And although I agree that the credits are very important.
I will not leave the project, and my teammates either, for a history of credits...

For you and most of your teammates it may be so, for many others it isn't. It's not only about people leaving, those do no real harm to the project, but it's also about people aborting MB WUs for to crunch more AP. I wonder how many MB WUs has already not been crunched and marked as bad with too many errors because of that. So giving the correct amount of credits might indeed have something to do with the quality of the science we are doing here.
ID: 1498129 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 1498130 - Posted: 1 Apr 2014, 7:42:17 UTC - in response to Message 1498129.  
Last modified: 1 Apr 2014, 7:42:26 UTC

but it's also about people aborting MB WUs for to crunch more AP.

+1
CreditScrew stimulates "cherry-picking" in harming form indeed.
SETI apps news
We're not gonna fight them. We're gonna transcend them.
ID: 1498130 · Report as offensive
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 34744
Credit: 261,360,520
RAC: 489
Australia
Message 1498131 - Posted: 1 Apr 2014, 7:46:13 UTC

I find it funny that 2 people who have posted here are guilty of cherry picking.

1 who aborts MB V7 work to get more AP work and the other who hoards AP work way above the set limits. :-(
ID: 1498131 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 1498134 - Posted: 1 Apr 2014, 7:50:19 UTC
Last modified: 1 Apr 2014, 7:59:14 UTC



http://lunatics.kwsn.net/2-windows/what-is-best-hardware-for-what-seti-application.0.html

Unfortunately, CreditScrew doesn't reflect this relative performance chart.
Though it coincide on ATi AP vs MB it stil pays more (much more!) for NV AP vs CUDA MB that directly opposite of that should be for relatively best device usage.
SETI apps news
We're not gonna fight them. We're gonna transcend them.
ID: 1498134 · Report as offensive
Profile betreger Project Donor
Avatar

Send message
Joined: 29 Jun 99
Posts: 11359
Credit: 29,581,041
RAC: 66
United States
Message 1498230 - Posted: 1 Apr 2014, 16:08:00 UTC - in response to Message 1498135.  

+ 1
ID: 1498230 · Report as offensive
Profile James Sotherden
Avatar

Send message
Joined: 16 May 99
Posts: 10436
Credit: 110,373,059
RAC: 54
United States
Message 1498461 - Posted: 2 Apr 2014, 5:58:35 UTC

And Im sure that letter was got forwarded to Dr. A. With a note saying, I have no clue what this means, So you take care of it as you see fit.
last year somebody else wrote a letter to the chancellor. See how far that got.

The only way action will be taken is if the user base drops down to, lets say the equal number of milkway crunchers. Im sure that would get soemones attention.

What I object to is one disgruntled super cruncher speaking for all of us.
[/quote]

Old James
ID: 1498461 · Report as offensive
Profile Bernie Vine
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 26 May 99
Posts: 9954
Credit: 103,452,613
RAC: 328
United Kingdom
Message 1498517 - Posted: 2 Apr 2014, 9:28:49 UTC

What I object to is one disgruntled super cruncher speaking for all of us.

Agreed
ID: 1498517 · Report as offensive
1 · 2 · Next

Message boards : Number crunching : Letter to Professor Stuart Bale, Director SSL


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.