Boinc's Death Knell?


log in

Advanced search

Message boards : Politics : Boinc's Death Knell?

1 · 2 · 3 · 4 . . . 9 · Next
Author Message
Profile Sirius B
Volunteer tester
Avatar
Send message
Joined: 26 Dec 00
Posts: 10227
Credit: 1,530,033
RAC: 266
United Kingdom
Message 1127046 - Posted: 12 Jul 2011, 23:40:49 UTC

Felt this was too hot a potato for N/C, so posted here.

Is this the beginning of the end for Boinc?

"I wish BOINC would allow projects to have their own credit assignment policies, but David Anderson is very clear about not wanting this option, and he is removing all the mechanisms for doing so." (quoted from another project)

One of my replies.....

"I'll support Plan B because as my signature shows, I enjoy crunching various projects that interest me. Not really bothered about credits, BUT it does have a value - Good natured competition between team members on their monthly Project of the Month.

Also, whenever I or others have a problem with Boinc or the individual project, there are many crunchers & project staff who provide assistance in resolving the issue.

That fact alone is what makes me continue to crunch. I can & will say this - should DA's policy continue to such a degree that it affects all Boinc projects, I will instantly detach from S@H. Compared to all the other projects, S@H is understaffed, underfunded, so in no way should it be THE NO 1 project.

DA opened Pandora's Box when he produced Boinc & as he now see's his pet project losing major ground, he's seeking "Eldorado" which in centuries, has never been found!

Good Luck DA, you're gonna need it!"

I orginally crunched S@H back in '99 using a Compaq Pro Desktop 386s with a 56k modem & found the whole idea brilliant. Years later, I search for S@H & find that the "Classic" is no longer available but Boinc is & found that even better.

Boinc is a manager enabling client applications to run with volunteers crunching the various WU's & so should not become a dictator as to how other projects award credits(or for the developer of Boinc to use his pet project as the standard bearer in the credit awards system - for that to be realised, then S@H MUST prove itself worthy - unfortunately, I don't think that will ever happen).

If DA continues on this path, not only will he be noted as the producer of Boinc, but also as the destroyer of Boinc!
____________

Profile ML1
Volunteer tester
Send message
Joined: 25 Nov 01
Posts: 8270
Credit: 4,071,566
RAC: 333
United Kingdom
Message 1127062 - Posted: 13 Jul 2011, 0:23:24 UTC
Last modified: 13 Jul 2011, 0:26:12 UTC

Boinc has been extraordinarily successful, both for the projects and for the forums and public outreach. Full credit there!

Meanwhile, the awarded credits counts have been and continue to be both great encouragement and a curse.

Myself, I curse and ignore the credits for their 'arbitrary' and highly variable nature. There's been various calls and suggestions for trying to bring some science into the credits system. So far, practical expediency has won out for something that is cobbled together as 'good enough'. (Hence the pun on the name of the units?) Fair enough on that, but I personally feel that a lot of science is lost in that the credits cannot be used as a reliable measurement tool.

I also wonder if s@h itself is being sacrificed to a 'greater good' and extreme testing and development of Boinc rather than pursuing a primary goal of SETI research. There is rather a long hiatus between a vast assembly of processed WUs and any subsequent results or direct SETI research publications...

Meanwhile, the s@h staff continue a dedicated miracle!


Time to rebalance?

Keep searchin',
Martin
____________
See new freedom: Mageia4
Linux Voice See & try out your OS Freedom!
The Future is what We make IT (GPLv3)

Profile Sirius B
Volunteer tester
Avatar
Send message
Joined: 26 Dec 00
Posts: 10227
Credit: 1,530,033
RAC: 266
United Kingdom
Message 1127083 - Posted: 13 Jul 2011, 1:53:19 UTC

I feel sorry for the S@H team though. They've been doing an excellent job with what little they have.

Unfortunately, this "new" idea of DA's has brought down Aqua twice already & if his "Control Freak Atitude" continues in the current vain, it's going to bring down a lot of other projects, AND that was not the original purpose of Boinc!

IMV, I'm thinking DA has become a political animal rather than a scientific genius.

Personally, I think he should savour the fact that he made it possible for many to do research via Distributed Computing an International Event, so should consider the fact that should he persist in that overbearing control style, that fact will disappear & he'll be forever known as "the Fool that destroyed Boinc"

Now that Pandora's Box has been opened, & should too many projects get brought down, I'm pretty certain that there will be someone out there working on a replacement for Boinc....

..ssssh, already in the pipeline.....

DOING

Dacom Open Infrastructure Networking Grid & totally copyright free!
____________

Profile Blurf
Volunteer tester
Send message
Joined: 2 Sep 06
Posts: 7413
Credit: 6,487,195
RAC: 4,374
United States
Message 1127085 - Posted: 13 Jul 2011, 2:04:09 UTC - in response to Message 1127083.

Unfortunately, this "new" idea of DA's has brought down Aqua twice already


Sirius....do you have some evidence that he was responsible???

Given I don't check their forums much but their own News Page states:

Last week the AQUA server had to be shut down because many configuration files were accidentally deleted.

____________


Profile Sirius B
Volunteer tester
Avatar
Send message
Joined: 26 Dec 00
Posts: 10227
Credit: 1,530,033
RAC: 266
United Kingdom
Message 1127087 - Posted: 13 Jul 2011, 2:15:59 UTC
Last modified: 13 Jul 2011, 2:17:25 UTC

Check the 1st post. Stated quote from other project. It wasn't a cruncher that posted that 1st quote but a project admin.

They would know better than us leprechauns.

Edit: - Also, don't forget his "famous" hiccup last year regarding credit. What makes you think he's let that slide?
____________

Profile Blurf
Volunteer tester
Send message
Joined: 2 Sep 06
Posts: 7413
Credit: 6,487,195
RAC: 4,374
United States
Message 1127089 - Posted: 13 Jul 2011, 2:27:59 UTC - in response to Message 1127087.
Last modified: 13 Jul 2011, 2:47:05 UTC

Check the 1st post. Stated quote from other project. It wasn't a cruncher that posted that 1st quote but a project admin.


Nowhere does that quote directly blame DA for the recent crash. Now granted I haven't always been a fan but your accusation is pretty serious. Again please display your evidence to support your accusation that he caused Aqua's recent crash.
____________


Profile Sirius B
Volunteer tester
Avatar
Send message
Joined: 26 Dec 00
Posts: 10227
Credit: 1,530,033
RAC: 266
United Kingdom
Message 1127093 - Posted: 13 Jul 2011, 2:50:48 UTC
Last modified: 13 Jul 2011, 2:53:11 UTC

WHY? The debate is not about Aqua but the tightening up of DA's credit in which he wants ALL projects to award credit similar to S@H.

I'll search for the proof if you can provide proof that DA is not removing the credit mechanisms from boinc...

...fair deal?

Edit: - that bit about DOING was not meant as sarcasism, it is actually being worked on.
____________

Profile Blurf
Volunteer tester
Send message
Joined: 2 Sep 06
Posts: 7413
Credit: 6,487,195
RAC: 4,374
United States
Message 1127095 - Posted: 13 Jul 2011, 2:54:39 UTC - in response to Message 1127093.
Last modified: 13 Jul 2011, 2:59:56 UTC

WHY? The debate is not about Aqua but the tightening up of DA's credit in which he wants ALL projects to award credit similar to S@H.

I'll search for the proof if you can provide proof that DA is not removing the credit mechanisms from boinc...

...fair deal?


But you made what appears to me to be a rather serious accusation as part of your discussion over his involvement in credits....I'd like to see you back it up. Either please do so or retract it.

As to the credit issue, it personally doesn't matter a great deal to me.
____________


Profile Sirius B
Volunteer tester
Avatar
Send message
Joined: 26 Dec 00
Posts: 10227
Credit: 1,530,033
RAC: 266
United Kingdom
Message 1127096 - Posted: 13 Jul 2011, 3:04:23 UTC - in response to Message 1127095.
Last modified: 13 Jul 2011, 3:06:33 UTC

WHY? The debate is not about Aqua but the tightening up of DA's credit in which he wants ALL projects to award credit similar to S@H.

I'll search for the proof if you can provide proof that DA is not removing the credit mechanisms from boinc...

...fair deal?


But you made what appears to me to be a rather serious accusation....I'd like to see you back it up. Either please do so or retract it.

As to the credit issue, it personally doesn't matter a great deal to me.



No need for me to retract it. It has been stated by a project admin that the boinc server updates for both June & July 2011 has caused serious issues with the credit allocation by awarding crazy amounts of credit. Boinc Dev's are aware of the situation, so why are you jumping onto DA's bandwagon?

Isn't that proof enough taking into account that, that was DA's intention?

As for the credit, it doesn't matter to me BUT what does matter is an individual acting like a dictator!

Boinc was produced to ASSIST the sciences not be locked down to suit the developer!
____________

Profile Sirius B
Volunteer tester
Avatar
Send message
Joined: 26 Dec 00
Posts: 10227
Credit: 1,530,033
RAC: 266
United Kingdom
Message 1127097 - Posted: 13 Jul 2011, 3:08:11 UTC
Last modified: 13 Jul 2011, 3:16:13 UTC

Also, & this is my personal opinion (& as this is the political board, I'm entitled to that opinion), I don't think S@H is up to the task of being the overall BOINC standard bearer!

Edit: DA is in fact boinc is he not? Doesn't he approve all the changes to boinc?

If not, then his ramblings regarding credit awards should not be happening.
____________

Profile Sirius B
Volunteer tester
Avatar
Send message
Joined: 26 Dec 00
Posts: 10227
Credit: 1,530,033
RAC: 266
United Kingdom
Message 1127102 - Posted: 13 Jul 2011, 3:19:04 UTC - in response to Message 1127085.

Unfortunately, this "new" idea of DA's has brought down Aqua twice already


Sirius....do you have some evidence that he was responsible???

Given I don't check their forums much but their own News Page states:

Last week the AQUA server had to be shut down because many configuration files were accidentally deleted.



Your post after this states you only read the news page, before you come out & state that others have made a serious accusation, I suggest you check the forums 1st, that way, heated debates can be avoided.

I never post anything on a whim or profess to know about subjects I do not.
____________

Profile Blurf
Volunteer tester
Send message
Joined: 2 Sep 06
Posts: 7413
Credit: 6,487,195
RAC: 4,374
United States
Message 1127107 - Posted: 13 Jul 2011, 3:36:28 UTC

Then I would suggest that you are wasting your hands by typing all of this. People have gone after DA on the forums here and it hasn't done a bit of good for anyone.
____________


Profile Sirius B
Volunteer tester
Avatar
Send message
Joined: 26 Dec 00
Posts: 10227
Credit: 1,530,033
RAC: 266
United Kingdom
Message 1127116 - Posted: 13 Jul 2011, 4:06:45 UTC - in response to Message 1127107.
Last modified: 13 Jul 2011, 4:09:24 UTC

Your still missing the point. Where does it say on any post that I'm after Dr Anderson?

If you cannot see the point I'm raising, then I can't understand why you're here let alone crunching!

The problem as I see it ( & I'll be damned if I haven't hit the nail on the head so to speak)....

Dr A & the Seti staff worked extremely hard to produce Boinc as they had seen how well the Seti classic took off & wanted to improve the program & also let others be able to do research at a much reduced cost.

THEY succeeded & I'm guessing they were highly elated at that success, & so they should be.

However, & I can truly understand this, while doing so,again guessing, but I've been there, so it more than likely happened, & that is due to that great team work, Dr A treats Boinc & possibly Seti as his baby, again so he should.

However, due to its international success, again all credit to Dr A & the team, it has outgrown that small team & even Dr A himself.

Unfortunately, he can't seem to accept that fact & is continuing to use S@H as a test bed.

how do we know that all this updates/upgrades are not behind some of the technical problems experienced by Seti?

Aqua has proved that the server updates affected them, so why not Seti?

Or even other projects?

Bill Gates developed Microsoft into what it is today..is he still there?

It is the same for all great companies/Universities. Great men/women developed them but either through stepping aside or death, those companies are still here with us today.

Dr A has produced a fantastic program & all I'm saying is: -

Stop Think Act.

Stop: - Halt all actions.

Think: - Am I on the way to destroying what I've helped create? Are my decisions/updates/upgrades affecting my own pet project? Are they causing problems for other projects?

Act: - If they are,what I can do to reverse the potential damage? do I continue regardless of the problems already caused? Should I stop using Seti as a testbed & let them regain the momentum & let seti regain it's standard?

As again IMV, if he continues in the same vain, Boinc will just become a sad footnote in history.

AND for your information Blurf, this is not a tirade, or an angst against Dr A - I want Boinc to continue & saddened that that was your reply, & a question for you - should a replacement for boinc be achieved, will you still crunch when boinc is dead & only the replacement is available?
____________

Profile ML1
Volunteer tester
Send message
Joined: 25 Nov 01
Posts: 8270
Credit: 4,071,566
RAC: 333
United Kingdom
Message 1127171 - Posted: 13 Jul 2011, 11:43:55 UTC
Last modified: 13 Jul 2011, 11:50:48 UTC

Mmmm... At least I've lost the direction and message of this thread...

My view/observation is that s@h is indeed used as a test-bed for all things Boinc. Especially so for the extremes of:


  • total number of participants;
  • number of active participants/hosts;
  • data throughput;
  • database load;
  • system resource requirements;
  • forums workload;
  • participant management;
  • and others?



s@h is also used as a 'unifying' benchmark to anchor some vague meaning to the credits system used. Wherein lies quite a turgid problem for projects that do not work in a sympathetic way to s@h and so we get the old apples vs oranges arguments...

After much past debate, my view is that the credits should instead be anchored on (and calibrated upon) compute resources used and scaled in value according to the utilised transistor count and the transistor switching count. That way, participants get rewarded for how well their particular hardware (and energy) is (are) utilised. However, that divorces any value of whether or not the use of those resources is 'worth' anything...

Perhaps there is a lingering comparison being made to the perceived 'value' of a s@h WU rather than making a reliable measure of the compute resource expended.

The present credits system may well be more political than scientific for the primary purpose of persuading ignorant paymasters to condescend to continue the funding for Boinc... Big 'credits numbers' work well for a political simplistic view. Note that the 'Cobblestone' allows for 'various' (simplistic) 'comparisons' to supercomputer performance...

Unfortunately, regardless of the science, the politics for funding is all a part of the game imposed upon science.


Meanwhile, the one snippet I've seen about Aqua suggests that the random credits may well be from uninitialised variables randomly confusing the credits count. Hence, that suggests that particular example is merely a normal debug problem rather than anything about doom laden upheavals.


All just a small part of normal development.

Keep searchin',
Martin
____________
See new freedom: Mageia4
Linux Voice See & try out your OS Freedom!
The Future is what We make IT (GPLv3)

Profile ignorance is no excuse
Avatar
Send message
Joined: 4 Oct 00
Posts: 9529
Credit: 44,433,274
RAC: 0
Korea, North
Message 1127214 - Posted: 13 Jul 2011, 15:11:44 UTC - in response to Message 1127171.

I figured I'd chime in here. I would like to see credit assigned equally across projects. Aqua thinks they can do it better and assign more credit for less work. Why? to garner more volunteers. To what end?

I would suggest having a multiplier system for new BOINC projects where the new projects can, for example, hand out 2.5X regular credits for the first 3 months of a project and reducing the multiplier by 0.5 for every 3 months until the credits are normalized. In this way you can get people to jump onto new projects for higher credits and perhaps get them to stay if they like the science.

for Aqua crashing it seems like they may have been attempting to circumvent the current BOINC credit rate and killed their own machines. Oh well I don't crunch there anyway.

If Aqua is so set on providing scewed value for their work then maybe they should consider creating their own interface of app much like F@H has always done.

All that being said having projects that throw out a random number for a credit really doesnt help anyone. If anything they should just eliminate the credit scale if that is how people want to run things
____________
In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope

End terrorism by building a school

msattler
Volunteer tester
Avatar
Send message
Joined: 9 Jul 00
Posts: 38322
Credit: 560,155,509
RAC: 652,863
United States
Message 1127228 - Posted: 13 Jul 2011, 15:32:47 UTC

Let me go on record here saying that for the most part I am on DA's side.
Why should he NOT require all projects under the Boinc banner to award credits on a consistent scale? After all, when it comes right down to it, they are all riding on the coattails of Boinc and Seti.

It has always kinda irked me that certain projects award ludicrous rates of credit for the work done there compared to Seti.
In part, because it has caused some to leave the Seti project not because they believe in another, but simply because they can generate dizzying numbers for their RAC pot.
Why SHOULD that continue? Just so other projects can try to lure participants with inflated credits? Did I read something about a certain popular project having 'double credit days'???

Equalize the credits and let the other projects stand or fall on their own merit or lack of same. Let the people choose based on the project, not the credit scale.

Their fear of that is the only reason another project admin would have to complain about not being able to give out whatever credit rate he feels like. Does not affect their science at all....
____________
*********************************************
Embrace your inner kitty...ya know ya wanna!

I have met a few friends in my life.
Most were cats.

Profile Sirius B
Volunteer tester
Avatar
Send message
Joined: 26 Dec 00
Posts: 10227
Credit: 1,530,033
RAC: 266
United Kingdom
Message 1127263 - Posted: 13 Jul 2011, 16:33:29 UTC
Last modified: 13 Jul 2011, 16:38:33 UTC

If Dr Anderson wanted total control & have all projects with credit awards similar to S@H, then he should not have made boinc open source.

As for using S@H as a testbed, it would make a lot more sense to reopen Boinc alpha & let the testers push his code & any changes to the limit.

Hmmn, closed due to it causing problems with the server? What does that tell an intelligent person about boinc?

Great on it's initial introduction a decade ago, but now.......
____________

Profile Gary Charpentier
Volunteer tester
Avatar
Send message
Joined: 25 Dec 00
Posts: 12136
Credit: 6,416,149
RAC: 8,081
United States
Message 1127268 - Posted: 13 Jul 2011, 16:41:00 UTC - in response to Message 1127228.

Suppose a project wanted to reward people who donate with credits at a faster rate for their cash donations? The more you donate the more credit per work unit. Is that an issue?

____________

msattler
Volunteer tester
Avatar
Send message
Joined: 9 Jul 00
Posts: 38322
Credit: 560,155,509
RAC: 652,863
United States
Message 1127273 - Posted: 13 Jul 2011, 16:46:10 UTC - in response to Message 1127268.

Suppose a project wanted to reward people who donate with credits at a faster rate for their cash donations? The more you donate the more credit per work unit. Is that an issue?

It should be. Since when are monetary donations a measure of work done for the project? Those are two separate issues, and should not be intermixed. RAC is a measure of work being done.

Perhaps better server access, or different colors of stars could be awarded to recognize levels of monetary support.
But the star thingy has been brought up by myself before, and was talked down as being divisive. Which is fine by me. One star fits all, whether the person has donated the minimum or a million.

____________
*********************************************
Embrace your inner kitty...ya know ya wanna!

I have met a few friends in my life.
Most were cats.

Profile Blurf
Volunteer tester
Send message
Joined: 2 Sep 06
Posts: 7413
Credit: 6,487,195
RAC: 4,374
United States
Message 1127318 - Posted: 13 Jul 2011, 19:02:07 UTC
Last modified: 13 Jul 2011, 19:07:58 UTC

Due to not having RAC, the Admin of Aqua (aka BoincAdmin) approved me to post a statement...partially from a post in Aqua--the other a personal statement to me:

I wouldn’t say that “DA crashed AQUA”. DA replaced our custom credit assignment mechanism with CreditNew, and after miscalculating a parameter, it started assigning insane amounts of credit. That is why we had to stop the work.


From http://aqua.dwavesys.com/forum_thread.php?id=854&nowrap=true#11742:


For AQUA, David Anderson has been a friend, not an enemy. We are software developers too and we know very well that working on a complex system like BOINC is hard. We would be surprised if there were never any bugs in such a system.

We may disagree with some of the recent changes in BOINC and were disappointed that we had to stop the work, but we don't consider disabling custom credit assignment as a personal insult (= we don't get emotional about it).

Our approach is to see if CreditNew can be made to work. Honestly, if there can be a standard and fair credit system then I am sure most project admins would be happy, because that is one less thing to worry about. Getting the scientific work done is the main concern of most project admins.

I know that the above may make some people angry, but please look at the big picture. BOINC credit is virtual. It is really just a token of appreciation for the help a volunteer has provided to a project. The most important thing here is that the amount of granted credit fairly represent the amount of help

____________


1 · 2 · 3 · 4 . . . 9 · Next

Message boards : Politics : Boinc's Death Knell?

Copyright © 2014 University of California