The credit controversy

[scole of TSBT]

Well the party may be over, so you might not want to spend any more money on ASICs until we know exactly what will happen. From the BU board...

- in discussion: removing anonymous platform, credits will be reduced by a factor x for all cgminer apps (not compared with GFLOPs of GPUs, its meaningless but need a better value of boinc combined stats)

Probably the change will make 90 - 99% of the users to leave, but I don't blame them. I wish there was a completely automatic, cheat-resistant, and fair credit system.

At least they can use the hardware to mine bitcoins for themselves.

I wish the discussion will be nice. We are all humans and we make errors.

PS. That factor is going to be pretty big. The BOINC developer told it should be 1612.

Henri.

Thread here...http://www.bitcoinutopia.net/bitcoinuto ... d=283#2695

I'll still crunch to help Milkyway though.

[scole of TSBT]

Project admins seem to indicate that based on project supporter comments, they are not inclined to reduce the credits as suggested by the BOINC developers. Some are suggesting some much smaller reduction in future credits to meet those opposed somewhere in the middle. Stay tuned.

[myshortpencil]

The Czechs are angry. (From http://lists.ssl.berkeley.edu/mailman/listinfo: The boinc project mailing list). I'm thinking about doing a few billion credits for CNT :)

On Oct 22, 2014, at 6:41 PM, David Anderson <davea@ssl.berkeley.edu> wrote:

Dušan:

I agree with much of what you say.
I created BOINC with the goal of helping scientific research.
BOINC has been funded almost entirely by the U.S. National Science Foundation;
they fund BOINC in order to support science, in particular American science.
If BOINC's contribution to science declines (as it has been)
this support will end at some point.

At the same time, I've never tried to limit how BOINC can be used,
and it's been used in various ways that I didn't intend:
- industrial research (AQUA@home from DWAVE systems)
- prime-number and crypto applications by hobbyists
- public/for-profit use (Charity Engine)
- in-house use within companies
- Bitcoin Utopia
I've provided technical support for these projects, including BU.

I don't think what BU is doing is immoral
(although I agree it's pointless and a waste of energy).
The main problem with BU is credit.
Credit is supposed to measure FLOPs,
and they grant credit for something that's not FLOPs.
The long-term solution is for them to grant credit separately: see
http://boinc.berkeley.edu/trac/wiki/CreditGeneralized

The short-term solution is for the cross-project statistics sites to not include BU.
I strongly recommend this.
I suggest that you and other team founders contact the admins
of these sites and urge them to do that.

-- David

On 22-Oct-2014 11:57 AM, cnt@czechnationalteam.cz wrote:
Dear Sir/Madam,


I would like to address the leaders of main BOINC teams and Mr. David Anderson as
well, as I have been appointed to lead the discussion regarding the project Bitcoin
Utopia.

We have recently been watching the current events with the project of BOINC with
great dissatisfaction.

The recent events that have been damaging the whole BOINC, degrading the results of
dozens of projects for more than last ten years and demotivating common users.

We would like to appeal as follows:

1.

Distributed computing was based on using the residual computer power/output for
human progress in many fields of scientific research. During many years,
hundreds of thousands people have joined the project and the possibility to
compare the performance within the projects has been one of the biggest
advantage of the BOINC system.

Rating each of the project has been so far in some bounds and there has not been any
degradation in credit in global statistics there. Even after computations coming on
GPU, the evaluation still corresponded with hardware increasing in output and a lot
of projects are still taking advantage of that.

But the Project Bitcoin Utopia is out of any standards having been built several
years. Its evaluation exceeds several times compare to other BOINC projects. The
statistic data, having been gathered in many years, is completely destroyed and it
does not have any sense.

During several weeks of running the ASIC devices the total credit from Bitcoin
Utopia is 1/3 to other BOINC projects and this rate is continuously increasing.


2.

The aim of the project is not useful. Neither for scientific purposes nor for
mankind at all.

It is just a commercial project that does not have any sense in the system of BOINC.

Firstly, the founders of the project are the only one who have the biggest profit
from it and they make a profit with a disproportion to the owners of BC pools.

Secondly, funding provided to the projects is also very controversial.

Funding flowing to the projects is lower than expenditures on the energy used.

From this point of view I am asking: “ What is that project useful for in the
system of BOINC? Why should we, thanks to it, let the whole system of BOINC destroy ? ”

3.

Evaluation of the project might be the main problem. Conducted work on the
project of Bitcoin Utopia as far as power/output is several times higher
comparing to CPU or GPU. But it is not the output in FLOPS that the rate has
been calculated in BOINC so far. The same over evaluation could have been used
in any other projects such as QCN, for example, or the project Radioactive and
by this way to bound to frequency of tremors on the measured radiation – just to
make up whatever and then give reasons for that.

The same way, for example, we should provide some bonus in our project Asteroid for
its users who then can find a model of ideal asteroid.

Is there any use for BOINC?

The owners of the projects have always kept some moral rules so far and they have
been in some devaluation bounds in their evaluation.


Let’s do something together so that we will stop this devaluation activities of
Bitcoin Utopia, retroactively reset to a natural level or just remove the project
from global statistics at BOINCSTATS or FREE-DC.

Let’s make some clear rules for future.

Nobody wants to devaluate our long term activities and effort, to discourage
longstanding users and to suppress or even to destroy the whole BOINC .


Dear colleagues, before your replying to this appeal, try to consult the matter
first within your team and do not express just your personal opinion on that.

Try to express the opinion of the whole team, like I am doing right now here on
behalf of Czech National Team o.p.s.



With Kind Regards,


Dušan Vykouřil

director of Czech National Team o. p. s.

[scole of TSBT]

They need to stop playing the "BU is hurting other science projects" card until someone shows credible info from one of the stats sites that shows a decline due to BU and not normal ups and downs in project interest. I looked at a few and didn't see any such trend over the past 4 months. Rosetta is down a little, but GPUGrid is up. Can have as much to do with CPU challenges pulling CPUs away from a project for a bit rather than losing interest due to BU.

And the "Credit is supposed to measure FLOPs, and they grant credit for something that's not FLOPs." complaint just doesn't make sense either. The same SHA-256 calculations can be done by a CPU, GPU or ASIC. BU follows the credit new guidelines precisely for the CPU and GPU WUs. The credit for ASIC WUs are proportional to the CPU credits based on FLOPS, even less I think.

[PinkPenguin]

Before GPUs and ASICs the credit system was a way of calculating the value of one's contribution and even those with a single computer could visibly see their contributions, set themselves objectives and feel that, in their own small way, they were contributing something worthwhile... The problem is that BU has managed to swamp the credit system (however fairly or unfairly the credits are assigned by BU) and made the smaller contributors seem or, rather, feel irrelevant. Feelings, however irrational, have to be catered for if Boinc wants to remain an inclusive community and not end up like Bitcoin which has almost become the exclusive province of large scale operations (if you want to mine in any way seriously it is probably better to go for the lesser currencies and change them to Bitcoin when and if you get enough).

From a purely economic point of view mining bitcoins at this stage is just not likely cost effective (maybe with an ASIC but certainly not with a CPU or GPU)... unless you are very lucky or have a four leaf clover. It would probably be more cost effective, if less fun, to donate the money to science (as has already been pointed out on this board).

These objections are not new and came out with the early GPU projects (GPUGRID, ACQUA, MilkyWay, PrimeGrid etc...). However the GPU projects did not have as radical an effect on credits as the BU project which can make use of ASICs and I am not aware of ASICs which can be used to mine other currencies let alone contribute to other scientific projects...

Having said this I would make no moral judgement on anyone who participates in BU - it is contributing to science in it's own way and if it makes it fun then it ain't a bad thing. The Czech team has a point and it is a valid point, though I don't think we should take ourselves too seriously (the moral high ground is usually a quagmire at best)... we should just think about objectives and how to encourage people to participate in BOINC and how to make it fun for all.

[scole of TSBT]

I should have left it alone. I understand the criticism about the value of the project and high credit, but I think the real value lies in getting people thinking about using other hardware options to do faster and cheaper processing . There is a high end molecular simulation ASIC, ANTON, that makes supercomputers look like a 286 processor. A little out of BOINC project budgets, but so were 5000 GFLOP systems 10 years ago.

[Alez]

Bok at freeDC has already stated he will not delist a project. Willie at BAMStats has BU crunching to repay the RAID card that went pop. Can't see him delisting BU. DA has no power over the stats sites as they are all third party.
We've had the conversation over and over. Time for others to accept the situation and deal with it how they want. If you don't agree with BU then don't run it. If you have no problems with it, run it. If all you care about is the science then do the science and stop moaning about the points that you don't care about anyway.
As for science being the looser, crap. BU has not impacted on either my CPU or GPU output for various projects. In fact I am more willing to stick with projects with crap points because I run BU.
How do you define science anyway ? Personally I think SETI is a waste of electricity as are most mathematical projects, but that's my opinion and in no way would I try and foist that opinion on others. Too many keyboard warriors pretending they have the moral ground, whilst deigning the real reason they are bitching .. deep down they do care about the points and are sore at loosing their places.

[Buster Gunn]

This is not the first time DA has accused a project of "ruining" the concept. GPUGRID got hammered and so did Milky Way when their former admins went BONKERS. I personally don't see the need to award outrageous credits. It's now turned into a "who wants to spend the most money" show and that takes the fun out of it for me.
Hey what do I know. I still remember when you got 1 cobblestone for 1 Seti WU completed. Most of the projects had similar accounting.

This reminds me of how I used to love the game of baseball until the teams started paying outrageous salaries to the players. Now it is just something I surf by while looking for the BBC or the History channels. This year is a little different. The KC Royals (one of the smaller salaried teams) is kicking the collective butts of the higher salaried teams.

Money can't but everything.

[myshortpencil]

http://stats.free-dc.org/stats.php?page=boincteams

Free-DC stats now includes a Combined Teams total credit score without bitcoin link. To which I ask, where's the Combined Teams total credit score without GPU?

[Alez]

A very valid point. We could also ask where's the Combined Teams Total Credit score for only DA approved projects.

[myshortpencil]

^ +1