Bigadv points change

[texinga]

May I ask what the role of the DAB representative would have been to their respective Teams with regard to this recent change in points structure? If that role was to include alerting their Team and seeking input to pass onto Pande, I don't recall seeing any such communication on my Team. If the role does not include that kind of Team communication, then I would have to ask who a DAB member is actually representing?

To Bruce, if Pande wants to better manage how Folders feel about changes like this, I would offer the suggestion to be clearer about some of the things they say. To this day, I don't really understand what they want a Bigadv Folder to do after they say something like "As judged from the high demand for bigadv work units, this has been very much a success, perhaps a little too much so. We would like to continue to offer a bonus for bigadv to offset the above factors, but we don't want demand for bigadv to overwhelm the rest of the project or imbalance the points system". So, what do they want us to do? Stop Folding Bidadvs as much? Are they expecting us to invest in more GPUs (which I would not do). More clarity is what several people are asking for. Saying that they are hearing us does not return that clarity. There needs to be more than listening if people want to clear up misunderstandings.

[road-runner]

I can see the writing on the wall, manufacturers are going to screw up folding like they did for us little guys at Hwbot and forum wars... I hate to see them getting involved but that is what happens when you start bringing in the manufacture sponsored forums and people...

Let me rephrase, they are already screwing it up... More to come stay tuned...

[Jester]

May I ask what the role of the DAB representative would have been to their respective Teams with regard to this recent change in points structure? If that role was to include alerting their Team and seeking input to pass onto Pande, I don't recall seeing any such communication on my Team. If the role does not include that kind of Team communication, then I would have to ask who a DAB member is actually representing?

To Bruce, if Pande wants to better manage how Folders feel about changes like this, I would offer the suggestion to be clearer about some of the things they say. To this day, I don't really understand what they want a Bigadv Folder to do after they say something like "As judged from the high demand for bigadv work units, this has been very much a success, perhaps a little too much so. We would like to continue to offer a bonus for bigadv to offset the above factors, but we don't want demand for bigadv to overwhelm the rest of the project or imbalance the points system". So, what do they want us to do? Stop Folding Bidadvs as much? Are they expecting us to invest in more GPUs (which I would not do). More clarity is what several people are asking for. Saying that they are hearing us does not return that clarity. There needs to be more than listening if people want to clear up misunderstandings.

If indeed demand for Bigadv was a victim of it's own success then there should've been a lot more thought and discussion put into any changes to it, while something that's causing unrest among members needs to be "dealt with quickly" there should be equal weight to try and get it right, as I see it there was an issue with points awarded mainly from a new series of Big work units, and an oversubscription to the bigadv project possibly at the expense of equally worthwhile projects,
If that was the case then why reduce the bonus alone by such a large amount, surely the same or better could've been acheived by an adjustment to the return deadlines that determine the bonus multiplying factor,
That would too see a reduction in points awarded and also have the "culling" effect on marginal machines that are just getting inside the current deadlines, if the thought of such "culling" seems overly draconian, then look no further than item 4 and 5 here: http://folding.stanford.edu/English/FAQ-BestPractices

[joebighair]

Ok I have read enough and I mean I read every page of this thread, I found this thread when looking for the meaning of QRB and I found it interesting to start with. I had to stop folding for awhile because of my work didn't allow me to take part like I wanted but I started to get back in it just 2 days ago and I find this @%!# going on. I have been folding for a little over 2 yrs, I started with a 3.2Ghz p4 and a ps3 moved up to a quad 6600 then to my current q9450 OCed of course , went with a couple of gpus and folded away. I started out to help the team that taught me how to OC my CPU. Great bunch of guys/gals over at OCwiki. What they told me about folding was that it was about the science and what good it could do and that even with my p4 and ps3 I was helping out. What I have read here makes me ashamed that I am part of this group! I take that back, all you that spent big money on machines to as someone put it to show your E-penis and whining about a drop in points should be the ones ashamed. F@H was about helping to advance science and help cure diseases it was never to be about points, the points where to be used to have a little fun and to give a little appreciation for the help. Now there are teams out there giving money so you join their team (think GPU’S) I won’t mention their name because that is what they want to promote there products on the back of something that was good. So all you bigadv 48 core, 2600K, 12 core hacking freaks ruin the sprit in which folding was founded. So my 2 cents is pack up your toys and go to another distributed computing project and let the rest of us get back to the job of helping just for the sake of helping. That is all and good day to you!!

[7im]

May I ask what the role of the DAB representative would have been to their respective Teams with regard to this recent change in points structure? If that role was to include alerting their Team and seeking input to pass onto Pande, I don't recall seeing any such communication on my Team. If the role does not include that kind of Team communication, then I would have to ask who a DAB member is actually representing?

Some DAB reps may not be as active as other reps. You'd have to ask your DAB those questions.

... if Pande wants to better manage how Folders feel about changes like this, I would offer the suggestion to be clearer about some of the things they say. To this day, I don't really understand what they want a Bigadv Folder to do after they say something like "As judged from the high demand for bigadv work units, this has been very much a success, perhaps a little too much so. We would like to continue to offer a bonus for bigadv to offset the above factors, but we don't want demand for bigadv to overwhelm the rest of the project or imbalance the points system". So, what do they want us to do? Stop Folding Bidadvs as much?

How about following the Best Practices and project recommendations? When top teams openly flaunt written procedures how to trick the fah client in to downloading bigadv-12 work units on to systems with less than 12 cores, that's probably one part of what they speak. Or bigadv-8 WUs on to 6 or 4 core systems.

The Best Practices are pretty clear.

[MtM]

How about following the Best Practices and project recommendations? When top teams openly flaunt written procedures how to trick the fah client in to downloading bigadv-12 work units on to systems with less than 12 cores, that's probably one part of what they speak. Or bigadv-8 WUs on to 6 or 4 core systems.

The Best Practices are pretty clear.

How about changing the rules and kicking teams out of the DAB if they commit repeat offenses like these? Being in the DAB seems pretty prestigious for the teams, as only the 'top' teams are admitted. How about 'top' teams not only with points but with behavior as well

I don't like how these things are allowed to be without consequences, or should I say without public consequence? And no, a point change isn't what I'd call a public consequence, as consequence's for these offenses should not be spread out over all donors, they need to punish those who broke the ethical rules of folding.

I would extend that to when a team publicly discusses dumping work unit's in public.

Shape up, or loose your DAB spot.

@texinga

Ask your DAB in public, send him a pm to this thread... no more backroom secrets

[Jester]

How about following the Best Practices and project recommendations? When top teams openly flaunt written procedures how to trick the fah client in to downloading bigadv-12 work units on to systems with less than 12 cores, that's probably one part of what they speak. Or bigadv-8 WUs on to 6 or 4 core systems.

The Best Practices are pretty clear.

How about changing the rules and kicking teams out of the DAB if they commit repeat offenses like these? Being in the DAB seems pretty prestigious for the teams, as only the 'top' teams are admitted. How about 'top' teams not only with points but with behavior as well

I don't like how these things are allowed to be without consequences, or should I say without public consequence? And no, a point change isn't what I'd call a public consequence, as consequence's for these offenses should not be spread out over all donors, they need to punish those who broke the ethical rules of folding.

I would extend that to when a team publicly discusses dumping work unit's in public.

Shape up, or loose your DAB spot.


@texinga

Ask your DAB in public, send him a pm to this thread... no more backroom secrets

I don't think punishing anyone for any of the above is prudent in a contributor based project, but when something needs to be changed like the Bigadv project points why not as I said earlier simply reduce the deadlines,
There should be few complaints about such a reduction in light of the best practices guidelines,
Or in plain English " we have to reduce the value of Bigadv to not have it dominate the whole project and have reduced the deadlines accordingly following our original donor machine guidelines".
If then a "claimed bigadv rig" doesn't "measure up" how can the owner complain if they've been using other than "best practices" ?
Are there any 8 cpu core users out there who can't finish current Wu's well inside their deadlines ?

[MtM]

I disagree, strongly.

Teams, especially those who are represented in the DAB have a moral responsibility. Why would it not be prudent? Are you referring to the 'threat' of top teams possibly saying bye to f@h if they are punished for repeatedly breaking the guidelines set out by PG? That would be the 'wurst' case scenario...

Wurst between parenthesis, as I don't think it would be that bad.

Let me explain why: I don't believe a team as a whole can be bad. There are always bad apples, and in some teams even those bad apples might have gotten some 'status' due to having allot of points on their account, or for being involved in the project for such a long time. These are the people who I would describe as senior folders. The problem is the imo the hierarchy in a team, some people are allowed to say things which goes against the recommendations given by PG. If 7im puts forward that statement I believe him on his word, so without even knowing the teams involved I do say: if they are part of the DAB, kick them out. You can't zero a team because of a few bad apples, and yet I believe that giving of the signal to them that they are hurting the project ( and being kicked out of the DAB is a pretty serious signal, more so it's a public signal which I feel might be in order ) is a necessity. If they are a top team and not in the DAB it's more difficult, but maybe it's time to reorganize the DAB so that it's not limited to the top x teams ( in a way where the team's point total is the deciding factor ), maybe all teams in the top 10/20/50/100 should have a chance to have a voice in the DAB. I'm not saying Bruce isn't representing the smaller teams as best as possible, but it's not humanly possible for him to represent everyone BUT the top teams.

Also, the choice for the top x teams was made I think with the assumption that those represent the biggest group of donors. But I don't think that's true anymore, or at least not something which can be assumed from a point total alone. If I would take 10 people I know of having some real firepower and start a new team with them, we would make top x teams in a short time. Does that mean those 10 people are really representing anything like a majority vote considering donor considerations?

This is the problem I see, and this is a possible solution:

Make the DAB bigger, 10 representatives from different teams, but allow them to be from the top 20 teams. Enforce a democratic election of team representatives for the top 20 teams each year, and a democratic election between those representatives on this forum to define the DAB for the coming year. They have to choose 10 people and will not be allowed to vote for themselves, and the votes have to be public ( and this applies to the team's representative election as well ). Any team which does not want to hold a democratic election will forfeit their chance of making the DAB for that year ( how else to enforce this? ).

What is the downside: allot of work, needing to make sure elections are done fairly. If this were to be done, the DAB would need some public statues, possibly a subforum here where they can publicly discuss things of importance.

The upsides: transparency, clarity, and being sure of having representatives who really represent the donors ( and just as important: donors who know their representatives and know what they are doing ). Also, it makes sure the top teams follow best practices to the best of their abilities.

Edit:

I forgot Bruce, he would be the 11th member on the DAB representing teams below the electability threshold. That also makes the DAB an uneven number.

Another upside which could be utilized is as follows;

There are still allot of complaints about lack off communication.

When reorganizing the DAB, don't just increase it's size but extend it's function from only an advisory organ to the organ used for communications back to the donors, facilitated by the forum section mentioned before.

Also, if people doubt: yes I would hold a DAB representative accountable if their forums have guides which are not following the best practices guidelines. This would mean a DAB representative would have to be capable of moderating his own team forums, not having to depend on other people to agree with his view about what is allowed and what is not. The DAB members would have allot of accountability, but to have that much accountability also means they need the ability to enforce their own believes on their team forums.

I know people who posted in the past about not being able to stop certain threads on their team forums as they are not the position to remove them. That has to change will this be able to work.

*edited for spelling

[k1wi]

Somewhere in the past two or three pages this topic has headed off on a bit of a [DAB] tangent...

[Jester]

Some valid points there,
A DAB representative shouldn't automatically be, or become a forum Moderator as such, but at least given Moderator priviliges to edit or delete posts or threads that
are in their opinion clearly against best practices guidelines, if the team as a whole doesn't accept that then they forfeit the DAB position to the next team down the ladder,
if the current DAB members are in a situation where they feel powerless to have a voice on their team forums they should say so, and if nothing changes resign from the
DAB on principal alone.

[Jester]

Somewhere in the past two or three pages this topic has headed off on a bit of a [DAB] tangent...

True, but within context of points value changes, and those who may have some influence in those changes...
or do we want more pages of "the points are too high"..... "no they're not".... etc...

[Nathan_P]

How about following the Best Practices and project recommendations? When top teams openly flaunt written procedures how to trick the fah client in to downloading bigadv-12 work units on to systems with less than 12 cores, that's probably one part of what they speak. Or bigadv-8 WUs on to 6 or 4 core systems.

The Best Practices are pretty clear.

I frequent several of the top team forums and have never seen anyone doing this on any of the teams, perhaps some proof would be in order before you go throwing around accusations like this. I can however name several teams/individuals who have tricked clients/ dumped WU etc - most of who frequent this forum.

On the point of DAB and the points, 2 DAB reps have said that they never saw much in the way of discussion about the points change. One of them is even a beta tester and he didn't hear much about it either.

[phoenicis.]

Whilst entirely agreeing with the need for greater accountability and home forum authority for a wider DAB, I don’t think we’re in a position to criticise the individuals following the core hacking practice. Although the guidelines seem perfectly clear, when you drill down to the source quotes, the practice is only discouraged and is declared not to be cheating. The members look to this and the fact that their uber overclocked 2600Ks are almost as fast as 980Xs and can easily meet the deadlines as reasons to continue. They do not see that it is damaging the project by providing slower returns and discouraging the building of multi-proc machines because of the shortage of 6903s/6904s to feed them. We’re never all going to agree and they’re just doing what they think is right given the information available.

If a clear statement was made by Drs Pande or Kasson saying that this is definitely cheating and not permitted then, if we’re talking about the same thread, the forum admin/mod (who is a DAB member) would be in a better position to condemn the practice and close the thread without suffering a chorus of protests by their team members.

Without a software fix, this is the only approach I can see to reduce core hacking. Previously on this thread, Bruce pointed to the pitfalls of reducing the deadlines. Slower machines, but with the right number of cores, would still pick the units up but may not be able to complete them on time.

All this assumes that PG considers that the practice actually is damaging the science. If not, then the rule should be removed from the guidelines.

There’s no room for ambiguity otherwise we just end up trying to support our own point of view by picking at words here or sentences there. It’s as if we were trying to interpret scriptures from two thousand years ago where there’s no chance of clarification being sought from the author.

The same could be said for the recent points change.

Was it predominantly driven by the science or donor complaints?

If science, why were the 6903/6904s released with extremely large bonuses just a month before the change? Why were the standard bigadv not causing an imbalance for almost 2 years of beta and then suddenly impacting in such a way to cause a step change? What are the numbers sitting behind the imbalance and how have they changed over time?

If complaints, why were the standard bigadv adjusted at all considering the lack of comment prior to the release of 6903/6904s? Why was the 2684 reduced almost to the level of a typical standard SMP wu? Was the donor feedback representative of the whole and why was the only poll on the subject, however imperfect, not heeded or repeated with a suitably neutral question? Why were expectations not set well in advance so as to avoid unwise purchases?

It’s normal to try seek understanding from the decision makers so that we, the donors, can avoid making the same errors again going forward. I’ve made many errors including getting overly passionate/emotional regarding the project, buying too much of the wrong kit mistakenly believing that points were a guide to the most important science and behaving like a prat due no doubt to a bruised ego. I’ll try to adjust my approach to make sure these don’t repeat themselves (although I can’t guarantee the last one ).

If any mistakes were made by PG, no problem, we’re all human. The key thing is to inject confidence by setting out plans to avoid a recurrence.

The Project’s time is incredibly important and there will no doubt not be time to provide all the answers. However, as requested by texinga, some clarity/transparency would be really useful in giving some of us a sense of clear purpose and the confidence to start investing in the project’s infrastructure again.

Edit: Spelling.

[MtM]

Somewhere in the past two or three pages this topic has headed off on a bit of a [DAB] tangent...

The DAB came into play after people raised concerns about transparency considering what PG states to be donor feedback. I think the DAB should be more visible to not only PG but to the donors as well, so that concern alone would have been enough for me to respond to in the manner I did. The statement from 7im allowed it to expand to not only visibility but also accountability. When combining those two, I get the result as I outlined above.

In my opinion, this isn't off topic.

If you mean that an actual discussion about how the DAB could be reorganized would warrant a different thread, I agree. But I will not start one since I'm not in a position to judge if it's really needed ( more people will need to reply to form a better understanding of where the community stands on this as a whole ). If at a later stage the need for change becomes really clear, that would be soon enough to start another thread. If it would be done now, the discussion would only be spread out over a multitude of threads all covering the same subject ( be it from slightly different angles: one which is still discussing if it's needed to change things, and one which discusses how these changes need to look like ). I think it's first important to finish the discussion about the need for change, as it will help with the second part where this change would more clearly defined and having both at the same time will be confusing and potentially disruptive to the other discussions going on.

What if someone doesn't like a proposed change and therefore starts to argue against the need for a change? I'd say; prevent people from using the outcome of the second part as an argument in the first It's already been established that allot of donor's aren't here out of altruism, I don't have a problem with expanding that to accepting people will always try to defend their own interest.

Some valid points there,
A DAB representative shouldn't automatically be, or become a forum Moderator as such, but at least given Moderator priviliges to edit or delete posts or threads that
are in their opinion clearly against best practices guidelines, if the team as a whole doesn't accept that then they forfeit the DAB position to the next team down the ladder,
if the current DAB members are in a situation where they feel powerless to have a voice on their team forums they should say so, and if nothing changes resign from the
DAB on principal alone.

But a resignation wouldn't fix this in the long run, a bigger change is needed for that. And if you allow resignations, would you just pass on the spot to the next team down the point ladder after I explained that point total isn't related directly to representing the majority of donors?

Maybe I should have added another representative other then Bruce, one who is electable from this forum. That would broaden the amount of donors directly involved with choosing a representative to everyone not limited to people folding for the top teams. I do not want to diminish the efforts put into the DAB by Bruce, but if I am advocating a change aimed at transparency as well as trying to make more donors feel they are being listened to by making the DAB membership based on open elections, the DAB shouldn't be only represented by those who fold in a team or those who can vote for a team representative.

Not sure about how this reads, I do believe Bruce is doing a good job listening to the people on these forums and it might be enough if there is more transparency about what is discussed in the DAB to make the anonymous/smaller team folders feel they are surely being listened to and being represented by him.

On the other hand, adding another team neutral member to the DAB might be a good thing when looking at how the DAB's internal structure considering outward communications. I would expect each DAB member to be responsible for the feedback back to their teams, so the representative of this forum should be responsible for keeping a forum section going dedicated to facilitate communications to and from donors. Is it fair to ask Bruce to both admin this whole forum and take on the task of being the visible forum representative? Right now, it might not look like allot of work since there is little to no transparency, so almost nothing is known about the DAB, but if you lift that obfuscation there comes a big need to structure a new flow of information. Not even asking if he's capable of it, I believe his dedication alone would take care of that, but I would like to be sure the work load does not create conflicts with other responsibilities?

That is not a moral question, but an organizational one.

@kiwi...

I'm going against my own words, discussing possible changes without having a clear indication for a need first. I'll take a step back now.

Edit:

@phoenicis

Good points!

Edit2:

As mentioned here things don't need further clarification. That part of phoenicis's post wasn't actually a good point at all and I should have corrected it.

[Jester]

Whilst entirely agreeing with the need for greater accountability and home forum authority for a wider DAB, I don’t think we’re in a position to criticise the individuals following the core hacking practice. Although the guidelines seem perfectly clear, when you drill down to the source quotes, the practice is only discouraged and is declared not to be cheating. The members look to this and the fact that their uber overclocked 2600Ks are almost as fast as 980Xs and can easily meet the deadlines as reasons to continue. They do not see that it is damaging the project by providing slower returns and discouraging the building of multi-proc machines because of the shortage of 6903s/6904s to feed them. We’re never all going to agree and they’re just doing what they think is right given the information available.

If a clear statement was made by Drs Pande or Kasson saying that this is definitely cheating and not permitted then, if we’re talking about the same thread, the forum admin/mod (who is a DAB member) would be in a better position to condemn the practice and close the thread without suffering a chorus of protests by their team members.

Without a software fix, this is the only approach I can see to reduce core hacking. Previously on this thread, Bruce pointed to the pitfalls of reducing the deadlines. Slower machines, but with the right number of cores, would still pick the units up but may not be able to complete them on time.

All this assumes that PG considers that the practice actually is damaging the science. If not, then the rule should be removed from the guidelines.

There’s no room for ambiguity otherwise we just end up trying to support our own point of view by picking at words here or sentences there. It’s as if we were trying to interpret scriptures from two thousand years ago where there’s no chance of clarification being sought from the author.

The same could be said for the recent points change.

Was it predominantly driven by the science or donor complaints?

If science, why were the 6903/6904s released with extremely large bonuses just a month before the change? Why were the standard bigadv not causing an imbalance for almost 2 years of beta and then suddenly impacting in such a way to cause a step change? What are the numbers sitting behind the imbalance and how have they changed over time?

If complaints, why were the standard bigadv adjusted at all considering the lack of comment prior to the release of 6903/6904s? Why was the 2684 reduced almost to the level of a typical standard SMP wu? Was the donor feedback representative of the whole and why was the only poll on the subject, however imperfect, not heeded or repeated with a suitably neutral question? Why were expectations not set well in advance so as to avoid unwise purchases?

It’s normal to try seek understanding from the decision makers so that we, the donors, can avoid making the same errors again going forward. I’ve made many errors including getting overly passionate/emotional regarding the project, buying too much of the wrong kit mistakenly believing that points were a guide to the most important science and behaving like a prat due no doubt to a bruised ego. I’ll try to adjust my approach to make sure these don’t repeat themselves (although I can’t guarantee the last one ).

If any mistakes were made by PG, no problem, we’re all human. The key thing is to inject confidence by setting out plans to avoid a recurrence.

The Project’s time is incredibly important and there will no doubt not be time to provide all the answers. However, as requested by texinga, some clarity/transparency would be really useful in giving some of us a sense of clear purpose and the confidence to start investing in the project’s infrastructure again.

Edit: Spelling.

I think most of us have been guilty of "spitting the dummy" ... "throwing the teddy out of the pram" and other "prat like behaviour" from time to time,
possibly due to a bruised ego but likely too that we care about what we are doing for the project,
And 10 pages or more after starting this thread that was later merged, I for one have seen little to nothing as encouragement
to see which direction to follow to help increase my contributions, I've again seen a few things however that means as always my current hardware will still be folding 24/7,
but maybe with a little less enthusiasm just now.