Brian

Off topic

What most do not know is that there is communications between Project Admins to this goal. There are credit adjustments being made that means no one has to start over. Over time it will not matter. People are with a Project because of what they feel their computer time is worth or their beliefs in the project goal.

It means that there is a recognized concern that you will not find in a project forum.

But you have still not answered my question. If credits are so meaningless and credit hounds such a minority why is there a need for project parity? See you guys have placed yourselves in somewhat of a conundrum.

You keep playing the game of on-ups-manship as if you are going to "outsmart" us and trip us up. It is that exact mentality that is causing conflict.

I've already explained that the reason for cross project parity is to prevent users, who put too much weight into credits, from using their weight to dictate to a project what they should do.

If nothing is done about cross project parity, then credits, which are reflective of the amount of science being done, can and will be used to sway favor of whatever reigning group resides at a particular project.

If you take away such a reason by making all credit/hr equal, then you effectively eliminate any power trips any user or group may attempt.

I see no difference.

...because you are not looking to see a difference. When one does not look, one surely cannot see. You are stubbornly convinced of your belief.

Unless I greatly misunderstand things, this would mean that all participants would get a boost, and the project would also get a boost, perhaps even helping the "latency" complaint (PhonAcq's buzz word, I think????) It would eliminate the advantage that loud complainers in high profile teams have over other people running the same hardware within this project, so you would perhaps increase the noise level there, but it would be a HUGE step forward in some sort of real cross-project parity effort.

If the credits are virtually meaningless and have no value at all, then how can they potentially be used as an influence? Either those in charge of the project can be influenced by the heavy weights, or they cannot...I certainly do not see how it can be both.

So this is a broad topic and likely to bring up unforseen facets of the problem

I would like to ask the original poster (PhoneAcq) if the thread he started was meant to only discuss this issue from a SETI perspective, or if we are allowed to discuss this in the broader BOINC sense.

There is some heavy pressure from certain quarters to restrict this to SETI-only issues.

Since this thread was moved here to "Politics" I would naturally assume that it is is now "Off-topic" (not SETI-specific) by definition, as is everything else in this forum.

If it's SETI-specific, why not move it back to the SETI NC forum, away from the off-topic discussions that most users never see?

I see no difference.

...because you are not looking to see a difference. When one does not look, one surely cannot see. You are stubbornly convinced of your belief.

That is a circular argument that I could easily toss back at you. You are looking too hard to find a problem where there is none simply because you are not content with the way it currently is. As such, you are stubbornly convinced that since you see it, any intelligent person would agree with you, and since I disagree, it must be I who is stubborn and refuse to "see". In fact I do see, I fully understand and I disagree.

<cut to end>

Please, spare me the "I just don't understand" arguments. I do, and I disagree with you. I don't think every project has to add SSEx code to be considered "fair".

Additionally, "off-topic" is very different from "disagreed with" or "not liked". If someone brings up a point, but decides that they don't want their point discussed and/or disagreed with in a reasoned manner, it is my opinion that they should probably not bring up the point in the public discussion, reserving it for a private message, or not at all.

Sorry for being so late about replying to this. It's been a hell of a day. (Check Matt's post in Tech News) And that's one of the problems. It's always a hell of a day. Therefore I haven't had a chance to digest the thread and reply to specific points. I'll try to get to some of that tomorrow. I also have a hardware donation announcement/update to make tomorrow.

The loss rate of about 1.5% per month is typical for times between press releases and other news. I take comfort in the fact that we are well above the 160,000 active users we had a year ago. But losing volunteers isn't ever a good thing, especially when we need more.

I don't think we can ascribe the loss to a single cause. Lack of project news certainly plays a part. As much as I try to convince the principle SETI@home people to make regular postings to the forums, Matt is the only one that seems to manage. I, with my post every month or so blog, am in second place. Jeff has posted once or twice to the tech news, as has Bob. Dan has never posted, and David has BOINC related fish to fry. Some of this is just the way we operate. Work till you drop and then work some more in a prone position. If I could just convince them that postings are part of work... Or myself for that matter... (You'll notice I'm doing this on my own time. But then again I do a lot of work for SETI@home on my own time.) The servers will work tomorrow if I don't post. They might not if I do.

Another big issue is the rate at which people replace their computers. If we assume that people replace their computers every 3 years, that works out to 2.8% a month. If nobody ever installed BOINC on a new machine, that's the rate at which we would lose volunteers. And since reinstalling BOINC, and reattaching to your projects, isn't the simplest thing in the world, some people may just not be doing it. We have a system that send emails reminding people if they have dropped off the project, or telling them where to get help if they never were able to return a result, but that was turned off for the donation drive. Matt should be turning it back on soon. We'll see if that makes a difference to our rate of decline.

As for the rate of news releases, I'm hoping to find an interested journalism student to work with us for the summer (at least). Having someone who's job it is to help us get the info and progress reports out to the world would help immensely. The perspective would help as well. Sometimes it's hard to see that what you're working on is interesting when it's your job. We've also got a task list for interested comp. sci., engineering, and physics students. Hopefully we'll find some good ones.

Regarding salaries (the post at the top as I write this), we do earn salaries. None of us are getting rich off of SETI@home. We would all be able to earn more money by quitting and getting jobs in industry. Some of our former employees have done just that. By staying with SETI@home, I manage earn significantly less than former students and post docs of mine. Six people currently pull salaries from the SETI@home grants and donations, but since nobody is paid full time we really have about 3 full time equivalent (FTE) employees. One more FTE would mean 33% more effort put towards the project. We'd like to hire that additional person, but we don't have the money to do it. So we do what we can with what we have and that leave us with little time to make posts telling what we're doing.

I'll try to read some of the posts that are farther down tomorrow and respond to your other concerns.

"I'm trying to maintain a shred of dignity in this world." - Me

i went ahead and just made a donation. maybe it will help get ntpckr up and running. now that i got all that off my chest the past couple of days, i will quietly go back to crunching my workunits.
i will also make quarterly donations in the future.

"I'm trying to maintain a shred of dignity in this world." - Me

Friends....the posts referring specifically to the issues about Boinc and credits have been unhidden which were hidden last night.

All we ask is that people try to stick to the topics involved. No bashing of David Anderson or anyone personally. Stick to the credit issues when discussing him and we should be good. As always the rules to the left when posting still apply in full.

Any concerns, please email the mod list at setimods@ssl.berkeley.edu

Thank You

As I indicated in PM, my Dad considers himself to be right and me wrong in nearly any discussion. You are currently utilizing the discussion tactics of my Dad. I can be actually agreeing with him, but because they are words coming from my lips, surely I am disagreeing, so he disagrees with my agreement for the sake of disagreeing.

The idea of the wrapper and the auto-switching means that if a user with support for FPU-only code comes along, they are given FPU-only code, including any inline optimizations for FPU. If another user comes along that has support for SSE, they are given SSE optimized code to execute. If yet another user comes along that has support for SSE2, they are given SSE2 optimized code to execute.

If one project elects to take this route and highly optimizes their code, but another project elects to only sort-of optimize their code, yet leave a way for other users to gain more performance, this seems to violate the idea you put forth, that if it can be done in the stock app, it should be done in the stock app.

I am actually advocating a complete leveling of the playing field. You, currently, are stating that you disagree with a level playing field, but that you do strongly want things to be fair. You cannot have one without the other.

I think the only real objection one can make to what I am stating is that it will make the cross-project adjustment factor perhaps very complex to figure out.

I suggest you read the following real slowly:

You're taking my words too strictly. "Stock app" can be "stock apps". Whatever is supported by the project. If they want to highly optimize multiple applications and deliver them all at once, each with a specific optimization, then fine, they can release multiple apps, as long as each individual app used gives the appropriate credit/hr. It is the credit/hr that must be the same, no matter the app or apps, for a level playing field. If a faster host can do more work in an hour, then they earn a higher RAC, just as it is now.

This is where I don't see the difference. I don't care if its multiple optimized apps or a single optimized app, or a non-optimized app, just as long as every app grants the same amount of credit/hr.

<snip>

You again, horribly misunderstand my position and every time you attempt to state my side of the argument, you show precisely how far off the mark you are.

As I stated before, you already have a predetermined set of thoughts,

If the credits are virtually meaningless and have no value at all, then how can they potentially be used as an influence? Either those in charge of the project can be influenced by the heavy weights, or they cannot...I certainly do not see how it can be both.

I thought you guys would have figured that out by now.

Just because I claim (and we all virtually agree) that credits mean nothing, doesn't mean people will not try to make them into something more than they are. There is nothing preventing a project who has few users with large amounts of average credit, indicating that they contribute a lot to that project, from allowing those users to have political sway.

Friends....the posts referring specifically to the issues about Boinc and credits have been unhidden which were hidden last night.

All we ask is that people try to stick to the topics involved. No bashing of David Anderson or anyone personally. Stick to the credit issues when discussing him and we should be good. As always the rules to the left when posting still apply in full.

Any concerns, please email the mod list at setimods@ssl.berkeley.edu

Thank You