Pages: [1] 2 3
picantecomputing
BAM!ID: 19319
Joined: 2007-02-14
Posts: 172
Credits: 2,486,623
World-rank: 127,812

2007-02-23 04:35:22
last modified: 2007-02-23 04:37:39

I have to admit, I've only been participating in BOINC projects for a bit less than 2 months now, so I'm somewhat of a noob. But when I look at the BOINC combined stats, as of this posting there are 933,242 total users, only 275,866 of whom are active - less than 30%. Obviously BOINC has been relatively effective in attracting new participants, but it seems to me that 30% is a fairly lousy retention number, given that the cost of taking part is so minimal. If the future growth of DC projects depends upon maximizing participation - not just on acquiring new members who may quit after 10K or 20K credits - then I wonder if perhaps the current credit system isn't effective enough? Again, this idea is totally out of left field and potentially riddled with problems - I understand that. Perhaps, to those more well-versed in the ins and outs of the BOINC community, this is a completely ridiculous and offensive idea. But I'm wondering if a stronger incentive system is needed to help stimulate continued - and accelerated - growth in BOINC participation as a whole.

Let's face it, the current system is based on a competitive, albeit friendly and cooperative atmosphere of participation. There's no getting around that. And for many (it seems about 30% or so), the credit system and other more personal desires to do some good are enough to keep crunching along. But losing 70% of participants in such a worthwhile cause seems outrageously high to me. So what I'm wondering is, has anybody ever thought of bringing either nonprofit or corporate sponsors in to back a limited award-oriented system? I know, it sounds nuts, but hear me out. What if there were several levels of credit "achievement" across BOINC projects, in the combined stats, for which one was rewarded with, just for example, submission into a random raffle for something like a new PC (for higher credit milestones, like a million or so) or something of lesser value (copy of Vista?) for lower credit milestones. There could be lots of other ways of doing this - that was just my first thought. And no, I don't work for Microsoft. There are tons of award possibilities. Those were just the first two I thought of.

And I hate to turn this into a corporate-sponsored thing. I realize BOINC is an advertising-free venue, and I appreciate that as much as anyone else. Really. Preferably something like this would be backed by a nonprofit that is well-aligned with BOINC's overall mission. And again, I'm aware this is fraught with potential problems. I'm really just trying to think outside the lines here a bit and am reacting to the 30%-active number, which seems pretty miserable to me. My only interest here is to spark conversation on how we might make BOINC a more productive and successful overall venture - from an intellectual perspective. It seems to me that so much focus gets put on individual and team credit scores, that maybe sometimes we in general lose sight of the health and effectiveness of BOINC as a united movement of sorts. Are we doing the best we can, as a group? Or are we too concerned with reaching the next individual or team milestone to really notice whether the system as a whole is functioning at maximum capacity? (Don't get me wrong - I love watching my numbers. I'm so totally not bashing that.) But if we could turn the 30% active number into 70% active, how much more would BOINC and its participant projects be able to accomplish? And wouldn't a reward system foster greater participation from new members as well? I realize it might sound a bit silly. I mean, would being eligible for some sort of prize really motivate more folks to crunch? I guess to that I would answer: The simple fact of acquiring ever-higher credit numbers keeps a good portion of us crunching away, so why wouldn't a "real" incentive be even more effective?

I have no idea if this has any validity at all, how it would get coordinated, overseen, or even off the ground. I'm really just throwing it out there to think about how me can make BOINC better. Period. Perhaps this is taking a bit of a marketing approach to what is a scientific endeavor, which may or may not sit well with some. It's just food for thought, so please, no angry responses. I'm just trying to explore an option that may or may not have been considered previously, one which could potentially help to take BOINC to the next level. In any case, I'd be curious to hear your thoughts, positive or negative.

PovAddict
BAM!ID: 115
Joined: 2006-05-10
Posts: 1013
Credits: 5,785,239
World-rank: 76,276

2007-02-23 04:48:43
last modified: 2007-02-23 04:50:09

I didn't carefully read your suggestion, but I want to give some comments. Problem with that is that it would always be the same the ones who win. You probably didn't know, but there are two users on SETI with above a thousand computers. Probably a whole company's building running under the same account. (Source: a paper about distributed computing, will give link tomorrow if I find it). Even discarding those, a user with a single computer has absolutely no chance...

Also, if the prize is a PC, that will give him even more chances for next one. And we're also giving more incentive for cheating.

Just my humble comments (at almost 2 am, maybe I didn't make any sense on what I said ).
picantecomputing
BAM!ID: 19319
Joined: 2007-02-14
Posts: 172
Credits: 2,486,623
World-rank: 127,812

2007-02-23 05:00:11
last modified: 2007-02-23 05:04:14

I didn't carefully read your suggestion, but I want to give some comments. Problem with that is that it would always be the same the ones who win. You probably didn't know, but there are two users on SETI with above a thousand computers. Probably a whole company's building running under the same account. (Source: a paper about distributed computing, will give link tomorrow if I find it). Even discarding those, a user with a single computer has absolutely no chance...

Also, if the prize is a PC, that will give him even more chances for next one. And we're also giving more incentive for cheating.

Just my humble comments (at almost 2 am, maybe I didn't make any sense on what I said ).


I'm sure there would be a way of imposing restrictions. Yes, the folks with more hosts will reach a million (or whatever threshold) faster. But I was thinking it wouldn't be one entry per host, just one per user, and they'd only be able to win once at that level.

And yes, it's true that cheating has occasionally been a problem. But it's not like your chance of winning would be 1 in 10. And maybe the drawing is only held once or twice a year for those who have crossed a particular milestone. So it's not like there would be 200 PCs or whatever given away per year - more like 1 or 2 or 5. And given that there's still significant competition at each milestone level, I think it would still be a rather random system, as far as who wins. I wasn't suggesting that if you go over a milestone, you automatically get the PC. Only that you'd be entered (once, regardless of how many hosts you're running) in a random drawing to win the PC.

And for those with 1 or 2 hosts (I only have 2 myself), achieving eligibility for the bigger drawings would be as difficult as achieving a million credits would be now, so I don't think it's any more anti-competitive than the current credit-only system. The point would be to encourage users to dedicate more resources toward moving to the higher levels, thereby increasing the capacity of the whole system, but I don't think there would be any more of a penalty for having lower numbers than there is now. I mean, sure I'd love to be able to crank out 10,000 RAC, but I don't feel penalized now, nor would I later, for not being able to. I earn the credits I make according to the hardware I can dedicate, but that doesn't make it an unfair system. The goal would be to encourage more long-term participation by ALL users, especially the ones lower down in the stats. If a few giveaways could drastically increase the capacity of BOINC as a whole, wouldn't it be worth it?

Thanks for the comments. Obviously this hasn't all been hashed out yet.
picantecomputing
BAM!ID: 19319
Joined: 2007-02-14
Posts: 172
Credits: 2,486,623
World-rank: 127,812

2007-02-23 06:55:06

Also, if the prize is a PC, that will give him even more chances for next one. And we're also giving more incentive for cheating.


Giving away a PC was just my first obvious thought and is just one of countless possibilities. If it seems an inappropriate choice, I'm sure we could come up with something more fitting. Any ideas, anyone?

Re: the whole cheating issue, on top of the fact that cheating by no means guarantees (or even makes it likely) that you'd win whatever award was available at your credit level, is cheating really so prevalent that it should be a deterrent to undertaking something like this, if the benefits were potentially significant? Perhaps the remote possibility of winning something might encourage the rare participant to attempt to cheat the system, but (1) aren't there checks and balances against cheating currently in the system? and (2) wouldn't the benefits far outweigh the potential drawbacks? There will likely always be a few bad apples (even now, for the sake of obtaining ill-gotten credit), but if such an incentive system could boost BOINC's overall output, is that reason enough to keep the status quo?

As I've said, perhaps this is not the ideal solution. Maybe more to the point is to ask whether it's possible to improve recruiting of new members and retention of existing ones (I would think definitely 'yes' on both), and if so, how do we go about it? There are of course the individual and team stats goals, and on a day-to-day basis, unless you're a project admin, that's probably where most of our attention lies - with the 'little picture.' I'm talking about the big picture and longer-term goals. Wouldn't all projects, current and future, benefit from a more robust system? Is 30% really an acceptable active/retention ratio? Is recruitment of new members really in high gear? With two or three times the capacity, wouldn't BOINC be capable of serving a proportionately greater purpose? From my perspective, BOINC is such a worthy cause and has so much great potential for discovery, why isn't it a household name? Why isn't a much, much larger percentage of the population taking part? And why are new members, who have obviously bought into BOINC and its goals to some degree, dropping out at a 70% rate? Is the status quo really in the best interests of BOINC - and scientific discovery - over the long haul? Just some questions to toss around. I'm sure someone out there can come up with "the" idea (or ideas) to make the most of a brilliant system. It seems pretty clear that we're not anywhere near the full potential of distributed computing. So the first question is, why aren't we? And the second is, how do we get there?
UBT - Halifax--lad
 
BAM!ID: 25
Joined: 2006-02-27
Posts: 366
Credits: 49,272
World-rank: 915,674

2007-02-23 12:10:39

and who provides the prizes? They cost money that projects do not have to spend
Join us in Chat (see the forum) Click the Sig
picantecomputing
BAM!ID: 19319
Joined: 2007-02-14
Posts: 172
Credits: 2,486,623
World-rank: 127,812

2007-02-23 16:57:34

and who provides the prizes? They cost money that projects do not have to spend


As I mentioned in the first post, ideally a nonprofit that is well-aligned with BOINC's goals, or alternatively a corporate sponsor - without turning BOINC into an advertising circus. No one wants that.
UBT - Halifax--lad
 
BAM!ID: 25
Joined: 2006-02-27
Posts: 366
Credits: 49,272
World-rank: 915,674

2007-02-24 09:17:57

and who provides the prizes? They cost money that projects do not have to spend


As I mentioned in the first post, ideally a nonprofit that is well-aligned with BOINC's goals, or alternatively a corporate sponsor - without turning BOINC into an advertising circus. No one wants that.



Can't see anyone been interested at all
Join us in Chat (see the forum) Click the Sig
picantecomputing
BAM!ID: 19319
Joined: 2007-02-14
Posts: 172
Credits: 2,486,623
World-rank: 127,812

2007-02-24 17:29:46
last modified: 2007-02-24 17:30:38

Bueller? Bueller?

[sound of silence]

Very well then...
LJN007
 
BAM!ID: 16525
Joined: 2007-01-07
Posts: 18
Credits: 424,808
World-rank: 330,132

2007-02-25 22:25:32

Bueller? Bueller?

[sound of silence]

Very well then...


I think the main reason to quit is we are all "rated" against all these "1000" computers user.

I had only two 500mhz compaq running SETI so I could not advance in the credit numbers system, now with a new AMD X2 at least I do not fall back in credit ranking
each day.
Maybe a class and or # of hosts per user rating system so you are not comparing users who have access to their companies resources?

Then the single who will never make it off the bottom has some chance of going in an upward direction.



It is to bad to lose so many users because everyone is ranked the same.


picantecomputing
BAM!ID: 19319
Joined: 2007-02-14
Posts: 172
Credits: 2,486,623
World-rank: 127,812

2007-02-26 03:34:32
last modified: 2007-02-26 03:42:05



I think the main reason to quit is we are all "rated" against all these "1000" computers user.

I had only two 500mhz compaq running SETI so I could not advance in the credit numbers system, now with a new AMD X2 at least I do not fall back in credit ranking
each day.
Maybe a class and or # of hosts per user rating system so you are not comparing users who have access to their companies resources?

Then the single who will never make it off the bottom has some chance of going in an upward direction.

It is to bad to lose so many users because everyone is ranked the same.


That's a really good point. I guess it would be an unfair advantage for users with access to corporate facilities to have a greater chance to profit (via some sort of an award). I do definitely see your point, and it is exactly those "smaller" crunchers that I think are abandoning the cause in huge numbers. Perhaps it's intimidating for some to see those with credits in the millions, with huge RAC, in turn causing them to see their own efforts as pointless and not particularly valuable. Which is TOTALLY WRONG, because the power of BOINC is the sum total of its members, large and small - not just the "big guys" at the top. One's contribution to the projects should be judged over the long haul (i.e., years), regardless of your RAC. If newer members could be made to understand that, we might have a better chance of keeping them invested.

So maybe you've hit on something. Maybe it's not the 1,000-host farms at the top that need reward incentives. Maybe, for the health of the whole system, it's the lower-RAC folks farther down that need the incentives. Perhaps it's reinforcing the message to them that their contributions ARE extemely worthwhile - AND appreciated by the entire community. Perhaps BOINC as a whole isn't doing enough to recognize the contributions of its members. For some of us, such recognition might not mean much - but I think that segment of the community is in it for the long haul anyway. Maybe the critical flaw in member retention is not actively encouraging those making smaller - yet still vital - contributions. I'd be interested to hear ideas about how we could make that happen. Not everyone is interested in joining a team, so that's only one small part of the solution.

What about some sort of giveaway for users reaching a significant (but reachable for smaller users) milestone, like 100K? Special recognition somewhere (other than the Shout box)? A "BOINCers do it in front of the computer" T-shirt? Okay, maybe not. But a little something might be enough to keep newer/smaller users interested in and connected to the BOINC community - keeping their resources dedicated to the overall cause. A giveaway would require some sort of sponsorship, but it might not even come to that.

Anyone have any brilliant ideas?
larry1186
 
BAM!ID: 6734
Joined: 2006-09-21
Posts: 92
Credits: 1,209,598
World-rank: 190,680

2007-02-26 16:33:27
last modified: 2007-02-26 16:33:54

Maybe a class and or # of hosts per user rating system so you are not comparing users who have access to their companies resources?


You mean something like a single-host users rank on Rosetta (for example). And this can show any number of hosts per user for any project individually. (I don't think BOINC combined is possible...)

If every user was allowed to climb the ranks, who would they climb over? Somebody has to fall back.

And if there was some sort of "real value" (prizes or awards) for credits, then there would be more incentive for cheaters to do their nasty thing and I don't think it would be worthwhile to try to set up some sort of counterfeit credit detection scheme since that would eat up resources already being spread too thin. And then the cheaters would always find a way around and then it becomes an ongoing war like what you see with virus/anti-virus software today. But some projects provide certificates indicating how many credits you have or how many hours or whatever (SETI and ABC I know of for sure) but that's available to all users regardless of credit.

I like unique T-shirts, and I like your suggestion "BOINCers do it in front of the computer", how about "Do you BOINC?" or "I like to BOINC." or "I'm curing AIDS right now." or "I'm finding a cure for cancer as we speak." or "I BOINC in my sleep." or... the list goes on... but I think these should be incentives to donate. i.e. for every $50 you donate you get a T-shirt or something... reminder: just ideas

@ LJN007: Please put your signature graphics in your signature, not the body of your message, that's why they're called signature graphics.
Don\'t get distracted by shiny objects.
picantecomputing
BAM!ID: 19319
Joined: 2007-02-14
Posts: 172
Credits: 2,486,623
World-rank: 127,812

2007-02-26 20:30:10
last modified: 2007-02-26 20:40:34

You mean something like a single-host users rank on Rosetta (for example). And this can show any number of hosts per user for any project individually. (I don't think BOINC combined is possible...)

...some projects provide certificates indicating how many credits you have or how many hours or whatever (SETI and ABC I know of for sure) but that's available to all users regardless of credit.

I like unique T-shirts, and I like your suggestion "BOINCers do it in front of the computer", how about "Do you BOINC?" or "I like to BOINC." or "I'm curing AIDS right now." or "I'm finding a cure for cancer as we speak." or "I BOINC in my sleep." or... the list goes on... but I think these should be incentives to donate. i.e. for every $50 you donate you get a T-shirt or something... reminder: just ideas


Yes, the stats for this kind of thing already exist. I guess I was thinking of something a bit more active that would reach out and engage users, in an effort to keep those marginal participants more involved in and connected to the community. For instance, BOINCstats (with the appropriate time and financial resources) could establish a Milestones page that would list users who have crossed significant milestones in the past week (perhaps starting at 25K, since a lot of users seem to drop out before that, then 50K, 100K, 250K, 500K, 1M...?). And perhaps at the lower end (maybe 50K), there could be a giveaway T-shirt or something. I know, I know, that costs money - I'm just talking about possibilities. I do also like the idea of giveaways for donations (and I love the T-shirt slogan suggestions). But I'm seeing donations as a separate issue from what I'm talking about, which is again to keep the smaller-RAC users engaged for a longer period of time, thereby making the whole system more productive. Maybe the giveaway is only a one-time thing per user, to keep it reasonable, and the milestones postings are an ongoing thing.

Also, another possibility for keeping new users in the loop and engaged in the community might be increased communication from the projects (or a site like BOINCstats). Perhaps an automated-type email saying 'Great work! You made it to 50K credits.' Or maybe something like the SETI certificate is automatically generated and emailed to users at certain milestones. Of course, it would have to be a feature that could be disabled for those who didn't want that kind of communication. Right now, all communication has to be sought out by users via message boards, etc. Maybe a little bit of acknowledgment in the other direction would help to strengthen one's loyalty to BOINC as a whole. I think recognition (a la the credits system) is key to BOINC's success, and there might be ways to expand on that that haven't been explored yet.

Again, all I'm saying is, we're losing 70% of users somewhere down the line, and many very early on. So first, why? And second, how can we prevent that from happening? All these ideas are directed at that goal - how can we keep more users participating longer.

Oh, and P.S., if BOINCstats were the host something like this (not to be presumptuous), it would give it yet another edge to attract more traffic from the community as a whole. Just a thought.
Honza
BAM!ID: 109
Joined: 2006-05-10
Posts: 154
Credits: 7,658,201,473
World-rank: 456

2007-02-27 06:59:29
last modified: 2007-02-27 06:59:45

All good discussion and nice ideas. Still - the most important remains unanswered...
and who provides the prizes? They cost money that projects do not have to spend

picantecomputing
BAM!ID: 19319
Joined: 2007-02-14
Posts: 172
Credits: 2,486,623
World-rank: 127,812

2007-02-27 07:12:03

and who provides the prizes? They cost money that projects do not have to spend


As I mentioned in the first post, ideally a nonprofit that is well-aligned with BOINC's goals, or alternatively a corporate sponsor - without turning BOINC into an advertising circus. No one wants that.


Admittedly, I don't have anyone specific in mind right now. That was sort of "phase 2" in my mind, after figuring out whether something like this would even be worthwhile.
vorik3
BAM!ID: 962
Joined: 2006-05-29
Posts: 5
Credits: 0
World-rank: 0

2007-03-07 13:32:56
last modified: 2007-03-07 13:34:28

You mean something like a single-host users rank on Rosetta (for example). And this can show any number of hosts per user for any project individually. (I don't think BOINC combined is possible...)


It sure is possible to do overall and using a single computer:
Rank with single computer

I only include people that don't hide their computers on ALL projects they participate in (ie, they could have one computer on most projects, but 100 on another. If they hide their hosts on the project with 100 computers participating, then it would look like they are just using 1 computer, which would be just wrong).
WimTea
 
BOINCstats SOFA member
BAM!ID: 360
Joined: 2006-05-14
Posts: 70
Credits: 48,683,742
World-rank: 17,497

2007-03-10 19:29:08

I thought this ranking with single computer is easily cheatable, though not intentionally, by using 1 comp as UL/DL station for many others. Given, it's not an easy feat to perform but it is doable - and sometimes necessary eg. when an employer allows BOINC to run om may computers but not any networking by it...

I agree with both Honza and others: who gives the prizes and how can (s)he be (or make) sure any ranking is and remains cheatproof as they are very vulnerable already?

Reading the previous again I guess my main point is that credits are themselves the prize as they express an amount of appreciation for your (comp(s)) effort.
picantecomputing
BAM!ID: 19319
Joined: 2007-02-14
Posts: 172
Credits: 2,486,623
World-rank: 127,812

2007-03-11 05:59:29
last modified: 2007-03-11 06:01:16

Reading the previous again I guess my main point is that credits are themselves the prize as they express an amount of appreciation for your (comp(s)) effort.


Point taken - it seems the whole prizes idea was a dud. So my next question would be, what are the current recruitment efforts to bring more new users to participate in BOINC? How do folks learn about it? And could those efforts be expanded to accelerate growth of the whole system? I just can't believe that the BOINC community as a whole is doing everything it can to expand its ranks and maximize capacity. It seems like SOOOO much attention is placed on stats (the individual and the team) that the strength and capacity of the whole system is shortchanged. Is anyone in charge of new member recruitment? Perhaps more importantly, does anyone even care? Or is this all about individual/team achievement? It seems such an obvious oversight that I can't believe more isn't being done.

Myself, around the end of last year happened to remember Folding@Home, which I had participated in briefly a couple of years ago (until it started crashing my PC), so I went looking for it again and found BOINC instead. Originally I heard about F@H via Google Labs. So my initial exposure to DC was completely random, and not due to any sort of "outreach" by the DC community. Not the best environment for wooing new members by the tens of thousands. Is word of mouth really what BOINC depends on for expansion? Seems a bit lame for such a great cause. (BTW, I realize WCG has its own outreach - I'm talking about BOINC as a whole).

So is anyone doing that kind of thing? If so, what are the current efforts? If not, WHY not? Individual stats are fine, but what matters most personally in the long run is how much is getting done by the collective whole, and how much more could be getting done, now and in the future, if BOINC's ranks were doubled or tripled (or more). As I mentioned before (way down in this thread), user retention seems to be a major problem, but currently one without a clear solution. So how about just stepping up new member recruitment instead? I'm trying to look beyond the here and now to ten or twenty years down the road. THINK of what will be possible with a ten- or twenty-fold increase in participation! New membership growth currently seems to be incremental at best, which in my eyes is shortchanging the potential of a brilliant system.

Thoughts? Ideas?
Honza
BAM!ID: 109
Joined: 2006-05-10
Posts: 154
Credits: 7,658,201,473
World-rank: 456

2007-03-11 20:13:31

Yes, there are efforts to get new users...and to make participants stay, which is perhaps even more important.

Back in Czech National Team days, we created t-shirt with BOINC logo and URL, adhesive tapes to place on doors/cars and other exposed areas, we had several meetings in Praha.
Now, there is a 3rd BOINCon next months; organized by members of boinc.cz and boinc.sk. Around 15-20 people from Czech and Slovakia spend a weekend together.
There were several internet and even magazine articles published by active BOINC participant or "jurnalists'. Once a national TV came to my house to do an interview [video feed wasn't used at the end...what a relief :-)] but a spot was shot during prime time. Couple months ago friend and I went doing an interview on internet radio.

Personally I can say yes, there is an effort promoting idea of distributed computing and BOINC in particular.
I agree there is too much attantion put on stats and too small on what science we are doing. But the stats may keep people continue doing their's computer job...
S@NL - FilmFreak
 
BAM!ID: 3062
Joined: 2006-07-11
Posts: 196
Credits: 73,875,476
World-rank: 13,202

2007-03-12 00:42:19

This is an other idea, to get more new participants. You could create a "referrer"-field in the db so users can refer their friends to www.project.com/index.php?referer=1234. When someone sings up through this link (and crunches 1000 credits, for instance) user 1234 gets a referral, so users will compete to get the most referrals.

You could also count the sum of the referrals credits, so referrers will try to keep their refferals active. This credit should be separate from his own credits of course.

Any thoughts/comments?
Racoon: Take your bike and leave the car, going straight to Erics bar.
http://weezepoel.tk
picantecomputing
BAM!ID: 19319
Joined: 2007-02-14
Posts: 172
Credits: 2,486,623
World-rank: 127,812

2007-03-12 02:42:23

This is an other idea, to get more new participants. You could create a "referrer"-field in the db so users can refer their friends to www.project.com/index.php?referer=1234. When someone sings up through this link (and crunches 1000 credits, for instance) user 1234 gets a referral, so users will compete to get the most referrals.

You could also count the sum of the referrals credits, so referrers will try to keep their refferals active. This credit should be separate from his own credits of course.

Any thoughts/comments?


Interesting idea, especially since the web is becoming evermore user-driven. I love the fact that it's expansion driven by the BOINC community (bottom-up) rather than by some BOINC oversight committee (top-down). I think we're all seeing (via YouTube, etc.) that viral - and geometric - growth is an extremely effective "marketing"/spread-the-word tactic, and the very same principle could be applied to expanding BOINC, as you mentioned. We'd still need someone (or some group) with the technical know-how and access to help make it all happen - particularly if it's going to be a cross-project effort. And you'd need buy-in from (and implementation by) project admins as well.

I just happened to think that one positive side effect of that kind of plan, if we were to include referral credits as a trackable/rankable statistic, is that teams could use their collective power not just to expand within the existing BOINC community, but also to acquire new members from outside as well. By using the millions (billions?) of collective connections of BOINCers across the world, it also would allow for a truly global effort executed at the local/regional level through users' respective online communities, according to how each user sees fit to spread the message.

So with all that said, maybe most of the marketing mechanism is already in place (short of tracking referral results) via the existing web. And maybe it's just the emphasis on recruiting new members (and being rewarded for them via referral credits) as a "goal" for BOINCers that is lacking. Another reason I like this idea (I like it more, the more I think about it) is that it uses the incentive of the credits system (already proven to be effective) to boost overall BOINC productivity.

So my next question is, who's got the know-how and influence to make it happen? What would the next step be?
picantecomputing
BAM!ID: 19319
Joined: 2007-02-14
Posts: 172
Credits: 2,486,623
World-rank: 127,812

2007-03-12 03:00:04
last modified: 2007-03-12 03:05:18

In addition to tracking one's own credits and referral credits separately, you could also have a "Total Contribution" category that equals the sum of both. So while a user may have crunched 20K credits him/herself, if they refer 20 other BOINCers to participate, their total contribution could actually be many times greater - might be a nice way to see the true results of one's participation, direct and indirect. As is, I don't really see that there's any incentive whatsoever, outside of individual initiative and dedication to BOINC in general, for anyone to get others involved. This might provide a nice incentive to remedy that.

As a side note, I noticed WCG has a send-to-a-friend feature to encourage referrals (without any tracking mechanism). It might be a good idea to add this to all project pages, as well as any others like BOINCstats. The easier it is for folks to make referrals, the more they'll make, and the more powerful the system.

I really think there's a TON of upside to implementing this idea, and very little downside that I can think of at the moment.
Kevin Erickson
 
BAM!ID: 229
Joined: 2006-05-12
Posts: 61
Credits: 173,806,346
World-rank: 7,323

2007-03-26 19:20:25

Some time ago, i think in '02 or '03, Seti@home (before BOINC) gave out awards for reaching a certain amount of hours of crunching or packets of crunching. You were able to download it and print it off yourself.

I love the Refer list, if this was implemented BOINC wide via new server updates and exported on XML then i think boincstats could pick it up.

As far as funding for prizes:
If there was a site with adds (create income) and some dedicated people(BOINCSTATSwilly?? JK), and people went there to check there stats/weather/news/gas prices/read blogs and clicked on the links then those non greedy people could give away prizes.

LJN007
 
BAM!ID: 16525
Joined: 2007-01-07
Posts: 18
Credits: 424,808
World-rank: 330,132

2007-03-26 19:40:35
last modified: 2007-03-26 19:51:26

Some time ago, i think in '02 or '03, Seti@home (before BOINC) gave out awards for reaching a certain amount of hours of crunching or packets of crunching. You were able to download it and print it off yourself.


Yes you are correct about Seti@home Classic. They started at 250, 500, 1000, 1500 and up. By starting so low the 500MHZ user had a chance to get one. Maybe the Certificate levels could be set to the number and type of Hosts you have active online. So the 50 host guys would have the same "slope" as the single host a "Xn" value is set for each computer speed to set the min. level for the !st ,2nd - etc award certificate.

I know my 5000 in Seti classic was a hard goal on a 500mhz compaq

If fact the project sent an email with the link to get the award

Dear Chewie @ Research & Stats,

Congratulations on recently completing your 500th SETI@home workunit. Through your support, SETI@home has grown to become the largest distributed computation on Earth. We at SETI@home greatly appreciate the 1.41 years of computer time you have donated to the project, and hope that you will continue your support. As a small token of our appreciation, you may now download and print an official certificate stating your achievement of this goal, available at:

http://setiathome.ssl.berkeley.edu/cgi-bin/cgi?email=snc@atlantic.net&cmd=print_cert&certnum=500&size=0

Thank you again for your contribution.

Sincerely,

The SETI@home Team


SETI@home member since 14 Nov 2001
Total credit 39,749.20
Recent average credit 201.72
SETI@home classic workunits 5,906
SETI@home classic CPU time 60,377 hours
Kevin Erickson
 
BAM!ID: 229
Joined: 2006-05-12
Posts: 61
Credits: 173,806,346
World-rank: 7,323

2007-03-26 22:59:51

I found my "Certificate of Computation" and it is from Feb 15 2005 (More recent than i thought). It came in 3 different types and I downloaded all 3.

S@NL - FilmFreak
 
BAM!ID: 3062
Joined: 2006-07-11
Posts: 196
Credits: 73,875,476
World-rank: 13,202

2007-03-29 21:15:32

So my next question is, who's got the know-how and influence to make it happen? What would the next step be?

I sent it to the developers mailing list. And this is David Anderson's reaction:
That's a great idea; let's do it.

There are still some problems on how to implement it, but they seem to be working on it.
Racoon: Take your bike and leave the car, going straight to Erics bar.
http://weezepoel.tk
Pages: [1] 2 3

Index :: BOINC :: This is a totally random idea...
Reason: