[Discussion] 4th Generation BOINC credit system

Message boards : Questions and problems : [Discussion] 4th Generation BOINC credit system
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 · Next

AuthorMessage
Profile pschoefer

Send message
Joined: 5 Aug 06
Posts: 48
Germany
Message 69684 - Posted: 21 May 2016, 13:36:42 UTC - in response to Message 69634.  

I think that there is a way to introduce benchmark jobs that use real data and real applications but first I need to find a way to untangle all of the CreditNew code so it can be replaced at all.

That's how Rectilinear Crossing Number fixed their problem with benchmark cheating some years ago. Not perfect, but about the best thing that can be done for variable-length tasks with no good (known) metric for the work done.

But still, it should be made as easy as possible for project administrators to use their own system instead (which might be as simple as n credits per task for fixed-length tasks).
ID: 69684 · Report as offensive
Chris S
Avatar

Send message
Joined: 13 Jun 10
Posts: 2797
United Kingdom
Message 69694 - Posted: 22 May 2016, 7:58:02 UTC

I hesitate to make any comment here for a number of reasons, but it might be useful to put the view from a simple end user and cruncher.

As I understood it, the rationale behind Boinc was to build upon the original Seti@Home Classic which had proved that distributed computing using the computers of the general public was a viable proposition, and an alternative to hiring Super Computer time. It was initially envisaged that the projects would only utilise unused processor cycles in the background, while the computer was not being used for its main task. Rather a naive notion I would have thought, and probably mainly used as a marketing tool to sell the DC concept to the masses.

It was also envisaged that every project under the Boinc umbrella would offer equal rewards in terms of notional credits for the time and cost of crunching. It was all based upon people choosing a project for its individual scientific worth. All fine on paper, but of course it didn't take into account human nature.

Some people did indeed choose their projects based upon their perceived scientific worth, but others were just more interested in being at the top of a leader board somewhere and didn't care what they crunched or where. They hopped around from project to project looking for the best return. Some managed to inflate the credit they got which was pure cheating. Project managers soon found that if they offered 3 or 4 times more credit than others did, then they got more users and more work done.

But I find the RAC figure very useful although I may be using it in the wrong way. Generally speaking the RAC on a particular machine stays fairly level over time, but of course varies if the project has no work or there is a glut of shorties or VLAR's. A regularly falling RAC tells me something is wrong somewhere, has the machine fallen over, has the GPU fan seized, has it lost it's network connection, is it stuck on a single task for days on end etc. A quick check directly or remotely, usually shows the source of the problem.

The other use of RAC is to compare your machine with others of the same specs on the same project. If their RAC is way above yours how have they done that. Is your machine not optimised as well as theirs, is it a driver version problem, config files, enhanced apps etc. You might as well get the best performance for the project you choose to crunch for.

Lastly people like me will never ever get to the top of the leader boards and I couldn't afford the kit to try, but I see nothing wrong with a little friendly competition within teams, provided it isn't taken too seriously. It's all come a long way from unused CPU cycles to dedicated crunching farms running 24/7 with dozens of GPU's, Growing like Topsy comes to mind but without any overall control that I can see. Even programmers on the same project disagree on how to do things! And of course it stretches the favoured projects supporting infrastructure to new limits.

As an end user I don't micromanage my kit, I just need a simple performance indicator to monitor for problems and to ensure that I get the best out of my set up, such as it is. At present RAC supported by credit new does that for me, although I am aware of BoincView and Account Manager etc. Whatever credit system is introduced it will not please everybody anyway, and lets be honest here, how many of the current 3.5 million active Boinc crunchers really care what type of credit system there is, or even if we have one at all. Is it just for the hard core vociferous minority?
ID: 69694 · Report as offensive
Egilman
Avatar

Send message
Joined: 27 Jan 16
Posts: 16
Message 69835 - Posted: 31 May 2016, 9:58:49 UTC - in response to Message 69694.  

Actually, the first cheating was discovered on Seti I. Those text based WU's were easy to scam in several different ways.

There were people all over the world amassing huge totals of credit for doing nothing.
I know this cause I was one of the six people on Ars Technica that helped expose it. That was before Boinc was ever put into operation.

Dr Anderson told us that Boinc would solve such problems but obviously it hasn't.

Now, with Gridcoin, your adding a financial incentive to cheat to the mix and want to change the whole credit paradigm to solve that for one specific groups advantage.

I say eliminate Gridcoin from boinc. That removes the financial incentive to cheat which will only make the problem worse. At the very least eliminate paying for crunch with bitcoin in any way.

This is how Gridcoin has grown to be the largest boinc team is by greed.
and yeah I am a member on the Gridcoin site and have seen how the creaters of gridcoin have the full intent to take over boinc and use it to their own ends.
First by becoming the dominant team in Bionc, and them monetizing boinc.

And lets not forget the profit in bitcoins of any type are based upon value/demand. More demand higher value. By completely monetizing Boinc, they create huge demand for their currency thereby increasing the value of the bitcoin they already hold. sure they pay a few sheckles out to the crunchers but that is far far far less than the value/demand they are creating. Which makes the gridcoin creators even more wealthy.

So answer me a question?

WHY would Boinc want to get involved directly with making a few people wealthy beyond the dreams of avarice when the original intent was to volunteer for science?

I'm also the world Number 1 cruncher in Collatz and Moo. My computers are not hidden and anyone can check the WU's and their crunch at any time.

Cheating has been going on since the moment distributing computing was formed. And will continue despite your best efforts to stop it. and that is simply to stay on top of a leaderboard. Much less being paid to be on top of that board.

Greed has driven Gridcoin to the great recruiting levels they have attained. greed is driving this initiative to adjust credit. and the only ones that benefit is the founders of gridcoin.

Everyone else will lose.

I say leave it alone, and eliminate pay for production from boinc. Remove greed as an incentive for crunching.

The scientific community will love ya for it as will most of the organized teams out there.
ID: 69835 · Report as offensive
Profile CM

Send message
Joined: 13 Aug 15
Posts: 63
Message 69837 - Posted: 31 May 2016, 11:34:59 UTC - in response to Message 69835.  


Now, with Gridcoin, your adding a financial incentive to cheat to the mix and want to change the whole credit paradigm to solve that for one specific groups advantage.

I say eliminate Gridcoin from boinc. That removes the financial incentive to cheat which will only make the problem worse. At the very least eliminate paying for crunch with bitcoin in any way.

Cheating was already present due to inner/inter team competition, do you also propose to ban charityengine and any team that offers prizes for BOINC computation?

Improving the credit system does not solve issues that only affect a single group - cheating affects all BOINC users/teams. If a new credit system was devised, it wouldn't be forced upon all projects - projects can chose what credit system to implement (there is no centralized control over BOINC projects), and several may not upgrade simply due to project inactivity or may be unwilling to upgrade due to perceiving it as unnecessary server maintenance.

Due to a lack of centralized power, you cannot eliminate Gridcoin from BOINC. You could campaign for each BOINC project admin to ban/delete team 'Gridcoin', but it's highly unlikely that many project owners will intentionally drive computing power away from their projects just to entertain a single troll.


This is how Gridcoin has grown to be the largest boinc team is by greed.
and yeah I am a member on the Gridcoin site and have seen how the creaters of gridcoin have the full intent to take over boinc and use it to their own ends.
First by becoming the dominant team in Bionc, and them monetizing boinc.

Your misinformation is based on a single cherry-picked forum post; I've already covered this topic: https://boinc.berkeley.edu/dev/forum_thread.php?id=10794&postid=68948 (Note that the current poll shows a majority of users wish to vote on the mandatory team requirement)

If BOINC was to transition to a 4th gen credit system or if we can work out how to eliminate cheating, the team requirement removal would only be a matter of managing a scalable solution on the blockchain (increasing neural network registration from 2k to 100k+) instead of for providing security from cheaters. Thusly, Gridcoin would no longer be a one team crypto, but rather a crypto for the entire BOINC community.


And lets not forget the profit in bitcoins of any type are based upon value/demand. More demand higher value. By completely monetizing Boinc, they create huge demand for their currency thereby increasing the value of the bitcoin they already hold. sure they pay a few sheckles out to the crunchers but that is far far far less than the value/demand they are creating. Which makes the gridcoin creators even more wealthy.
[...]
Greed has driven Gridcoin to the great recruiting levels they have attained. greed is driving this initiative to adjust credit. and the only ones that benefit is the founders of gridcoin.

I hold no Bitcoin; it is only used as a trading pair on exchanges - bitcoin is not utilised at all within the gridcoin system.

I don't think wishing to offset some of the electricity costs racked up by BOINC computation is greedy, and the founders of gridcoin are not the only ones to benefit from it continuing to exist (There's no supply cap, in 10+ years the current supply will be a minority compared to the sum of newly generated coins. Not to mention the fact that there was no instant/pre-mine; all existing GRC represents past BOINC computation).


So answer me a question?

WHY would Boinc want to get involved directly with making a few people wealthy beyond the dreams of avarice when the original intent was to volunteer for science?

BOINC is not involved in neither Gridcoin's development nor community (as far as I'm aware, there may be several anon BOINC devs in our community).

The BOINC platform/software is not dedicated solely to science, nor is it mandatory that projects must have a scientific purpose; there are several non-scientific BOINC projects currently - do you oppose their existence too?

Regarding people becoming wealthy 'beyond the dreams of avarice', thats rather far fetched considering Gridcoin currently is worth (1/4030=0.025%) the value of Bitcoin.


Cheating has been going on since the moment distributing computing was formed. And will continue despite your best efforts to stop it. and that is simply to stay on top of a leaderboard. Much less being paid to be on top of that board.

Are you justifying the existence of cheating purely based on its presence in the past? Your second sentence reads like a threat, are you upset that you or your team members will no longer be able to fraudulently claim a high leaderboard position?

Surely if we can identify solutions to cheating, their implementation would be beneficial to the entire BOINC community? Whilst it may be a cat&mouse game, why should we not oppose cheaters? Do you take the same view on computer security updates (considering it's a similar never-ending cat&mouse game).


I say leave it alone, and eliminate pay for production from boinc. Remove greed as an incentive for crunching.

The scientific community will love ya for it as will most of the organized teams out there.

As I've stated above, censorship/oppression of communities based on their purpose is unlikely to occur.

Please stay on topic of the 4th gen credit system instead of derailing the topic towards gridcoin.

To those reading: Sorry for having to reply to this post (went off topic myself).
ID: 69837 · Report as offensive
Egilman
Avatar

Send message
Joined: 27 Jan 16
Posts: 16
Message 69848 - Posted: 31 May 2016, 17:12:44 UTC - in response to Message 69837.  

When you need to insult/abuse your critics to defend your position from serious criticism.

Usually there is something funny (crooked) going on in the background.

Let me ask a question. if this is so great, why not take the open source boinc platform and devise you own crunching system and leave boinc out of it?

Why the need to take over the platform?

And yes there are programmers what will cheat any system you can devise. I mean Moore's law? you obviously have some smarts.

Windows ring a bell? they have been trying to lock windows down since they first wrote it, but every time they release a new version, the next day a hacked version is available on the undernet. The millions they pay for security and it's hacked before it is even released.

There is no absolutely secure boinc. A pipe dream all your trying to do is get boinc dev to work with you in your money making scheme.

You will lose the boinc community and you will never get them back

More importantly you will lose the reasearchers, as they will take the open source platform and develop their own.

Does any one forget what happened to the medical program when the rumor hit that they were for profit? it took them years to get their reputation and crunchers back.

You guys can be greedy all you want. you can also dupe the community into supporting you.

But just remember, if you really were altruistic and just trying to get the little guy some power deferment, why cite the current value of gridcoin? and it's potential?

You guys are in it for the bucks.

Your whole response and each insult nails that home.
ID: 69848 · Report as offensive
Vortac

Send message
Joined: 31 May 16
Posts: 3
Croatia
Message 69857 - Posted: 31 May 2016, 21:19:00 UTC

Egilman, you are right - fully commercial distributed computing networks are already in development, but not by Gridcoin. Most likely they will be based on Ethereum, whose market value (as we write this) is 500 times bigger than Gridcoin and still growing. So, your worries are indeed justified - people are in it for bucks and tomorrow they will happily switch to the network which pays more. It's just not happening here. You are giving altruism lessons to the wrong people.
ID: 69857 · Report as offensive
Profile Ageless
Volunteer moderator
Project administrator
Volunteer tester
Help desk expert
Avatar

Send message
Joined: 29 Aug 05
Posts: 12104
Netherlands
Message 69858 - Posted: 31 May 2016, 21:19:56 UTC

These forums are for discussion of anything that goes about BOINC. No discussion is ever set in stone about its direction, so if the direction deviates somewhat to the left or right, fine. Eventually it'll always comes back on topic.

If you do not like what others say in the discussion, or if someone's opinion differs so much from yours, you don't go calling them a troll. If you truly cannot get over what they say, put them on ignore. The option is available in the forum preferences. That way their posts are hidden from your view, you don't have to worry about what they say anymore, and you don't have to go call them names for absolutely no reason.
Jord
Please do not private message me for tech support. Use the forums for that. Tech PMs will be ignored.
ID: 69858 · Report as offensive
Profile CM

Send message
Joined: 13 Aug 15
Posts: 63
Message 69867 - Posted: 31 May 2016, 22:51:09 UTC - in response to Message 69858.  

These forums are for discussion of anything that goes about BOINC. No discussion is ever set in stone about its direction, so if the direction deviates somewhat to the left or right, fine. Eventually it'll always comes back on topic.

If you do not like what others say in the discussion, or if someone's opinion differs so much from yours, you don't go calling them a troll. If you truly cannot get over what they say, put them on ignore. The option is available in the forum preferences. That way their posts are hidden from your view, you don't have to worry about what they say anymore, and you don't have to go call them names for absolutely no reason.

Cool, permission to maintain going off topic granted! It's somewhat fair to categorize Egilman's posts as trolling though considering he doesn't acknowledge answers to his questions, but fair enough after this post I'll cease namecalling, it was quite crass of me.

When you need to insult/abuse your critics to defend your position from serious criticism.

Usually there is something funny (crooked) going on in the background.

What insult? Cheekily implying that you're a cheater for defending cheating? Or implying that failing to acknowledge previous answers/responses to your questions/posts (in this thread and the 'monetization' thread) somewhat constitutes trolling?


Let me ask a question. if this is so great, why not take the open source boinc platform and devise you own crunching system and leave boinc out of it?

Why the need to take over the platform?

What specificity are you referring to by 'this'? The general concept of a 4th gen credit system?

According to the BOINC project governance model, decisions are based on community consensus & and disputes are decided upon by the PMC. https://boinc.berkeley.edu/trac/wiki/ProjectGovernance

Simply put there is no reason to fork BOINC at this point (nor should there be), the concept of a 4th gen credit system has not been defined never-mind there being a community breaking lack of consensus on the matter.

Your opposition to further development/improvement of the credit system has been noted & should be taken into account in the future (to maintain fair community consensus). How community consensus is measured would be an interesting side-topic.


And yes there are programmers what will cheat any system you can devise. I mean Moore's law? you obviously have some smarts.

Windows ring a bell? they have been trying to lock windows down since they first wrote it, but every time they release a new version, the next day a hacked version is available on the undernet. The millions they pay for security and it's hacked before it is even released.

Your analogy of Moore's law to cheating a credit system makes little sense, computers are getting more powerful every year thus BOINC cheating will always occur? Poor analogy.

Windows is one of the most successful brands of operating systems on the market. Of course they have been fighting a continuous battle against vulnerabilities in their operating systems (all popular software is subject to continuous attack). Had Microsoft never patched any of the publicly exposed vulnerabilities their operating systems would not nearly have the same market share they currently hold & Microsoft would likely not be the same company it is today.

Simply because new vulnerabilities/issues may present themselves in the future doesn't mean that existing vulnerabilities/issues (that are public knowledge) should be ignored/maintained. Stagnation is never the solution to security flaws.


There is no absolutely secure boinc. A pipe dream all your trying to do is get boinc dev to work with you in your money making scheme.

You're right that there is no absolutely secure BOINC - there will always exist flaws in software (past/present/future) that will be taken advantage of by malicious actors, but using that as an excuse not to take positive action against their existence doesn't make sense.

What's wrong with interacting with the BOINC community/developers? Do you wish for this to not be an open community? Many of the users that have expressed their views in this thread have zero affiliation with Gridcoin yet share somewhat similar views that a 4th gen credit system could be a good idea (we have yet to reach consensus on it's definition).


You will lose the boinc community and you will never get them back

What makes you say this? At this point, our team/software has yet to reach the majority of the BOINC community.


More importantly you will lose the reasearchers, as they will take the open source platform and develop their own.

I thought that you wished for us to develop our own fork of BOINC, now you wish for researchers themselves to fork their own BOINC platform?

The vast majority of BOINC users don't care about the credit system, only a small percentage do, so I doubt that a mass exodus would occur. If it did, it would be great publicity for BOINC & volunteer computing.


Does any one forget what happened to the medical program when the rumor hit that they were for profit? it took them years to get their reputation and crunchers back.

Got a link to a thread on this? What project was this? Do you oppose any company that runs a BOINC project that isn't expressly non-profit? IBM is a for-profit company yet their WCG project is one of the most popular BOINC projects.


But just remember, if you really were altruistic and just trying to get the little guy some power deferment, why cite the current value of gridcoin? and it's potential?

I cited the value of Gridcoin as a response to your post where you implied that Gridcoin was "making a few people wealthy beyond the dreams of avarice"; Avarice - one of the '7 deadly sins' being incapable of dreaming about a marketcap 0.025% of Bitcoin's marketcap is a bit far fetched.


You guys are in it for the bucks.

Your whole response and each insult nails that home.

My views/posts are that of my own and do not represent the entire Gridcoin community.
ID: 69867 · Report as offensive
Egilman
Avatar

Send message
Joined: 27 Jan 16
Posts: 16
Message 69899 - Posted: 2 Jun 2016, 5:11:39 UTC - in response to Message 69857.  
Last modified: 2 Jun 2016, 5:16:27 UTC

Egilman, you are right - fully commercial distributed computing networks are already in development, but not by Gridcoin. Most likely they will be based on Ethereum, whose market value (as we write this) is 500 times bigger than Gridcoin and still growing. So, your worries are indeed justified - people are in it for bucks and tomorrow they will happily switch to the network which pays more. It's just not happening here. You are giving altruism lessons to the wrong people.


Ethereum has a long way to go yet, but looks to be the future of the cloud, computing in general and Microsoft is betting on it.

But Distributed Computing as we know it is probably a decade away before it flips to Ethereum.

And I know I'm talking to a wall. But it get so frustrating to see everything good in the world going down the money rathole.

Gets frustrating. WE hated exposing the cheats in Seti I we were afraid of what it would show on our own team, Thankfully none (except for the proof of concept accounts we set up for verification)

And it was Rosetta that took the hit for the rumor that they were going for profit. (I was working at the UW when it happened)

There is always going to be those that look for the angle. I just wish it wouldn't happen here, Individual contribution to science is dying out to the corporate/government sponsored and financed conglomerate.

One of the last bastions of freedom.
ID: 69899 · Report as offensive
Vortac

Send message
Joined: 31 May 16
Posts: 3
Croatia
Message 69902 - Posted: 2 Jun 2016, 11:09:23 UTC
Last modified: 2 Jun 2016, 11:10:15 UTC

You gain some, you lose some. Fully commercial distributed computing networks will certainly be profit-driven, corporate-controlled and non-independent. But they will attract more people to the idea of distributed computing, hopefully providing scientists with more computing power in the long run.

In the end, I think the best idea is to let the users decide if they want to donate their CPU time completely for free (pure BOINC) or to join something like the Golem project and let the Big Company run unknown, proprietary code on their machines, with little or no control over the process, but also getting paid for it. Or join Gridcoin and get a token reward, but also lose some of the BOINC freedoms (cannot join another team, cannot get rewarded for every BOINC project etc). Or join Folding@home and get rewarded in Curecoin or Foldingcoin. Or join Sia and rent your hard-drive for Siacoins. There are endless possibilites and the more powerful machine you have, more options will be available to you. Let the users choose.
ID: 69902 · Report as offensive
Richard Haselgrove
Volunteer moderator
Volunteer tester
Help desk expert

Send message
Joined: 5 Oct 06
Posts: 2424
United Kingdom
Message 69903 - Posted: 2 Jun 2016, 11:23:58 UTC - in response to Message 69902.  

I suspect the project scientists who commission the research in the first place will do the choosing. I've suspected for a while that many of them would prefer the more controlled environment of a commercial platform, rather than the Wild West of BOINC with all these message boards to worry about. Witness the way SIMAP was at the forefront of BOINC for Android with their HTC 'Power to Give' collaboration - but very rapidly closed down their entire BOINC operation, soon afterwards.
ID: 69903 · Report as offensive
Vortac

Send message
Joined: 31 May 16
Posts: 3
Croatia
Message 69906 - Posted: 2 Jun 2016, 14:49:47 UTC

Well, it's certainly possible BOINC will diminish when fully commercial and more controlled platforms become available. As Egilman said, people are in it for bucks. And scientists too, we might add.

Anyway, in this context, Gridcoin could actually strengthen BOINC. I guess it would be possible to acquire the computational power of the Gridcoin team quite quickly with some Gridcoins and for a scientist with a limited budget that would certainly be more affordable than Amazon EC2 or Microsoft Azure (which are still quite expensive). It hasn't been done yet, but it's feasible enough, although still clunky and unstraightforward for someone who isn't familiar with cryptocurrencies and/or BOINC. Golem project looks far more advanced on paper, but it's still in early development, while BOINC is here and working.
ID: 69906 · Report as offensive
SMH17

Send message
Joined: 14 Jun 16
Posts: 1
Italy
Message 70183 - Posted: 14 Jun 2016, 14:07:23 UTC

Paying the research with a cryptocurrency as Gridcoin can only help the BOINC purpose. If a day the value of this cryptocurrency will grow certainly will attract greedy and bad people that are only profit driven, but in any scenario will drive on BOINC projects an huge amount of additional computational power to help science progress in the same way the avarice moved tons of computational power wasted in Bitcoin and others cryptocurrencies mining.
It's a win-win, so make no sense claim such ban, prevent all possible cheating attempts is needed in any case.
ID: 70183 · Report as offensive
zero

Send message
Joined: 26 Jul 16
Posts: 1
United States
Message 71018 - Posted: 26 Jul 2016, 2:47:44 UTC
Last modified: 26 Jul 2016, 2:56:45 UTC

I'm not sure if this is a dead discussion or not, so I apologize, but I feel the need to post some input.

I have just discovered BOINC through Gridcoin, and it's a shame I hadn't discovered it earlier, but there are reasons for that. When BOINC was first introduced, I was in middle school. Looking at Google trends, the publicity of BOINC has declined since 2005, and not being a part of the scientific community, I don't know when I would have encountered it. Now that I've discovered it, I think it's one of the greatest things I could contribute to. It's a centralized group of causes that I can get behind. I have a gaming rig and a work computer, and I finally feel like they are living their potential.

And because of Gridcoin, it's not just a win- it's a win-win. It helps me justify my contribution, when I am not really in a position to be donating money or time at this point in my life. The fact is, I hope that the payoff is at least breakeven, but I'm okay with losing money because it aligns with my core beliefs. And the reality is, I'm sure this will be the case with most of the participants on the Gridcoin team. The average person isn't in the position to justify running up their electric bill or taking the time to monitor their computer cycles. Especially if they aren't already scientifically or technically inclined. But that doesn't mean their contribution isn't helpful or needed. And it doesn't mean that they wouldn't be willing or enthusiastic about being a part of the greater BOINC community, if given a minimal incentive.

Egilman, I completely understand your sentiment. You have been doing for a long time here what you are seeing some people come in and do in the hopes of making money. It seems to taint the feeling of community around contributing selflessly to projects that may change the world. I know that there are definitely people with the wrong, and even malevolent intentions that are and will be associated with Gridcoin, and they may not have entered this space without it. But this quote struck me:

Individual contribution to science is dying out to the corporate/government sponsored and financed conglomerate.


I think this is an accurate statement, but even further I don't think the average individual has ever had a chance to meaningfully contribute to a greater scientific effort. Again, most people don't have the time or resources. It's an unfortunate fact that science takes money. This need perverts at every turn: http://www.scientificamerican.com/article/dr-no-money/

The Gridcoin project especially feels to me like an opportunity to fight against exactly this issue. To me at least, the project's priorities are aligned with BOINC completely. At the very grandest hopes for Gridcoin, one would dream of a currency by science, for science. Individuals creating value through computer cycles, and users like yourself can compound their contributions, by also contributing the earned currency. An incentive system to help galvanize a population around contribution to science, whether you have a room of servers or an android phone.

As you point out, this brings all kinds of negative people to the table, but that is inherent in expanding the scope to the average person. Really, every corner of the scientific community could go a long way by reaching the average person. And the fight to combat the effects of a negative influence is a worthwhile one, because the value is twofold: increase in reach, and improving the product and message.

Which finally brings me back on topic- this credit system does already have value. Though this is the first time I've ever "mined," I have a long history of dealing with virtual points and currencies. If someone is trying to game the system to get more of it, you know it has value. It may only be in exposure or in status, but it's valuable. They are spending time and effort purely to gain these numbers.

Which means that it is worth protecting, if you are going to have it as part of your community. Even a purely ornamental measure is important to the people it influences, if only in the subconscious. Not to mention that anyone who has run any tests or dealt with analytics knows the compounding effects of bad data.

I think the effort to prevent cheating in the BOINC credit system is a worthy one, and happens to align directly with the priority of the Gridcoin project.

I don't have enough technical knowledge to contribute to task verification for granting credits, but if one of the goals of this discussions is creating a system that is comparable cross-project, then why not create a derivative metric off of the standardized cycle-driven credit system? Gridcoin is doing it in one way, but if the goal is pure comparison, there are plenty of options depending on how you want the output to look. You could have a variable logarithmic scale normalized to each project. I just don't think the answer has to be the one true credit. I always see a value in a base standard, with derivatives for comparison.

Just my .02; glad to be a part of the community here.
ID: 71018 · Report as offensive
Profile CM

Send message
Joined: 13 Aug 15
Posts: 63
Message 71065 - Posted: 28 Jul 2016, 15:37:16 UTC
Last modified: 28 Jul 2016, 15:37:37 UTC

To anyone interested, the Gridcoin community is holding a Gridcoin Hangout on BeyondBitcoin's mumble server on Saturday 30th. The 4th generation BOINC credit system will be one of the topics discussed, I'd really appreciate it if BOINC community members could attend. Cheers :)

https://steemit.com/beyondbitcoin/@cm-steem/gridcoin-hangout-001-rsvp-and-suggest-topics
ID: 71065 · Report as offensive
ChristianB
Volunteer moderator
Volunteer developer
Volunteer tester

Send message
Joined: 4 Jul 12
Posts: 305
Germany
Message 71133 - Posted: 2 Aug 2016, 7:51:23 UTC

I couldn't make it to the Hangout. Are there some minutes from the discussion?
ID: 71133 · Report as offensive
Profile CM

Send message
Joined: 13 Aug 15
Posts: 63
Message 71174 - Posted: 3 Aug 2016, 22:19:24 UTC - in response to Message 71133.  
Last modified: 3 Aug 2016, 23:17:17 UTC

I couldn't make it to the Hangout. Are there some minutes from the discussion?

The hangout has been uploaded to soundcloud: https://soundcloud.com/gridcoin-community-hangouts/gridcoin-hangout-001

The 4th gen BOINC credit system was the first topic of discussion (First 20 mins), basically summing up the topic for our community.

Here's a list of topics discussed: https://steemit.com/gridcoin/@erkan/topic-voting-for-gridcoin-hangout-001#@erkan/re-erkan-topic-voting-for-gridcoin-hangout-001-20160729t114852991z

The highest voted topics were prioritised in the hangout, so that's a guide to what was discussed.

I'll schedule another hangout and keep you noted.

Edit: A second hangout has now been scheduled: https://steemit.com/gridcoin/@cm-steem/gridcoin-hangout-002-rsvp-and-suggest-topics
ID: 71174 · Report as offensive
Profile CM

Send message
Joined: 13 Aug 15
Posts: 63
Message 71591 - Posted: 15 Aug 2016, 23:09:43 UTC
Last modified: 15 Aug 2016, 23:09:56 UTC

Posted a link to this thread to the boinc_dev mailinglist & several developers have begun discussing this topic:
http://lists.ssl.berkeley.edu/pipermail/boinc_dev/2016-August/022235.html
ID: 71591 · Report as offensive
Profile Ageless
Volunteer moderator
Project administrator
Volunteer tester
Help desk expert
Avatar

Send message
Joined: 29 Aug 05
Posts: 12104
Netherlands
Message 74283 - Posted: 21 Nov 2016, 21:41:12 UTC

Just giving this thread a bump up, so it can stay open and I don't have to reopen it again tomorrow.
ID: 74283 · Report as offensive
Profile CM

Send message
Joined: 13 Aug 15
Posts: 63
Message 74284 - Posted: 21 Nov 2016, 22:04:10 UTC

*bump*

Has anyone had any further thoughts regarding a possible 4th gen credit system?

We've discussed it a bit in the Gridcoin community hangouts.
ID: 74284 · Report as offensive
Previous · 1 · 2 · 3 · 4 · 5 · Next

Message boards : Questions and problems : [Discussion] 4th Generation BOINC credit system

Copyright © 2017 University of California. Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.