| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: Suggestions / Feedback
Thread: Reward |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 8
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
One problem i have always had with this grid computing, is that i never know if my computer found anything useful. I know everything it finds is useful but i mean something that will be of greater importance then most. With that i would like to suggest to possibilities of "rewards" to peoples whose computer discovers something ground breaking or revolutionary. They could be of prize or even their name posted on the main page with the basic problem or ground breaking development that the user's computer solved. Also, if the person does not want his or her name displayed then they can choose.
What i am basically saying is that i would love to know if my computer finds something grounds breaking and if so, what is it. Everyone should know what i mean by ground breaking. i know this is for the greater good and all info is the best info but i mean something revolutionary. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
hmmmmmm ... tough question ... here is a back of the envelope calculation:
lets say iit takes 40 work units to predict the structure of a protein. Roughly half the proteins we fold fail to converge to a prediction that we trust (sometimes the simulations get stuck or just don't work for a host of reasons). Ok so this is rough, but with every 80 work units we get an answer for a protein. So I would say joining the grid for a week to a month and you should have produced at least one "good" result ... one usable structure prediction. we've done a fair bit of pre-computing to just send out the proteins most likely to result in usable results. That's part of why we can expect such yields. |
|||
|
||||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I think the basic idea is to 'tag' each result with the id of the user, and then on the other end have analysts mark or comment on results that were particularly useful. Obviously this shouldn't be something which bogs down the researchers, but it wouldn't need to be if only exceptional results were identified.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Usually its not any single result/person, but the outcome of many, many related or similar "interesting" candidate molecules that directs the research labs to focus effort in that area. From there the more precise computer models or real labwork can be done.
Keep in mind that the computer simulation models being used for most of the computer world are merely approximations of the real thing and are inherently inaccurate (usually because the precise human understanding of the natural processes are not yet exact). Sometimes multiple different computer models need to be done with the same molecules just because different programs are inaccurate in different ways. The results of computer simulations help researchers to know that there may be something of interest to look at must still be done manually in real experiments. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
But still, a simple email saying something like "Congratulations! A work unit sent by <machinename> on <date> has been deemed valuable for further research."
Something that keeps us in the loop. Instead of just leeching free CPU cycles off. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Screw that. I want my name attached to the protein. My screen name. NoEchonium... or DiRiboNoEchonius... you all get the idea.
Dig it.  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
screw that... just post it for all to see... u know... <machine name> provided totally useless information!!!
|
||
|
|
joatmon
Senior Cruncher Joined: Nov 17, 2004 Post Count: 185 Status: Offline Project Badges: 
|
for something like seti@home, it's easier to do because a work unit may be one that includes info about a true extraterrestrial rf source, and that was one of the appeals of seti@home, the incredibly remote chance that the random dataa your PC crunched might stand out. (kind of like buying a lottery ticket) As I understand it, this project is less sensational, no single work unit is going to produce special groundbreaking results. not as glamorous as discovering ET, but we're all contributing to a common goal nonetheless.
|
||
|
|
|