Tuesday, December 10, 2013

Is Boinc losing its ability to draw projects?

This is both a post to raise awareness, and a question that I am not sure I know the answer.  Long story short, I realized how lopsided mine, and the Atlantic teams credits were recently, and I wondered if there was anything I could do to throw a little spice into the mix by adding another project.

That is where the issue came in, besides WCG and GPUgrid, there seem to be few to no projects that offer somewhat wide base appeal and focus on humanitarian issues with mass appeal.  As much as I am a fan of all sciences, you must admit, in terms of questions we need answers to now to vastly improve the lives of everyone on the planet, are mostly medically or biologically based.  Which seems to have few projects out there (few scientists with the know how/ time/ funding to do so?).  Moreover even WCG seems to be starving for projects and if anyone could attract top talent I'd imagine it would be them.  World Community Grid is down to 6 projects, 3 current and 3 intermittent.

Moreover the power of GPU computing has been recognized, almost to the point that it is almost making supporting GPU computing on any single project a self defeating task, in the sense that the people willing to complete the tasks will finish them in a fraction of the time it seems to talk your team to find a new item to study, then code the project for the GPU tasks. In fact it appears the only projects that can sustain GPU tasks either have a dedicated team of researchers/ programmers designed to churn out item after item after item as part of their funded research, or are actually tied to something that has such a massive amount of test cases that they can be considered near inexhaustible. (Scanning space for a given items.)


So is this lack of projects that *matter* and to a larger extent projects in general that there are currently as opposed to just a few years ago, a sign that the ability for BOINC or BOINC like systems are not effective?  Or could it in fact be that they are actually too effective?  I do know, and have realized that in large quite a few people lose faith in BOINC after awhile, I mean for one look at BOINCstats and just look at how many thousands of thousands of people you can flip through from the bottom of the ranks in which a majority of people listed have zero credits listed for the past month and an average credit that is very low or zero.

I for one certainly hope more projects are in the works-- in general, or at the very least for World Community Grid.


No comments:

Post a Comment

World Community Grid Signature