Martin wrote: > Lynn W. Taylor wrote:
It doesn't matter if Paul is correct if his abusive tone overshadows his message. > Both points are correct in that: > > Boinc provides a framework for running science applications on untrusted > and unreliable hosts. That framework should include tools to ensure that > the results returned are the "correct" results from having been > correctly run. I agree with this up to a point. Where I disagree is covered nicely in your next paragraph. It is up to BOINC to transport the files to and from the project, and to schedule the work consistent with deadlines and resource shares. It provides facilities like having different hosts perform the same work so the results can be compared. > Whether or not the results are useful or "correct" for whatever science > is being explored for the project itself is indeed up to the people > running the project! Exactly. I tend to think in terms of s...@home because that is where I participate mainly. "Results" in that context don't require an intensive amount of rigor. In chemistry, you can measure the pH of a liquid accurately, but if you just need to know if it's acidic, you stick in a piece of litmus paper, and if it turns red, you've got an acid. s...@home is a litmus test. It's supposed to find things that might be signals and eliminate things that definitely aren't. > There is also a third "correct" in whether for the runtime > statistics/credits, the credits score can be considered to be correct or > just arbitrary random numbers that vaguely look to be in proportion to > CPU runtime as compared to a s...@h WU... If we're talking about BOINC-as-black-box -- about BOINC from a project scientist's view, this is pretty unimportant. That's one of the reasons we have inaccurate credit. From a participant view, it's a huge deal. That's why I'm in favor of anything that can dial-in credit automatically for a project without having to tie up the project scientists. They're busy enough doing their research. -- Lynn _______________________________________________ boinc_dev mailing list [email protected] http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev To unsubscribe, visit the above URL and (near bottom of page) enter your email address.
