Done. 53%

Done 55%

Voted yesterday. Just checked: 63%

THANK YOU *everyone* who has voted! You've helped put us in first place - where we hope to stay until voting closes on 21 April!

I'd also like to address a couple of comments made in this thread..

SekeRob - runtime fluctuations are typical especially anytime we launch a new project or an existing project makes changes to their work unit sizing, for example. It typically takes at least a few days for big contributing members and organizations (such as IBM) to adjust their project selections to maximize their contributions. In fact, as of yesterday, runtime levels have almost fully recovered from the Help Stop TB launch.

keithhenry - I'm sorry that you feel that we launched Help Stop TB before it was ready, just to capitalize on World TB Day. The decision to launch on World TB Day was made jointly by the researchers, our team and communications folks from both organizations. And yes, in making that decision, we weighed both the technical readiness of the application with the optimal conditions to get this research effort as much attention from new volunteers as possible. But put simply, we wouldn't have launched if we didn't feel ready. In fact, World TB Day was not the original launch date we had set for this project, which we initially hoped to launch in late Feb or early March. We kept delaying the launch until we were comfortable we were technically ready.

Our technical team has been working on this application and with the research team for many months, during which we performed several levels of testing including more than one round of beta testing. With each round of testing, issues were identified, the most serious of which were addressed and fixed. I don't deny that we have some teething problems but (I'm sure developers will agree) no application is released defect-free on launch day (or ever!). It's a constant and iterative process of prioritizing defects, addressing the most critical ones, until you're comfortable with the remaining issues (e.g. because they're not severe enough, or don't impact the vast majority of your users, to justify further delaying a launch).

Also, and I believe you acknowledge this anyway, as valuable as beta testing is, it doesn't always catch every problem - not necessarily because we don't run beta for long enough, but because the beta community may not fully represent the breadth of devices across the wider volunteer community. So we release new research applications, knowing that we may uncover new issues, and ready to address them. For example, we're currently investigating an issue with devices of a certain processor type (luckily not commonly used) which is having a slightly higher invalid rate that we'd like to see.

Finally, from the researchers' perspective, World Community Grid often gives them an opportunity to conduct research at an unprecedented scale and underestimate how quickly our volunteers go through research tasks. This means that there's only so much they can prepare for ahead of time and they often (as is the case with Help Stop TB) need to run initial batches, examine the results, before building more work and automating or at least stabilizing the process of generating more work units. That's why, for example, we sometimes have to start a project slow and speed it up once the researchers can provide a steady supply of work.

I hope this helps explain our rationale and approach.

I just voted (current percentage : 65%).