Bill Foster actually said...

Well, I was interested that it has repeatedly come up in this testimony that data errors are--a dominant source of them are misidentificatio...

I think it's a very bad predictor of success, but it's one of the--it's easily measurable, so we use it, but it's not at all good.

But there's a real need for something like ARPA-E to fill gaps in the private-sector research and development.

Thank you. And I guess I'd like to start off by just seconding all the praise that's been showered on ARPA-E for its achievements to date an...