Date: Thu, 17 Jun 1999 08:21:38 -0400 (EDT) From: James Stasheff Subject: Re: more Hopf postings Hovey writes: "If we did what the [National Science] Board wants us to do, we would fund 800 people instead of 1,400. "I completely disagree with the tenor of these remarks! Perhaps I am wrong about this, but it seems to me that doubling the size of each grant while giving half as many total grants would have a completely negative effect on mathematics! I don't understand why the NSF is so stupid as to want to do this--it seems like very poor management and not very cost-effective. Perhaps somebody can explain this to me, or perhaps together we can complain about it." Stasheff: Having served ont he NSF MATH advisory board, I can explain (without agreeing) some of what is involved here: 1. Notice the advice came fromt he Natl SCIENCE Board. One of our problems is that we are lumped together with the other HARD sciences which have a very different culture. Lab sciences have huge grants comparted to ours, but administrative costs within the NSF are NOT proportional to the size of the grant. Half as many grants doesn't necessarily mean half as many people, but up to now big pure math grants have gone to large groups, which leaves those of us with one or two colleagues at most in the same area at a disadvantage. Recent NSF proposals (I forget the acronym) which permit funding one grant across several institutions is a first step around the problem. Any other ideas? Complaining won't do it - that's been tried before. Suggestions of following the Canadian model - grants for travel and conferences and NOT summer support have been shot down repeatedly. Lewis: "We fund proposals, not individuals". That's an exaggeration at face value as is Mark's reaction. I have seen proposals where a big name proposes `to continue my work'. That won't/didn't fly. Proposals are funded to a great extent on reviews; reviewers can evaluate the proposal and the individual in any proportion they deem fitr. Unfortuantely the `grades' E, VG, G etc play too heavy a role in the NSF decisoin. Jim Stasheff jds@math.upenn.edu Until Dec 1999, I will be visiting U Penn but for hard copy the relevant address is: 146 Woodland Dr Lansdale PA 19446 (215)822-6707 As of Jan 3, 00 Jim Stasheff jds@math.unc.edu Math-UNC (919)-962-9607 Chapel Hill NC FAX:(919)-962-2568 27599-3250