Comment by spekcular | Hacker News Reader

Comment by spekcular | Hacker News Reader

Parent Full thread

The author writes: "Is there a hypothetical world where I would be an academic working on the Birch and Swinnerton-Dyer conjecture right now? Sure. It's probably a world where high-flying students are given, upon graduation, some sort of "mini-Genius Grant". If I had been awarded a five-year $62,500/year grant with the sole condition of "do research", I would almost certainly have persevered in academia and — despite working on the more interesting but longer-term questions — have had enough publications after those five years to obtain a continuing academic position. But that's not how granting agencies work; they give out one or two year awards, with the understanding that those who are successful will apply for more funding later."

These awards certainly exist; the Miller institute at Berkeley, the Simons fellows in NYC, and the Clay Fellows are a few examples. (Some not quite for 5 years, but more than 2. Also, the pay is more like 90k.) The NSF math postdoc is also close to this (2 years full funding if you elect to not teach those years, plus a third year where you have to teach.) Regular NSF grants and NSF CAREER awards are also longer, IIRC, though perhaps less relevant to fresh PhDs.

So, in fact, these "mini-Genius Grants" do exist.

It's worth emphasizing this part of the quote from the author:

> But that's not how granting agencies work; they give out one or two year awards, with the understanding that those who are successful will apply for more funding later.

I've worked for a funding agency. It's tough to give out five year grants with no deliverables other than "be a genius". When you are funding basic research you have to be prepared for the idea that lots will fail, and many people will decide to leave. How do you know who is a genius when they are finishing their PhD? Everyone finishing a PhD at Oxford/MIT/Stanford/etc has a very impressive resume and the reality is you can't fund them all.

Two year grants seems a reasonable compromise here.

How do you know who is a genius when they are finishing their PhD? Everyone finishing a PhD at Oxford/MIT/Stanford/etc has a very impressive resume

Impressive resumes, sure. But my experience has been that even when students are starting their doctorates there's very clear tiers -- even out of the pool of doctoral students in Oxford.

(In case anyone is thinking this is hubris on my part: I would say I was towards the top but not at the top. The most impressive student I met in Oxford was a mathematician by the name of Lillian Pierce -- and she wasn't even doing her doctorate yet.)

Suppose you wanted to arrange to give the top 10% or so of math Ph.D.s who graduate from Oxford the sort of no-strings-attached five-year grant you were talking about. How would you set up the decision process for deciding which graduates to award the grants to? Would the divisional chair at Oxford assign the grants? Would the graduate students vote on who should get them? Would the Royal Society choose the recipients?

I don't want to claim that nobody was trying to influence your opinion about who the best students were, when you were at Oxford; of course it's common for people to tout their own achievements and those of their friends. But I suspect that if you were in the position of directing millions of dollars a year of no-strings-attached funding, and people knew that, the amount of effort that went into influencing your opinion would be an order of magnitude greater, and might be sufficient to confuse you.

(This pickle is presumably obvious to you but the solution to the pickle isn't obvious to me, so I'm hoping you can outline a solution.)

This kind of decision is already routinely made by committees awarding NSF postdocs, Clay fellowships, etc. What's wrong with the process they use (and independent expert panel)? Do you think they select the wrong mathematicians?

cperciva's criticism of this process is, as I understand it, that the grants they give are too short, and that this, together with the selection criteria, creates a too-strong incentive to pursue short-term, low-risk results.

> But my experience has been that even when students are starting their doctorates there's very clear tiers -- even out of the pool of doctoral students in Oxford.

This might be true, but it's really really hard for someone who doesn't work with them day-to-day to distinguish.

Your article pointed out the same problem in getting novel worked reviewed - it was extremely hard to find people for that. I know in my case it was even worse - funding proposals generally come from research group leaders, and there at least you have a body of work to look at and can judge their likelihood to succeed.

But if you are trying to look at an individual's work how can you judge it when all you have to go on is peer reviews and maybe interviews with supervisors or peers?

Supervisors' motivations are already complicated - most of the time they would want an individual in their group funded because their work is generally aligned with that research group's field.

It's a pretty tricky problem.

> This might be true, but it's really really hard for someone who doesn't work with them day-to-day to distinguish.

It's really not (in mathematics, the field under discussion). It's not hard to look at the CV of a graduating PhD student in math and tell whether they're a semi-reasonable candidate for the Clay Fellowship, or the other fellowships listed above.

Among the small group who makes that cut, of course you need to rely on letters of recommendation, expert assessments, and so on. But the initial cut is fairly straightforward.

It's interesting. We were mostly funding at the PhD level, and in computer science.

By contrast, I read the links for all the current Clay Fellows[1] and noticed that maybe 60-70% of them had one thing in common: They had proved some long standing (30 year+) conjecture. I agree that's a pretty good sign.

Of course the program they did their PhD through is a pretty good initial screen too:

Princeton 4

Harvard 4

Standford 1

Imperial College London 1

MIT 1

UT Austin: 1

[1] https://www.claymath.org/fas

Reply on news.ycombinator.com