Vanity of Vanities, all is Vanity

New comment on your post #366 “Class notes on algorithms…”
Author :
Comment:
Yes, that helped a lot. It was important to consider the proposal in context with the other ones and looking at the “bang for the buck.” A proposal seeking $1Million done by a private company certainly ought to contain more deliverables than one from academia seeking $150K, most of which would be spent on graduate students.

As an aside, it would be interesting to have a “citation index” for proposals. That is, when a paper is cited in a proposal, it would seem a stronger form of flattery because the proposer is putting his prospects for funding on the line by citing a paper, and would not wish to appear foolish. But often the authors of the cited papers are unaware of these citations. I know two people now who have cited your papers as part of a fundraising inititative, and I’m sure you are completely unaware either one.

Yes, that helped a lot. It was important to consider the proposal in context with the other ones and looking at the “bang for the buck.” A proposal seeking $1Million done by a private company certainly ought to contain more deliverables than one from academia seeking $150K, most of which would be spent on graduate students.

As an aside, it would be interesting to have a “citation index” for proposals. That is, when a paper is cited in a proposal, it would seem a stronger form of flattery because the proposer is putting his prospects for funding on the line by citing a paper, and would not wish to appear foolish. But often the authors of the cited papers are unaware of these citations. I know two people now who have cited your papers as part of a fundraising inititative, and I’m sure you are completely unaware either one.

New comment on your post #366 “Class notes on algorithms…”
Author : Sariel Har-Peled
Comment:
I hope you are kidding when you are refereeing to me as a top-notch theoretician. I am not really a theoretician, and I am top notch only if you hold the notch upside down…

Anyway… Oh man. Refereeing is the hardest thing we have to do in our job. In some sense it is the way we can have the most influence on our field. To begin with we have our own bias, then it is a game that involves other people (for example, in NSF refereeing, all proposals are judged to be excellent, or extremely excellent, or splendidly extremely excellent, and ola-la-la-la excellent. If you get only excellent, it means your proposal sucks. [I am stretching reality a bit, but only by a bit.])

Anyway, relating to your question, the first step is to realize that you have to judge proposals and papers in comparisons to the other proposals and papers in the pool. There is no point saying a proposal by say X on this topic would have been much better, since X might be lazy in writing proposals or just not really interested in this topic (although this is a valid consideration for the funding panel chair – he can contact X and ask him to apply). How is this proposal compared with the other proposals? Is this guy going to work with other people that can do the less theoretical part of the job?

In some sense, my feeling is that one has to judge how much value one would get for the money. A proposal can be wonderful, but if the author is clearly unable to deliver (thats why track record is important) what she promises, then it might be completely useless. If on the other hand, the PI can not deliver but her questions are amazingly interesting, then just getting the questions out (and getting strong people to work on them) is worth the money. Ask yourself, in then years from now, if I fund X or fund Y, what (in expectation) would be the higher yield. What is the time frame? In a time frame of 20 years, theoretical ideas are more important, but in a time frame of five years a practical deliverable might be more valuable.

As for myself, I always try to be one level more positive than I feel about such things. In any case, you can put all your considerations in your “referee report”. Thats what the refereeing process is all about…Let the panel/chair/whatever make the final decision. Just make sure that your grading/rank is in sync with the rest of the people on the panel/whatever…

And do not take too seriously the written guidelines. People are probably ignoring them anyway.

Hope this helps…

I hope you are kidding when you are refereeing to me as a top-notch theoretician. I am not really a theoretician, and I am top notch only if you hold the notch upside down…

Anyway… Oh man. Refereeing is the hardest thing we have to do in our job. In some sense it is the way we can have the most influence on our field. To begin with we have our own bias, then it is a game that involves other people (for example, in NSF refereeing, all proposals are judged to be excellent, or extremely excellent, or splendidly extremely excellent, and ola-la-la-la excellent. If you get only excellent, it means your proposal sucks. [I am stretching reality a bit, but only by a bit.])

Anyway, relating to your question, the first step is to realize that you have to judge proposals and papers in comparisons to the other proposals and papers in the pool. There is no point saying a proposal by say X on this topic would have been much better, since X might be lazy in writing proposals or just not really interested in this topic (although this is a valid consideration for the funding panel chair – he can contact X and ask him to apply). How is this proposal compared with the other proposals? Is this guy going to work with other people that can do the less theoretical part of the job?

In some sense, my feeling is that one has to judge how much value one would get for the money. A proposal can be wonderful, but if the author is clearly unable to deliver (thats why track record is important) what she promises, then it might be completely useless. If on the other hand, the PI can not deliver but her questions are amazingly interesting, then just getting the questions out (and getting strong people to work on them) is worth the money. Ask yourself, in then years from now, if I fund X or fund Y, what (in expectation) would be the higher yield. What is the time frame? In a time frame of 20 years, theoretical ideas are more important, but in a time frame of five years a practical deliverable might be more valuable.

As for myself, I always try to be one level more positive than I feel about such things. In any case, you can put all your considerations in your “referee report”. Thats what the refereeing process is all about…Let the panel/chair/whatever make the final decision. Just make sure that your grading/rank is in sync with the rest of the people on the panel/whatever…

And do not take too seriously the written guidelines. People are probably ignoring them anyway.

Hope this helps…

New comment on your post #366 “Class notes on algorithms…”
Author :
Comment:
Sariel: I apologize for being off topic. But I need advice. I’m a fan of your work and your blog. But let’s suppose that you were a reviewer of grant proposals for a government agency. Suppose further that one of the major factors for evaluation is “relevance to [the agency's] objectives, including the applicablity of the proposed investigation to technology needs” and you see that the guy is completely theoretical [he even cites a Har-Peled paper!!] but not a star.

Half of me wants to recommend the proposal for at least having taste in topics. The other half wants me to reject it with the comment that if [funding agency] wishes to fund theory, they should get a top-notch theoretician instead. But you guys don’t write such proposals.

Thoughts?

Sariel: I apologize for being off topic. But I need advice. I’m a fan of your work and your blog. But let’s suppose that you were a reviewer of grant proposals for a government agency. Suppose further that one of the major factors for evaluation is “relevance to [the agency's] objectives, including the applicablity of the proposed investigation to technology needs” and you see that the guy is completely theoretical [he even cites a Har-Peled paper!!] but not a star.

Half of me wants to recommend the proposal for at least having taste in topics. The other half wants me to reject it with the comment that if [funding agency] wishes to fund theory, they should get a top-notch theoretician instead. But you guys don’t write such proposals.

Thoughts?

I just posted my class notes for the graduate algorithms class I am teaching this sememster on the web. I spent a lot of time on them this semester and I think they are acceptable.

Since we are not on this topic: Why do people like the TV series “battlestar galactica”? It does not make any sense at all. It makes less sense than a Bond movie. What I really would like to know is who is stronger – Aslan or the Cylons? I can just see the “scientist” from BS going around saying “there is no Aslan” every ten seconds for the next 10 episodes of this series, while being sexually attacked by a wardrobe.

New comment on your post #355 “Min-cost network flow”
Author : Sariel Har-Peled
Comment:
The Ahuja book while nice is not self contained. I am a great believer in self contained and succint representation of stuff. Long recursive books put me off…

The Ahuja book while nice is not self contained. I am a great believer in self contained and succint representation of stuff. Long recursive books put me off…

New comment on your post #355 “Min-cost network flow”
Author : Deepak
Comment:
check out this book – Network Flows, by Ahuja, Magnanti and Orlin. One of the best books on network flow algorithms, and has detailed description of min-cost flow problems and solution strategies.