Math Grant Proposal Writing Tips & Ideas

[Dragonfall]

Here's one to your mathematicians out there: What do you write in your grant proposals? I mean, you can't really outline an experiment which can only be done with the grant money or anything.

 

[Choppy]

I'm not a mathematician, but I don't understand your line of reasoning.

The point of a proposal is just that: a proposal. When you're applying for funding, the funding agencies want to know that you a specific plan for the money they are willing to issue and that you have a reasonable chance of achieving the goals you set. Hence the proposal itself will generally contain a relevant literature review and the results of any preliminary or pilot work. These are then used to frame the project outline that follows.

 

[Dragonfall]

I mean the only money you can ask for is a salary for yourself, and possibly money for travel and students. It's not like mathematicians need a lab, 5 technicians and millions worth of equipment.

 

[aostraff]

Why does a grant proposal have to involve a large sum of money? But to answer your question there are other factors I can think of such as paying other people, purchasing software and hardware.

 

[Civilized]

I mean the only money you can ask for is a salary for yourself, and possibly money for travel and students.

Add in computers and you're right, that's about it. In the US people with the title "Research Professor" are supported almost entirely by their own grants, and of course regular professors also apply for grants so that they can higher more grad students or do less teaching.

It is not as if mathematicians or theorists have to write demeaning proposals in which they beg for a salary, they just describe why their research fulfills the grant description: "We propose to work on this and this, for these reasons, and in conclusion we thank for the commitee for considering our proposal." The grant will require some degree of accounting of the expenditures, and everyone expects salaries to be a part of this.

 

[jquark]

there are a great many experiments in mathematics that require as much computer as can be obtained. if you want to model the universe you need a second universe of equal mass. approximations will be required by people decended from us 10 billion years from now, it simply must be approximated. nevertheless scope needs be maximized for resolution increase in detail. then there are things like just three objects which cannot be solved to begin with, and requires thousands of iterations with even so little material. space trajectories require this, and places that do that have the programs hardwired on chips. you'ld be suprised how much data one can generate. and how much time it takes. the full use of a supercomputer for a year is around standard if one can get that. i mean one single program running that long say at 60% capacity is typical for some models such as the early universe. weather forcasting in america uses three at severe storms center last i heard. (supercomputers) modelling water freezing, explosions, and so forth take a great deal of space. pure mathematics can be hard to separate from that, because they arent real, they are models, and then you have to tune them. different algorithms are more efficient. so sometimes there are researchers, who tell programmers what to do, and that makes up a pretty good team.