How not to get an ERC grant Stefan

How (not) to get an ERC grant Stefan Dziembowski

Disclaimer 1. My advice is based only on my own experience. 2. My field is computer science. Some of my advice may not be applicable in other areas.

Part 1 How to get an ERC grant?

Typical situation in grant competitions (simplified) outstanding projects Here a lot depends on luck can be improved! threshold good projects other projects

Typical scheme more important candidate project application Questions: 1. What to choose the topic? 2. How to present it? host institution

Questions �How to choose the topic? �How to present: ◦ your CV? ◦ your project?

Main question Does your project match the call?

Example ERC grants aim to support "Frontier Research", in other words the pursuit of questions at orbeyond the frontiers of knowledge. [. . . ] In particular, [. . . ] pioneering proposals addressing new and emerging fields ofresearchor proposals introducing unconventional, innovative approaches and scientific inventions are encouraged.

Hence: an application of a type: “In our last paper we have proven that blah holds for n = log 2. 34 m. We now plan to examine the case when n = logλ m, for λ > 2. 34. ” may be rejected regardless of the other merits.

What to put into your application? Invent your own problems! The reviewer should not get the impression that it is: yet another application on X. . .

Also: Avoid long-standing open problems The reviewer will write: “the project is too ambitious” ≈ “so many people already failed to solve it, hence so will you”

Questions �How to choose the topic? �How to present: ◦ your CV? ◦ your project?

Wrong approach Do not lower yourself to use “marketing tricks”. . . (it can work if you are a Nobel prize winner)

Right approach Challenge : Distinguish yourself from a crowd of similar candidates Moreover: Don’t loose points for stupid reasons

Basic strategy Look on your application “through the eyes of the reviewer” Question: What are his criteria?

The reviewer is also a human being! One should assume that he is competent and responsible. However it can happen that he: �doesn’t have time, �is irritable, �doesn’t know the topic in every detail.

Main principle Help the reviewer! �Do not assume that he will guess what you had in mind! �Do not ignore any mandatory part of your application. (otherwise he will have to give you zero points for this)

A non-expert should also understand some part of the application. (at least the beginning of it) Psychological effect : once he stops understanding he will at least get a positive impression about your application. advice frequently put comments like: “more information on this topic can be found in [. . . ]”

On the other hand: An expertshould get an impression that you are on top of the field. You need to show that you know the most recent results. Your application cannot be too general.

Questions �How to choose the topic? �How to present: ◦ your CV? ◦ your project?

Do not be too modest! Write sincerely what you are the most proud of. Example : If someone very famous cited you then write „These results were later used by X in [. . . ] (see page n)” Do not hope that the reviewer will find it out himself!

Questions �How to choose the topic? �How to present: ◦ your CV? ◦ your project?

Project �State-of-the-art �Methodology, �Resources and objectives

state-of-the-art The field X is very important because. . . Recently we witnessed an enormous progress in it [describe] objectives The goal of this project is to solve the following important problems [a list] sometimes the difference is unclear methodology to achieve it we will perform the following work [a list of research tasks ]

Methodology vs. objectives methodology we plan to prove Theorem X this will make the world a better place because [. . . ] objective

State-of-the-art �omitting �on someone’s contribution is risky the other hand : we don’t have much space. Solution: �mention all the important results at least briefly. �stress that the description is not as deep as we would like „be cause of the lack of space ”

Other advice

Timing �Do not write your application in the last moment. �Show it to your colleagues.

English Use correct English. On the other hand : remember that reviewers are likely not to be native speakers. Use simple language.

Style Being a little bit enthusiastic does not hurt: „We believe that new exciting results are just behind the corner”

Make the work of reviewer easier Use many references like: „As described in Section for ε > 0. 5. ” n, this is interesting only Don’t assume that the reviewer will read all your application at once.

Document format [1/2] Use all space available. This will allow you to say „be cause of the lack of space. . . ” Moreover: the reviewer will get an impression that you have much more to say.

Document format [2/2] Make it look like a scientific paper. In mathematics, physics and computer science: write it in La. Te. X.

Abstract Do not write it in the last moment. This part of your application will be read by the largest number of people Use it also for marketing!

Part II How not to get an ERC grant? (or: why there are so few ERC grants in Poland? )

Why there are so few ERC grants in Poland? Short answer: because the ERC evaluation is not based on the “ministerial list” , but on the scientific merits.

Polish system In numbers we trust! Key ingredient: the ministerial list of journals NATURE 50 ACM COMPUTING SURVEYS 50 JOURNAL OF THE ACM 45 INFORMATION AND COMPUTATION 25 Cement Wapno Beton 20 Drewno 15 Zeszyty Naukowe. Studia Informatica / Uniwersytet Szczeciński 7 STOC 0 FOCS 0 SODA 0 LICS 0 CAV 0 POPL 0 CRYPTO 0 37

How do the ERC people look at it? Anyone competent in the evaluation of computer science research and researchers knows that the role of. Web of Sciencefor computer science lies somewhere between irrelevant and harmful. prof. Bertrand Meyer ETH Zurich ERC Advanced Grant in 2011 president of Informatics. Europe cacm. org/blogs/blog-cacm/181755 -what-is-your-research-culture-part-3 -the-web-of-science/fulltext

Polish system prefers quantityover quality Example : Iuventus Plus grants for young scientists (around 200 grants each year, max 300. 000 PLN, duration: 2 years) First selection based purelyon a formula: 39

The journal evaluation committee The points are assigned by a committee composed of 13 members. Naturally: several areas of science are not represented.

The committee proposes “objective criteria” 13 pages like this: To be fair: in these circumstances they probably couldn’t produce anything better. . .

Result: In computer science the criteria from the “ministerial list” have little todo with the criteria of the international scientific community. A Polish scientist dilemma: do research. . . or collect points? One essentially needs to build two CVs (one for the scientific peers and another one for the bureaucrats ).

Point of view of some institutes A scientist that focuses on ambitious goals is a “parasite”. Better hire someone who produces a lot of mediocre results.

Why there almost all no Starting Grants outside of Warsaw? Univ. of Warsaw PAS Toruń Univ. of Warsaw IIMCB Warsaw PAS Warsaw IIMCB Warsaw Possible answer : because only large and “rich”institutions can afford to hire people who do not perform well according to the ministerial criteria.

Conclusion (for the policymakers) This system needs to be changed! Best solution: base it on the peer review. Or at least: adjust the “point system” to the international standards.

Conclusion (for the scientist) 1. When choosing a place to work : go to an institution that is not obsessed by the “ministerial list”. 2. Build your CV according to the international standards. Try to follow the publication habits of your colleagues at the top institutions(MIT, Stanford, Harvard, Cambridge, . . . ). Rule of thumb: your research is “internationally competitive” if it gets noticed at these places. Remember : for an ERC reviewer one “strong” citation from an MIT professorcounts more than 100 citations coming from some provincial university in Poland !

Thank you!