How to do research To provide some hints on what to do and what not - - PDF document

SLIDE 1

24/01/2018 1

How to do research

Scuola Superiore Sant’Anna

Giorgio Buttazzo

g.buttazzo@sssup.it

Aim of the course

To provide some hints on what to do and what not to do when you are interested in:

• doing research
• writing papers
• presenting your results

2

Outline

• 1. Understanding research
• 2. How to write scientific papers

– Making a good structure – Typical mistakes

• 3. Getting into the publication process
• 4. How to make presentations
• 5. How to write research projects
• 6. Simulating a small conference

3

Meaning of research

4

Research = Re - Search

• Re

 again

• Search 

examine carefully, try, test, probe  Detailed study of a subject aimed at inventing new solutions for specific problems (engineering approach).

Research

 Detailed and systematic study aimed at discovering new knowledge about our world (scientific approach).

Scientists vs. Engineers

5

Scientists

They are interested in

• the world as it is
• understanding the behavior of

what already exists

Engineers

They are interested in

• changing the world
• solving practical issues
• inventing something that

does not exist

Basic vs. Applied Research

6

Basic research

aimed at advancing our knowledge with little concern for any immediate practical benefits. It is common to distinguish between:

Applied research

aimed at achieving practical outcomes that are useful to the society. Such a distinction, however, creates confusion on the meaning of Applied Research, often considered as the application of known results for making industrial products. Such a distinction, however, creates confusion on the meaning of Applied Research, often considered as the application of known results for making industrial products.

SLIDE 2

24/01/2018 2

7

Another important aspect of research is to produce results that can be generalized to slightly different situations.

Basic vs. Applied Research

Applying known results for developing specific products is not applied research. Applying known results for developing specific products is not applied research.

Hence it is good to clarify that

Good research is always characterized by a rigorous scientific study aimed at advancing knowledge and producing new results that (sooner or later) will be useful for the society. Good research is always characterized by a rigorous scientific study aimed at advancing knowledge and producing new results that (sooner or later) will be useful for the society.

Theory vs. Practice

8

What’s more relevant for the society? The theory of electro-magnetic field or a practical study to optimize a screwdriver?

Believing that a theoretical study is not useful to the society is completely wrong!!

Fundamental science

9

No modern technology would exist without fundamental science:

Galileo Newton Faraday Maxwell Schrödinger Heisenberg

They had no idea about the implications of their results. What motivated them were not applications, but curiosity! They had no idea about the implications of their results. What motivated them were not applications, but curiosity!

Einstein

Pasteur’s Quadrant

10

Fundamental Understanding Consideration of use Low High Low High

Pure basic research Use‐inspired research Pure‐applied research Bad research

Niels Bohr Thomas Edison Louis Pasteur

by Donald Stokes, 1997

What is Engineering?

11

• Cost-effective

Considering design trade-offs on resource usage

• Solutions

Emphasis on building devices, but also methodologies

• Practical problems

Problems that matter to people and improve human life

• Application of scientific knowledge

Systematic use of analytical techniques ( generalization)

“Engineering is the development of cost-effective solutions to practical problems, through the application of scientific knowledge”

12

Real system Real system Mathematical description of the system

Relevant variables Relations among variables Definition of

model

The abstraction process

Abstract World Real World

SLIDE 3

24/01/2018 3

13

A simple robot model

L

= link length

(x, y) = tip coordinates 

= joint angle Relevant variables

x = L cos y = L sen

Relations among variables

Parameter Output variables Control variable

(x, y)  L

The abstraction process

14

Real system Real system

Analysis Analysis

solution

Problem instance Problem instance Abstraction

assumptions model

Abstract World

implementation

Concretization Solved problem Solved problem

 

Real World

15

solution

Analysis Analysis

implementation

Problem instance Problem instance Abstraction

assumptions model

Concretization

Experiments Experiments

Solved problem Solved problem

Real system Real system

Evaluation

Good knowledge of Math and Physics is important

The abstraction process

Real World Abstract World

Types of contributions

16

• 1. Building a new model and

provide the related analysis

• 2. Solving a problem, finding

an algorithm

• 3. Building a tool
• 4. Increase the performance
• f an existing approach
• 5. Performace study
• 6. Buiding an application

Mendeleev’s periodic table, public key in cryptography, Liu & Layland model) Bubble sort, Rate Monotonic scheduling, Priority Inheritance Matlab, Simulink, Schedulability analyzer make it faster, smaller, cheaper, more resource efficient Compare and evaluate different algorithms and approaches

(by decreasing impact)

Developing a working system Space of all research areas

17

The path to success consists of three simple elements: Find what interests you that you can do well and is needed by the people.

• Lui Sha

your talents your interests social needs Your research topic should be there

Elements of success Elements of success

18

your talent your interest social needs

Important questions to answer:

• What is easier for you, writing a complex

software program or proving a difficult theorem?

• What excites you?
• What excites the community?
• Does it fall into your skills?

Understading yourself

SLIDE 4

24/01/2018 4

19

Be the best on something (anything)

average skill disciplines

Music is more important than Soccer, but it would be crazy if Messi would quit soccer to compose a minor song. Music is more important than Soccer, but it would be crazy if Messi would quit soccer to compose a minor song.

Elements of success Importance of idols

20

Many people became good in something because they were astonished by the ability of someone elected as their idol.

Idols can also be found in science when you are illuminated by the paper written by someone! Idols can also be found in science when you are illuminated by the paper written by someone!

Research step by step

21

• 0. Find a

topic

• 1. Understand

the field and the application domain

• 2. Think of

the problem

• 3. Formalize

the problem

• 4. Find a solution

to the problem

Research step by step

22

• 5. Validate your

solution in practice

• 6. Compare your solution against existing
• nes
• 8. Communicate the solution

to the world

• 7. Extend the solution to

more general cases

• 1. Understanding the field
• Why is it important?
• What are the main application areas?
• Identify the key components (HW and SW) of your

system.

• What disciplines are needed to understand details?

Do you have the required expertise? If not, which courses do you need to attend?

23

• 1. Understanding the field
• Are there ethical issues involved (e.g., weapons,

animals)? Better to clear these things up before starting.

• Talk to people that already worked on the topic to

know tricks, shortcuts, and bottlenecks.

• Read

the documentation related to the main components (HW and SW). You must become the Master on that subject.

24

SLIDE 5

24/01/2018 5

• Read the literature related to your problem. Search
• n the right conference proceedings (ask your

supervisor about the top conferences in the topic). Conference papers have more recent results.

• Get familiar with the work done before (if any): use

tools, read the source code, run demos, test the system functionality, discover critical points and take notes on strange behaviors.

25

• 1. Understanding the field

Common mistakes

• Think, plan, read, and never

get into the main game

26

Reading is important, but it is also important you start experiments, simulations and analysis. You may find issues are quite different than you thought.

• When reading papers, be critic,

never take everything as a dogma.

The highest impact on research is due to novel ideas that criticize imprecise thinking.

 Read papers as a reviewer!

• 2. Thinking of a problem
• Which features would you like to have
• n your system?

27

• What is the novel contribution you

would like to give to the field?

• Make a wish list, from big dreams to

smaller (but more realistic) goals.

• Go back to the literature to see if some

problem has already been solved.

Common mistakes

28

• Choosing a wrong direction

Your supervisor may have the experience to advise on this.

• Too simplistic assumptions

Nobody will be interested in a solution that is not applicable to a real situation.

Common mistakes

29

• The problem has already been solved

Carefully look at the literature!!

Don’t get desperate, you may

 Find a better solution;  Extend the existing solution to more general cases;  Slightly change the problem to cover more realistic cases;  Find another problem.

• 3. Formalizing the problem
• If the problem is too complex, break it into a set of

smaller sub‐problems.

• Build a model of your system.

30

What is a model? It is an idealized representation of a physical system.

“The best model of a cat is another, or preferably the same, cat.” ‐ Rosenblueth & Wiener (Philosophy of Science, 1945) “The best model of a cat is another, or preferably the same, cat.” ‐ Rosenblueth & Wiener (Philosophy of Science, 1945)

SLIDE 6

24/01/2018 6

• 3. Formalizing the problem

31

What is a model? A model is a representation of something. It captures not all attributes of the represented thing, but rather

• nly those seeming relevant for a specific purpose.

“Confusing a model with reality would be like going to a restaurant and eat the menu” Golomb’s Law on mathematical models “Confusing a model with reality would be like going to a restaurant and eat the menu” Golomb’s Law on mathematical models

• 3. Formalizing the problem

32

What constitutes a good model?

• It should be expressive: an accurate representation of

reality;

• It should be tractable: provide results in a bounded

amount of time.

The Empire cartographers were so smart that produced a map of the land as big as the Empire, but the King found this map completely useless. The Empire cartographers were so smart that produced a map of the land as big as the Empire, but the King found this map completely useless. In this sense, all models are wrong but some are useful. In this sense, all models are wrong but some are useful.

• 3. Formalizing the problem

33

Unfortunately, expressiveness and tractability do not get along very well

expressiveness Untractability (complexity)

Useless models (too far from reality) Useless models (too complex to be analyzed)

Different approaches

There is an important difference in the use of models between scientists and engineers.

34

A scientist seeks a model to match a physical system. A scientist seeks a model to match a physical system. An engineer seeks a physical systems to match a model. An engineer seeks a physical systems to match a model. For instance, logic gates are deterministic models that describe the behavior of electrons in silicon circuits. But the usefulness of this model depends on our ability to build silicon structures that are faithful to the model.

• 3. Formalizing the problem
• Building a model imply the following tasks:
• Clearly identify the assumptions you need to simplify

reality (but not too much);

• Define the system interface (variables are exposed to

the user);

• Define the metrics for evaluating the outputs of your

system and its performance.

35

• 3. Formalizing the problem
• Distinguish variables between:
• System parameters (variables you don’t want to change);
• Input variables (primary variables affecting your method)
• Design variables (variables you want to change to apply

your control actions);

• State variables (variables describing the system state and

behavior);

• Output variables (variables you want to measure to

evaluate the performance of your method).

36

SLIDE 7

24/01/2018 7

• 3. Formalizing the problem
• Find relations among the variables: how outputs are

influenced by inputs?

• If relations cannot be easily derived by the physics,

make experiments:

• Decide a single input variable;
• Decide a single output variable;
• Fix the others as parameters;
• Vary the input variable in a given range;
• Measure the output for each value of the input.

37

Example 1

Determine the timing of a traffic lights in a crossroad:

38

STOP STOP

S1 S2 S1 S2

What are the relevant variables?

G T

• duration of green (G)
• period (T)
• flow in each road (i)
• queue lengths (qi)
• average car speed (v)
• average delay of a car

Example 2

Design a communication protocol in a computer network:

39

What are the relevant variables?

• number of nodes

n

• network topology
• average load



• message size

M

• message rate

T

• network bandwidth

B

• end-to-end delay

D

• error probability

p

Example 3

Design a control algorithm for a robotic device:

40

Relevant variables

Ball & Beam

• ball position:

x

• ball speed:

v

• beam angle:



• servo angle:



• sampling period: T
• settling time:



• overshoot:



• ball mass: m
• ball radius: r
• beam length: L
• ball's moment of inertia: J
• gravitational acceleration: g

System parameters

• 4. Finding a solution
• Experiments can give good hints for finding a

solution.

• Even theoretical results can come to your mind by

first looking at experimental data.

• Sometime, experimental data can be performed to

find counterexamples. Counterexamples brings usually bad news, but they are extremely useful to identify critical situations and countermeasures to avoid them.

41

• 4. Finding a solution
• If

the solution is an algorithm, evaluate its computational complexity. If it is too high, simplify the assumptions or try heuristic approaches.

42

Write notes and comments every week to monitor the progress of your research:

• problems encountered (how you solved them);
• experiments you carried out;
• assumptions you are making;
• scenarios you have considered, etc.

Everything will be useful when you write a paper.

SLIDE 8

24/01/2018 8

• 5. Validating the solution
• Implement the solution;
• Apply the solution to one or more specific cases;
• Evaluate the performance according to the metrics

previously defined, under several conditions (different values of the input or state variables).

43

• 5. Validating the solution

Often, evaluating the solution on a simulated system is more practical and reasonable. In this case:

• input or state variables should be generated as

random numbers within given ranges and realistic distributions.

• output variables should be averaged on several runs

for each input configuration.

• Remember to record both mean and variance for

each set of output values.

44

• 6. Comparing the solution
• Identify all the differences

(assumptions, model, complexity, performance);

• Implement the best methods you have found.
• Run the same set of experiments for each approach.

If your solution performs worse than the others for all possible scenarios, then you are in trouble.

45

• Search the literature to find different

solutions to the same problem;

• 6. Comparing the solution
• Before giving up, make sure you compare the

approaches with respect to different metrics:

• ften it’s reasonable to sacrifice performance for

a lower complexity or a more realistic model.

• Sometime the improvement is observed only within

a certain range of the input variables. This can still be interesting, since that range can

• ccur for particular systems, for which your

solution can be more appropriate than the others.

46

• 7. Extending the solution

Go back to the assumptions you made to simplify the system model:

• Are there assumptions that are too artificial or

restrictive?

• How the solution and the results would change by

relaxing those assumptions?

• How could the model be enriched with more

variables describing more details?

47

• 7. Extending the solution
• Are there other components to be considered?
• Is

the analysis tight? Is there space for improvements?

• Are experimental results complete?

48

SLIDE 9

24/01/2018 9

• 8. Communicating the solution

As soon as you have got an interesting result, write a technical report, do not wait until the whole project is finished.

• Remember to register the technical report and put it
• n your web page.
• If the work is done together with your supervisor, ask

him/her to revise it and put it on his/her web page, which has much more visibility to the external world. This ensures you the paternity of the result, in the case somebody should publish a similar work.

49

• 8. Communicating the solution
• Then, select a conference with your supervisor and

write a paper for that conference. This is important for receiving a quick feedback (normally after 2‐3 months) from a number of reviewers.

• In the meantime, you can keep working to extend

your work.

• If the paper is accepted, you have to make a

presentation at the conference. This is a very constructive experience, which is important for a number of reasons.

50

• 8. Communicating the solution

51

Why presenting a paper at a conference is important?

• You have the unique possibility to join a community
• f experts and exchange interesting ideas;
• After your presentation, some people can

give you a feedback to improve your work;

• You have the possibility to attract the

interest of big gurus in the field and make contacts with interesting people.

• 8. Communicating the solution
• With the feedback you received from the reviewers

and from people at the conference, you can think of writing a stronger paper for a journal.

• The issue on how to write a research paper will be

covered in detail later on.

52

General advices

• If you have a hard deadline, don’t rely on someone else.
• Most

students underestimate the importance

• f

physical aspects, such as:

• Using a sensor to acquire data into your program;
• Making an actuator to move and build a function

that controls it as desired;

• Ordering devices and have them on your desk;
• Making a new printed circuit from scratch;
• Soldering components and testing them.

53

General advices

• Write down an idea when you have it, because you

may easily forget it by the time you get to the office. Some ideas come early in the morning, so keep paper and pencil on your bedside table. Some ideas come at the restroom, so be ready.

• Do not be afraid to admit your ignorance. In academic

research one is continually venturing into new areas.

• Perfectionism and attention to detail is a virtue.

Taken to excess, it is a vice.

54

SLIDE 10

24/01/2018 10

How to get ideas

Make a selective reading: quickly go through the paper to see if there is anything interesting, then look for specific points. Questions to always have in mind:

• What is the main problem to be solved?
• What are the main assumptions.

If you don’t understand the math, make a step back to learn and then return to study the approach.

55

• Read widely, but remember, reading is not what

research is about. Too much reading could be a major danger that could prevent new ideas to flourish.

How to get ideas

• What to read?
• conference papers are the most up to date, but
• ften details are available in technical reports.
• Most of the time, the key idea is not in the details,

but in the methodology.

• When details are important and you cannot find

them, contact the authors directly by Email:

Often they feel honored to be considered and will be happy to provide additional information on their work.

56

How to get ideas

• Reading approach
• Some degree of skepticism is appropriate when

approaching most papers.

• Don’t trust anything you read.
• Verify the theoretical results and judge the

assumptions trying to imagine them applied to your system.

57

How to get ideas?

• Think of some practical issue

– Start from problems that really bother you: Dissatisfaction is the major source of inspiration

58

How to get ideas?

• Be driven by passions

– Passion is one of the strongest forces in Nature

59

Search the web using combined keywords, like:

• music & statistics
• painting & algorithms
• biking & sensors
• soccer & measurement

You will find interesting unexpected results.

How to get ideas?

• Observe Nature (Animals, plants, phenomena)

All solutions found by Nature are characterized by – High Efficiency – Beauty and elegance

60

SLIDE 11

24/01/2018 11

How to get ideas?

• Look into other fields

– Get knowledge from other fields: different problems may have similar solutions

61

At the beginning, documents were written using a single font: Courier Many different fonts are available today much nicer to read!

An example:

How to get ideas?

62

An interesting story

• Simply out of curiosity, Steve Jobs decided to attend a

calligraphy course, where he learned all about typefaces, fonts, letter spacing, and what is needed to make printing a beautiful form of art.

• The course had no practical value that he could

imagine, and he took it just for pleasure.

• Ten years later, his experience was crucial for the

design of the Macintosh user interface, which set a standard for all word processors.

How to get ideas?

63

What do we learn?

• Curiosity means pursuing something that does

not necessarily have an obvious purpose.

• We examine something merely because it seems

interesting.

• Later, we may discover unexpected connections

with what we are doing that can form the basis for new ideas.

How to get ideas?

64

Other examples

• Elastic tasks model
• Gravitational scheduling

Some hints

• It’s very useful making lists, and then organizing them.

Examples are lists of questions, possible experiments, possible surveys, sources of data, etc.

• Imagine the results of experiments before making

them: if the results were to be ..., then we would conclude .... otherwise our conclusion would be ....

• Think of an experiment that nobody else has done,

and do it. The emphasis here is on doing something because it can be done, not because you actually understand the implications.

65

• Take advantage of being in a Lab
• Talk to people to know what they are doing.
• Ask successful researchers in your department if

they have particular methods for getting ideas.

• Attend seminars: you can find interest topics and

you learn how to present (or not to present).

• Brainstorm with your colleagues about problems

they have.

• Collaborate with someone on a common problem.

66

What to do if you don’t have ideas?

SLIDE 12

24/01/2018 12

• Criticize a paper by someone else:
• Are there assumptions that are too artificial or

restrictive?

• How could the model be enriched with more

variables, describing more details?

• Are there other components to be considered?
• Is

the analysis tight? Is there space for improvements?

• Are experimental results complete?

67

What to do if you don’t have ideas?

• Build on top of existing work:
• Can I improve the experiments?
• Can I improve the theory?

 From deterministic to statistical analysis.  From fixed priorities to dynamic priorities.  From uniprocessor to multiprocessor or distributed.  From single goal to multiple goals (feasibility + energy)

• Can I build a tool that does it automatically?

68

What to do if you don’t have ideas?

• Reduce complexity
• System are becoming more and more complex
• Sometime too complex to be properly used or

fully understood

• Simplification is a very promising process in the

future at many levels:

 Less computational complexity  Smaller program size  Simpler interfaces

69

What to do if you don’t have ideas?

Seek for simplicity

Achieving simplicity is very difficult, because

• Simplicity implies:

– deep understanding – distilling – beauty and elegance

• It needs time,

– Good ideas need incubation and deep thinking

70

Examples in science

71

E = mc2 F = ma F = k x V = R I

factorial(n) { if (n == 0) return 1; return (n * factorial (n-1)); }

Achieving simplicity

Strategies to handle complexity:

• Partitioning
• Isolation
• Abstraction

72

SLIDE 13

24/01/2018 13

Partitioning

It is the process of dividing a system into independent parts that can be studied in isolation.

73

Isolation

It is the process of focusing on a specific component, to understand its behavior in relation with the others:

74

Abstraction

• An Abstraction is a concept that connects a set of

items as a group or category.

• Abstractions reduce the information content, retaining

what is relevant for a particular purpose.

75

“ball”

Abstraction

In Computer Science, abstractions allow program designers to separate methods from implementation mechanisms, so that they do not depend on software or hardware: Examples

• A queue
• A resource manager
• A virtual processor

76

Achieving simplicity

77

Techniques to simplify a process:

• Reduce functionality
• Hide extra functions
• Organize your knowledge

Reduce

Today, new functions are included just because it is possible, not because they are needed.

• Reduce functionality

78

SLIDE 14

24/01/2018 14

Reduce

• Balance between essential and fancy

79

set mode clock alarm down up date temp

Crucial issue to be addressed

• What to remove?

Reduce

• Simplicity saves you time. And time is value.

80

UP DOWN clock alarm

Hide

• If you cannot reduce any further, hide non

essential functions.

81

Crucial issues to be addressed

• What to hide?
• Where to put?

Hide

• Extra functions can be revealed only when

needed

82

Hide

• Hiding can be hierarchical

83

Organize

• Group similar items into categories

84

SLIDE 15

24/01/2018 15

Organize

Making a taxonomy is crucial for grasping the key features of a set of similar items:

• State of the art papers
• Neural learning paradigms
• Scheduling algorithms
• Control schemes
• Power‐aware strategies
• …

85

Organize

When I started working on Neural Networks:

86

Back propagation Kohonen networks Self-organizing maps Reward-Based Functions Recurrent networks Associative networks Competitive learning Hopfield networks Reinforcement Learning Q-learning

Organize

Slowly some order emerged from the chaos…

87

Learning paradigms supervised competitive reinforcement

Hopfield Back-Prop. Kohonen. ASE-ACE Q-learining. SOM.

Organize

88

Rate-Monotonic Deadline Monotonic Sporadic Server Priority Inheritance Priority Ceiling Slack Stealer Least Laxity First Fixed Priority Scheduling Polling Server EDF

The same thing happened on Real‐Time Systems:

Organize

The same thing happened on Real‐Time Systems

89

Scheduling algorithms

Fixed-priorities

RM DM EDF LLF CBS PS

Dynamic priorities periodic aperiodic periodic aperiodic

DS TBS

Organize

Strategies useful for achieving organization:

• List all the features
• Sort
• Group into categories
• Integrate (reduce to fewer groups)

90

SLIDE 16

24/01/2018 16

Organize

The key of software organization:

91

VERY IMPORTANT a TAB character is not equivalent to N spaces

Organize

92

void sort(int *v, int n) { int i, j; // array indexes int min, k; // value and index of minimum element int temp; // temporary variable for (i=0; i<n-1; i++) { min = v[i]; // find minimum bewteen i and n-1 k = i; for (j=i+1; j<n; j++) if (v[j] < min) { min = v[j]; k = j; } if (k > i) { // swap v[i] with v[k] temp = v[i]; v[i] = v[k]; v[k] = temp; } } }

Organize

93

void sort(int *v,int n) { int i,j; // array indexes int min,k; //value and index of minimum element //temporary variable int temp; for( i= 0 ;i <n-1;i++ ){ min = v[i]; //find minimum bewteen i and n-1 k =i; for (j =i+1;j <n;j++) if(v[j]<min) { min=v[j];k=j;} if( k> i){ //swap v[i] with v[k] temp = v[i]; v[i] = v[k]; v[k] = temp; } } }

Something that should NEVER be produced:

The best example

94

Research components

Applic Spin-off researchers Projects papers books tools conferences teaching people money students

Services are also required

• Reviews
• 3 reviews per paper per year
• Project reviews
• Teaching
• seminars, lectures, courses
• Take it as a pleasure for tranfering knowledge
• Organization
• conferences, worshops, special tracks, demos,

summer schools

96

SLIDE 17

24/01/2018 17

Research hierarchy

Find industrial contacts Meet with industries Propose projects Get money, get positions Teach courses Write Papers Write Proposals Organize demos Teach courses Full professor Associate professor Assistant professors Post Doc PhD Undergrad students

Some Ethical issues

98

• Don’t write papers without involving your supervisor
• He/she is responsible for your research
• Not including him/her may be seen as a problem
• Don’t take initiatives without consulting your supervisor
• Do not establish contacts with new people/companies/groups,

since they may affect/involve your supervisor anyway.

• Do not accept invitations to conferences/talks/courses, even if

they are for free.

Some Ethical issues

99

• Answer all Emails (from your collaborators)
• Not answering is considered a bad behavior
• Acknowledge reception of all messages
• Notify your absence from the lab
• You are supposed to stay in the lab at working hours
• If you cannot come for any reason, please send an Email

Some Ethical issues

100

If there are problems with your supervisor,

do not violate the mentioned ethical rules, but

• 1. Talk to him/her trying to solve them;
• 2. Talk to the PhD coordinator;
• 3. Ask the Advisory Board to change supervisor,

explaining the motivations.

If there are problems with your supervisor,

do not violate the mentioned ethical rules, but

• 1. Talk to him/her trying to solve them;
• 2. Talk to the PhD coordinator;
• 3. Ask the Advisory Board to change supervisor,

explaining the motivations.