PhD Process: Part II Personal, Communication and Environmental Aspects

1
PhD Process Part IIPersonal, Communication
and Environmental Aspects

• Jayant Haritsa
• Computer Science Automation
• Indian Institute of Science, Bangalore

2
The PhD Process
REAL WORLD
Part II
RESEARCHCOMMUNITY
INSTITUTION
ADVISOR
Part I
THESIS
3
PERSONAL ASPECTS
4
Motivation for PhD

• VTU requirement for promotion
• Cultural capital (add the Dr prefix)
• Masochistic mindset (long pain period)
• Enjoy discovery / writing / talking

5
Work Ethic

• INDEPENDENT research
• It is YOUR thesis, not the Advisors!
• The advisor is not a NANNY!
• CANNOT do PhD as side business
• Focus on ONE problem at a time
• 2 0.5 ltlt 1 1
• WELCOME criticism of your work

6
Survival Pre-requisites 1

• Ability to look at life in a perverse (creative)
  way
• (Mark Twain) Age is an issue of mind over
  matter. If you don't mind,
  it doesn't matter !
• Patience and mental strength to negotiate the
  troughs (inevitable for every PhD)
• Trust your gut feelings in spite of negative
  results
• Luck favors the prepared mind (Pasteur/Hamming)
• Good sense of humour you will definitely need
  it!
• Ability to smile (through gritted teeth) at
  advisor

7
Survival Pre-requisites 2

• Industry experience NO!
• "Real problems" only imply short-term
  applicability
• May even be counter-productive since immediate
  reaction is to start programming, rather than
  conceptual thinking
• PhD registration ? Research Scholar
• PhD is training to do research AND actually
  doing research

X
8
VTU Faculty Specifics

• Difficult to devote time due to academic and
  familial responsibilities
• Make sure to set aside a fixed daily time slot
  for research and make this sacrosanct
• Course students can participate in the research
  work under your direction and thereby accelerate
  the progress

9
Good PhD Thesis

• Relevant to society NO!
• Relevant to industry NO!
• Right question Did I have fun thinking about
  the problem and did I devise elegant solutions
  that I am proud to show my mother (and she didn't
  find any mistakes in the proofs)?
• Doing a PhD is hard enough without the burden of
  additional expectations ...
• All good dissertations find their way into the
  real-world sooner or later

10
ENVIRONMENTAL ASPECTS
11
BAD ADVISOR CHOICE

• Knows less about the topic than you do(even at
  the beginning)
• Not visible in the research community
• Says Yes to a student without assessment
• Popular with students due to low expectations,
  not subject mastery
• Most notable achievement in last five years has
  been getting his PF transferred

12
GOOD ADVISOR CHOICE 1

• Junior Young Turk
• Great enthusiasm and involvement
• Current with research topics
• Remembers PhD tribulations
• Expects the world from you
• Micro-management of your work

13
GOOD ADVISOR CHOICE 2

• Senior Big Shot
• Broad perspective of research area
• Can advertise your work well
• Provides academic freedom to explore
• Perenially rushed for time
• Will not write your thesis

14
BAD CHOICE of FRIENDS

• ondu idly-vada / by-two coffee dost
• Unless your thesis topic is Topological sorting
  arrangements of circles and donuts
• Indian philosophy buff
• Indian philosophy is an intellectual mechanism
  for providing extremely sophisticated reasons as
  to why work cannot be done
• Lazy genius
• Will drag you into the mud with him

15
GOOD CHOICE of FRIENDS

• Technically competent and ambitious
• a healthy spirit of both cooperation and
  competition
• Willing to call a spade a spade and criticize
  you to your face
• Brings out the best research in you

16
RESEARCH COMMUNITY

• Take every opportunity to meet with the movers
  and shakers in the field
• attend local international conferences such as
  COMAD, HiPC, FSTTCS, IndoCrypt,
• attend summer schools (Yahoo, Microsoft, )
• apply for six-month internships in research labs
  after you are about half-way through your thesis
• send technical reports for comments outside
• for funding, approach academies (INAE/IASc /NASc
  /INSA) and DST/CSIR/KSCST/VTU

17
REAL WORLD

• Family
• starts calling you a visiting professor
• question your sanity with cold-blooded regularity
• Well-wishers
• take great delight in asking you Is the thesis
  done yet? Can we call you Doctor now?Do all
  PhDs take this long?

18
COMMUNICATIONASPECTS
19
WRITING and PRESENTATION IMPERATIVES

• It is your duty as a scientist to share your
  discoveries with others
• Your work is understood only from what you
  present / publish
• You are evaluated only based on what you write in
  your reports, etc.

20
Acquiring Writing Skills 1

• http//dsl.serc.iisc.ernet.in/haritsa/geninfo/tec
  hwrite.ppt (by Vikram Pudi)
• Essential for both publications and thesis
• Spelling Automated Spellcheckers
• Style Elements of Style by Strunk White
• Clarity, elegance and flow
• Technical precision is paramount
• e.g. rampant misuse of optimal and ideal
• Avoid ornate language this is not a literary
  exercise
• e.g. The linked list appeared like a chain of
  jasmine flowers

21
Acquiring Writing Skills 2

• Organization
• Modular (like programming)
• Linear narrative (no item songs ?)
• Minimize redundancy
• Choose good titles/acronyms (like variable names)
• Time Will Tell Leveraging Temporal Expressions
  in IR
• Real-time commit protocol called PROMPT (Permits
  Reading Of Modified Prepared-data for Timeliness)
• Read and revise word-by-word!
• 10 page paper usually takes a month to write
  satisfactorily

22
Acquiring Presentation Skills

• Slides should be in large font / visible colors
• Ideas should be conveyed through pictures /
  animation as far as possible
• Focus on conveying the main idea, and not every
  single detail
• Rehearse the entire talk a few times
• Memorize the inter-slide transitions
• Carry out dry-runs with your peer group

23
PUBLICATIONS

• ESSENTIAL to publish at least TWO papers before
  submitting thesis
• Progress milestones ? confidence buildup
• Honest feedback from anonymous experts
• Early warning of potential disaster situations
• Ideas for future work
• Thesis gets written/revised at a steady pace,
  not delayed to the end, when it can prove to be
  an overwhelming task
• Makes the resume speak for itself

24
PUBLICATION FORA

• Unique feature of CS The top conferences and
  journals are viewed as equivalent fora.
• Evaluating Computer Scientists and Engineers for
  Promotion and Tenure
• David Patterson (Univ. of California, Berkeley)
• Lawrence Snyder (Univ. of Washington, Seattle)
• Jeffrey Ullman (Stanford University)
• In those dimensions that count most,
  conferences are superior. Conference
  publication is both rigorous and
  prestigious.

25
CS Conference/Journal Quality

• Most reliable, comprehensive and recent are the
  rankings by Australasian CORE
• 1500 conferences and 2000 journals
• Ranks them into A, A, B, C categories
• Website http//www.core.edu.au/
• Goal should be to publish in A and A categories
  
• ICDE database conference (A) 15 accepts

26
Further Reading

• How to Get a PhD A Handbook for Students and
  Their Supervisors
• E. Phillips and D.S. Pugh, OUP
• Getting a Phd An Action Plan to Help You Manage
  Your Research, Your Supervisor and Your Project
• John Finn (Routledge Study Guides)
• Authoring a PhD How to Plan, Draft, Write and
  Finish a Doctoral Thesis or Dissertation
• Patrick Dunleavy (Palgrave Study Guides)
• http//www.phdcomics.com/

27
Take Away
28
QUESTIONS ?

29
(No Transcript)
30
END