PhD Completion Milestones

Dr. Laurie Williams

 

 

Here are some milestones for completion of a PhD dissertation with me as your advisor.  It is important for you to stick to this schedule to have a high-quality dissertation, to contribute valuable research to the community, to obtain feedback on your work from lots of smart reviewers, and to be as competitive as possible for the job you want upon graduation.

 

Years 3 can overlap with any of the years below if you plan to graduate in four years. Those pursuing an academic career need to think in terms of 5-6 years so that you can get as many strong publications as possible.

 

Year

Activities

Year 1

If you are considering me as your advisor, we need to get to know each other and work together in some capacity.  Come to our graduate student weekly meeting so you can meet your peers and learn about the work we do. In addition, you could be a student in a class I teach, be a teaching assistant for , or we have an independent study project that we work on together.  At the end of this we can mutually decide whether or not we feel we can have a productive research relationship.  I will not sign any plan of work to be your advisor before the completion of this activity.

 

Fill your schedule with courses*. Take your four core coursesr and at least one statistics course (STAT 511). This is a good time to take independent study with me if you are not taking a class from me. Take department seminars so you get this requirement out of the way and get a window into research.

If you want to pursue an academic career, your first and second years are good for being a TA so you can get teaching experience before needing to really focus on your research. You could also consider teaching a course over the summer.

 

It's a good idea to have an industry internship during your first and second summers to get experience in industry before needing to really focus on your research.

Year 2

Fill your schedule with courses*. Take your two 700-level classes; finish your statistics courses for your statistics.

 

Take 3 credits of CSC890 in the fall semester and 3 credits of CSC690/CSC890 in the spring semester. Your CSC890 research should focus on an early study of the area you would like to research for your dissertation or on research you are doing for your RA assignment. Plan your CSC890 research to complete early enough such that you can submit it to a conference during the spring semester.

 

Submit your CSC890 research to any of the following conferences: ACSAC, Agile, ASE, CCS, ESEM, FASE, FSE, ICSM, ISSRE, ISSTA, RE, Security & Privacy, USENIX Security or other jointly-agreed substitute. Of these, ASE, CCS, FSE, ISSTA, and RE have very competitive acceptance rates, so the others may be a better choice for your first paper: ACSAC, Agile, ESEM, ICSM, ISSRE. Not to worry if the paper is not accepted :-). You will be learning about what is expected from the community in terms of reseach rigor, salience/novelty of topic, and writing.

 

Plan to submit a paper to one of the following conferences based upon your work from spring paper time through the summer, could be on your summer internship if you have one: ICST (due October), ICSE (due August/September) or WWW (due October). Note that ICSE is very competitive so you may consider ICST (or its short paper track) or WWW for this paper.

 

If you want to pursue an academic career, your first and second years are good for being a TA so you can get teaching experience before needing to focus only on your research. You could also consider teaching a course over the summer.

 

It's a good idea to have an industry internship during your first and second summers to get experience in industry before needing to really focus on your research.

Year 3:

Submit papers according to the following schedule:

  • Early fall: Submit your paper to ICST, ICSE, or WWW per plan above.
  • Fall semester: Submit a systematic literature review on your general research topic to the Information and Software Technology (IST) journal.
  • Spring semester: Submit to any ONE (quality over quantity!) of the following conferences: ACSAC (due June); Agile (due February), ASE (due May), CCS (due April), ESEM (due March), FSE (due March), ICSM (due April), ISSRE (due May), ISSTA (due February), RE (due February). During the spring semester, there are also opportunities to write papers for ICSE workshops and/or a Fast Abstract for ISSRE to get (additional) feedback on your work and to get yourself out there talking to people in your community.
  • Spring/Summer: Plan to submit a paper to one of the following conferences based upon your work from spring paper time through the summer, could be on your summer internship if you have one: ICST, ICSE, or WWW. It is wise to strategize to hold your very best work out for submission to ICSE rather than submit to a conference in last spring.

This plan has you writing three papers/year, four months/paper**. You should be able to write really strong papers if you focus on writing three really great papers/year. Some of these may be revisions of previously-submitted papers if they had not been accepted earlier. Plan your year such that you have these three breakpoints with something to report and get feedback on.

In Year 3 or 4 (but at least one year before your graduate), submit to a doctoral symposium at a relevant conference, such as ICSE, ICST, FSE, ASE or ESEM. Through this submission (and hopeful participation) you will obtain valuable feedback on your research idea from respected researchers.

Year 4 thru (Graduation - 1) year

Same as above, except you should write for ICSE and you should write a journal paper should be to TSE, TOSEM,or EMSE (preferably) or JSS, IST. One year you should write a paper for IEEE Computer, IEEE Sofware or IEEE Security and Privacy as a means to get your research out beyond the academic audience of most conferences so that you can influence the state-of-the-practice.

 
Graduation - 1 year

1.  Write your proposal document. Your research plans must include industrial and/or open source validation of your work.

2.  Hold your proposal defense at least one calendar year prior to expecting to graduate.

3. Same publication requirements as above except you should focus on publishing in the most competitive conferences (ICSE, ASE, CCS, FSE, ISSTA, OOPSLA, RE) and journals (TSE, TOSEM).
4. Finish all data gathering analysis before your final semester!
5. Plan your job search. You will need to apply for academic positions by December of the year before your graduation. It is a good idea to get your applications in for industry positions at the same time. I will be happy to brainstorm where to apply with you, and of course to write letters of recommendation for you. We will practice you job talk in reading group.

Graduation semester

1.       Write your dissertation.  We can work on it in bits and pieces throughout the semester.  However, you must give it to me in its entirety a minimum of 3 weeks before your defense date. That allows us two weeks for finalizing the whole document prior to giving it to your committee.  You committee will take the final week to read it. 

2.       Defend your dissertation.

3.       Revise your dissertation and obtain approval from the thesis editor.

 

 

* To be my student, you must include Software Engineering (CSC510) in your core courses, and take Software Testing and Reliability (CSC712) and Software Security (CSC515) and Software Engineering as a Human Activity (CSC710). If you have already taken Software Engineering, you must take a course that has CSC510 as a prerequisite. You must also take at least two statistics class to prepare you for the validation efforts of your research (STAT 511 and STAT 512). Having an adequate statistical background is essential for the types of model development we do and for the validation of our work.

Suggested core courses

Suggested 700-level: CSC710 and CSC712 (discussed above).

Also recommended: CSC522 (Automated Learning and Data Analysis)

** A continual stream of submissions is essential for obtaining external feedback on your ideas and to copyright the progess of your research. Of course, you would love to get all your papers accepted, but please realize that periodic paper rejection is generally inevitable. You are learning and getting feedback on your work regardless of whether the paper is accepted or rejected.