Advice to Graduate Students and Researchers John Owens

On Having a Web Page

Simple rule. If you are ever interested in publicizing yourself and your work, have a web page. That web page should have your name and your contact information (work address, work email, and perhaps phone). You might think about putting some background on it as well (your list of publications, perhaps, and your research interests). You should also consider putting a picture of yourself on the web. People will then know what you look like. This is useful if you want to meet those people and publicize yourself and your work to them.

As a new professor, I am frequently getting email from students asking for research advice, positions in my group, admission to my university. The first things I want to know are those items above. The web is a wonderful way to put all this information in one place to make it easier for the people you're trying to reach to find it. Ideally, what you want to happen is that someone can type your name into Google and have your web page come up first.

In the event your employer or university doesn't give you the opportunity to make a web page, there are many sites out there; I recommend Github Pages.

On Random but Relevant Advice From Others

Pat Hanrahan has an excellent talk on how to write a systems paper and Kayvon Fatahalian has an equally great set of tips for giving clear talks.

Olin Shivers offers dissertation advice (as well as possibly relevant advice on the thesis defense and automatic weapons, as well as the finest advising philosophy and acknowledgements on record).

David Patterson has several interesting nontechnical talks, in particular Writing Advice and How to Have a Bad Career in Research/Academia.

Jonathan Shewchuck discusses the three sins of authors in computer science and math.

Butler Lampson has several Hints for Computer System Design.

Pat Hanrahan recommends the previous work highly as well as No Silver Bullet: Essence and Accidents of Software Engineering by Fred Brooks.

Roy Levin and David Redell tell you How (and How Not) to Write a Good Systems Paper. And Alan Smith tells you how to evaluate one.

Ivan Sutherland's "Technology and Courage" is a must read for all graduate students, especially those at the beginning of their studies.

Richard Hamming's "You and Your Research" (copy 1, copy 2, copy 3) is also an excellent read.

André DeHon's How to be a Winner: Advice for students starting into research is aimed towards students doing systems research.

Joel Spolsky of the excellent Joel on Software provides some great Advice for Computer Science College Students.

PLoS has some great advice articles:

John Regehr and his colleagues explain the 5+5 Commandments of a Ph.D..

For creativity in experimental methods, Jeff Van Bueren's 2000 paper can't be beat. And Martin Schwartz makes a powerful case for the importance of stupidity in scientific research.

Jennifer Widom gives some tips for writing technical papers, and I also enjoyed Michael Ernst's thoughts on writing a technical paper. The Science of Scientific Writing discusses how to make scientific writing more readable (with great examples).

I see that the UCSD VLSI CAD Laboratory has assembled a list of resources that includes many of those above. Tao Xie and Yuan Xie maintain a incredible list of advice links. And my colleague Chen-Nee Chuah also has a terrific list of resources.

On Recommended Books

Two vital books, recommended by many and certainly by me:

On Publishing Papers on the Web

When publishing papers on the web, always create a web page associated with the paper. When you link to a paper on the web, always link to the web page associated with the paper. (When you reference a paper in another paper, of course, just reference the place it was published.)

(Why? When people look for your paper on the web, they most often don't want the whole paper. Instead, they want to see the author list, or get in touch with you, or see where it was published, or just to read the abstract. They don't want to download a 2 MB PDF file. Also, a well-constructed web page associated with a paper will index on web searches much better than a 2 MB PDF file.)

Think of that web page that you create as an advertisement. When you create that web page, it should contain the following information:

Finally, when you place a paper on the web, make sure the paper reflects where it is published. The version you put on the web will look exactly like the published paper except for this addition. Putting this in italics on the top of each page is most useful (i.e. "Appears in the 2002 Journal of Irreproducible Results"). This is for readers who print your paper, read it, put it aside (or lend it to someone else), and want to know where it came from to reference it etc.

The Robotics Institute at CMU does this right (example).


Random Stuff

Matt Might's Academic job search advice.

On asking for letters of recommendation for graduate study.


John Owens | Last updated .