Technical Presentation

Summary: Working in small teams, you will present a technical topic (e.g., a network protocol) not otherwise covered in this class.

Contents:


Introduction

By the end of the course, you will have studied key problems that have shaped the development of the Internet, as well as many protocols and ideas that address those problems. You will apply what you have learned by teaching a new topic to the class. 

Specifically, you will present either the findings of a research paper or an Internet protocol defined by an IETF RFC. I may suggest additional sources to provide context or clarification.

My goals are for you to
Because the end of our semester has become compressed, I am also seeing what you can prepare with relatively little time to do so, and I understand that is the situation.

Topics

Here is a selection of primary sources, organized roughly according to their correlation with the timeline of our course.  Your team should rank its top 4 choices.

Randolph Y. Wang, et al. Turning the Postal System into a Generic Digital Communication Mechanism. Proc. of ACM SIGCOMM 2004. August 2004. Retrieved April 23, 2007 from http://dsh.cs.washington.edu:8000/distance/sigcomm04.pdf. [Unlike RFC 1149, this is serious.]

S. Floyd, et al. Quick-Start for TCP and IP. IETF RFC 4782, January 2007. Retrieved April 23, 2007 from http://www.faqs.org/rfcs/rfc4782.html.

Dina Katabi, Mark Handley, and Charles Rohrs. Internet Congestion Control for High Bandwidth-Delay Product Networks. SIGCOMM 2002. Retrieved April 23, 2007 from http://nms.lcs.mit.edu/~dina/xcp.ps.

Craig Partridge. A 50-GB/s IP Router. IEEE/ACM Transactions on Networking, 6(3):237-248, Jun. 1998. Retrieved April 23, 2007 from http://www.ir.bbn.com/projects/mgr/50G.pdf.

N. Haller, et al. A One-Time Password System. IETF RFC 2289, February 1998. Retrieved April 23, 2007 from http://www.faqs.org/rfcs/rfc2289.html.

Anirudh Ramachandran and Nick Feamster. Understanding the Network-Level Behavior of Spammers. SIGCOMM 2006. Retrieved April 23, 2007 from http://www.cc.gatech.edu/~feamster/papers/p396-ramachandran.pdf.

Michael G. Reed, Paul F. Syverson, and David M. Goldschlag. Anonymous Connections and Onion Routing. JSAC 1998. Retrieved April 23, 2007 from http://www.onion-router.net/Publications/JSAC-1998.pdf. [CM, Heather, Ian]

Steven J. Murdoch. Hot or Not: Revealing Hidden Services by their Clock Skew. CCS 2006. Retrieved April 23, 2007 from http://www.cl.cam.ac.uk/~sjm217/papers/ccs06hotornot.pdf. [Mark, Elijah, Nick]


Mechanics

Collaboration and academic honesty: You will work in teams of 2-3. This team should be different from your project team. Each team is required to make an appointment with me to discuss the primary source a few days before the presentation.

Even if the author's talk slides are available on the web, you may not use their slides. This is partly for reasons of academic honesty, but mainly because you will give a better talk if you decide for yourself how to present the material. (I learned this the hard way.) I also want to see how you interpret and evaluate the work.

You may, of course, use figures from the paper (or from the author's slides) with proper citation.

Deadlines: You must choose a team and rank your top 4 preferred topics by Friday, April 27. Technical presentations will take place during our class meetings on May 7, 9, and possibly 11; additional deliverables will be due at the same time. No late work will be accepted.

Presentation guidelines

  1. The length of the presentation is 20 minutes.  This is realistic for a technical conference presentation; it means you will need to focus on the most important aspects of the paper.

  2. A short question/answer session will follow the presentation.

  3. Visual aids in the form of OpenOffice Impress (or Powerpoint) slides are appropriate for a technical presentation. Remember, these visual aid which should complement what you plan to say.

  4. You should plan to use the classroom computer/projector for your presentation.

  5. Each member of your team should speak. (Depending on how you split up the presentation, one member may have a longer speaking time than the others, which is okay with me.)

  6. Your presentation should focus on the paper assigned to your team.

  7. The intended audience for the presentation consists of upper-division computer science students who have basic networking knowledge. You may assume the audience is familiar with topics covered during class and in our readings. Assume the presentation is being conducted at a technical conference and it is your job to motivate the research problem, describe the solution, and evaluate the results.

Grading: Your entire team will receive a group presentation grade. I reserve the right to adjust an individual's grade up or down depending on the individual's contribution to the team. The presentation is worth a total of 100 points and counts as 10% of your overall grade in the course. 

The presentation itself is worth 80 points. 10 points will be earned by completing presentation evaluations of the other teams. The remaining 10 points will be earned by submitting 2 potential final exam questions. 

Each of the following categories are worth up to 20 points of the presentation grade:

Deliverables: 

  1. Email me your slides for the presentation within 24 hours after your presentation (or before, if you have them ready).

  2. Write 2 potential final exam questions based on your presentation. These should be short-answer questions, to which a good answer is a few words to a few sentences. The questions should be answerable by audience members who pay attention during your presentation. These questions, in verbatim and/or altered form, may appear on the final exam. Questions are due (in hard copy form) at the beginning of lecture the day of your presentation.

  3. For each presentation that you attend, complete and turn in an evaluation form at the end of the class period.


Suggestions for your presentation

  1. A typical structure for a technical presentation includes the following sections: Introduction, Problem Statement, Solution, Results, Evalution, and Conclusions.

  2. I strongly recommend that a single person presents a contiguous chunk of the presentation. In other words, Person A should not talk for 5 minutes, then go to Person B, then go back to person A.

  3. The first speaker usually introduces the entire set of presenters at the beginning of the presentation.

  4. Read your paper several times to gain an understanding of the research methods and results. Your meeting with me is a good time to ask questions.

  5. You may need to read articles in the references section of the paper to understand the context. Ask me if you are not able to obtain the paper through the library. Also, the web is a great resource for finding other articles related to your paper. Note that many authors place their peer-reviewed papers on their websites; but note also that much of what is "published" on the web is not peer-reviewed. 

  6. You will not be able to present every detail in the paper in 20 minutes---NOT EVEN CLOSE. Choose the most important parts of the paper and give them quality time in the presentation. I will reward quality of topic coverage more than quantity of topics covered.

  7. Practice. It will be clear to the audience if your group has rehearsed the presentation. Practice the presentation within your team, practice it with your friends, practice it to a wall.  You should also practice to make sure your presentation is about 20 minutes in length. Practice your presentation with your slides.

  8. In general, people tend to speak more quickly in front of an audience, so keep that in mind when you determine the overall timing of your presentation.

  9. Have fun and relax while you are speaking. (I know this may be difficult. I will not be expecting the same level of presentation as I would see at a professional meeting.)


Acknowledgments

This assignment draws on a similar assignment given by Tammy VanDeGrift at the University of Portland.

Janet Davis (davisjan@cs.grinnell.edu)

Created January 19, 2007
Last revised April 23, 2007