Thursday, September 29, 2005

Cocktail Conversations

I once met a professor at Chicago that would say "My business is war, and business is good." I have a food scientist friend from college who did his doctorate on starch and had a catch phrase "Everything you eat is healthy, safe and nutritious." But when I start having a conversation with non-scientists it often goes like this:
  • Them: "What do you do?"
  • Me: "I'm a professor at the University of Chicago."
  • "That's neat. What do you teach?"
  • "Computer Science."
  • (a) "Oh. Excuse me, I see someone I know," or
    (b) "I'm setting up a wireless network in my house.", or
    (c) "I don't use the computers much but my kids are really into it."
I'm sure many of you have had similar experiences. On occasion they will ask me about my research and I will regale them with stories about traveling salesman and Arthur and Merlin, but that rarely goes far. It could be worse, I might have done my research on Hopf algebras.

How do you discuss your research with non-specialists? I'm sure some of you solve this problem by not having any non-CS/Math friends. But if we can't easily discuss our work one-on-one how do we convince the public that our research is important to them and society at large.


  1. What is Hopf Algebra?

  2. Actually, I have had numerous similar conversations with scientists from other fileds (ph.d's from other fields or doctors). TSP is the most
    frequently used example, then come other easily describable optimization problems. Most of them can be easily related to cost/profit.

  3. Many people know natural numbers, addition, and multiplication. So, you might easily explain what a prime factorization is. But you can't explain the computational problem of factoring primes unless your conversational partner knows what an efficient algorithm is. I think one can't understand but only get used to algorithms. Essential TCS should be taught in school therefore. Some of these communication problems would disappear.

  4. I usually talk about P vs NP. It's simple to explain at a high level ("What if you could prove that lots of useful problems can't be solved quickly by a computer?"), even scientific people often haven't heard the problem stated precisely, and there are lots of cool philosophical and societal ramifications.

  5. One typical thing I use is
    shortest path algorithms using
    mapquest/google as examples.
    I used to use TSP before the
    web stuff took off.

  6. When I tell people I do computer science I usually get "whoa, there's lots of money in that field," so for a while I switched to telling people I did math, but I'd get the reply "good, I could use a new accountant," or "that's nice, what are you going to do with that?" Now I try to talk about P and NP on a high level as in the previous post. TSP is an easy place to start since people seem to relate to the problem set up. Thanks to those Citibank commercials most people are afraid of identity theft so cryptography is also something the non-technical tend to take some interest in.

  7. I mostly end up talking about the work I do on radio frequency identification privacy. This tends to go OK, since in many cases people have heard about WalMart or the US Department of Defense applying RFID technology. There are also colorful (and true) stories about using RFID to track children, RFID in passports, etc. If I am lucky, I can get to the point where I can set up some of the crypto problems involved. I never get to the solutions.

    Occasionally I will try to explain the work I've done on new types of digital signatures. This tends to go less well. I end up having to get across the idea of a digital signature first, and by that time we're out of the polite answer territory.

  8. I usually have to end up telling people quickly that a PhD program in CS is "much more like mathematics." My own area isn't as close as computational complexity is, but it's a good proxy for "no, I don't know how to set up your network."

    For my own research I explain to people "my goal is to make software more like computers. You can put in all of these different parts from different companies, yet they are compatible and work together." I also talk about cars, and how an air-conditioner designed for a particular car might need to be L-shaped or T-shaped, because it needs to exist in relation to the other parts of a car; and how that relates to software development problems.

  9. Here's another idea - you could rehearse
    the portrait of an ideal mathematician.
    Except in first person. (See also the fictional exchange contained therein.) Probably not recommended unless you are discussing with people you already know well...

  10. If you did Hopf Algebras, wouldn't you have an easy segue into quantum physics, which everyone likes to hear about?

  11. I usually avoid P versus NP to start. TSP is a particularly bad example since it is so easy in practice. (After all, the optimal TSP is known between all the county seats in the lower 48 states and there are good practical (and sometimes theoretical) algorithms to get very close in other circumstances.

    People usually are interested when I just throw out the term "Computational Complexity". After giving a one sentence definition in which I make sure to include the words "mathematical model" to lay the ground for warding off the "how do I use my PC?" questions, I usually mention crypto, data structure/DB/web search applications.

  12. When people ask me what I do for a living, I say I'm an engineer. If pressed, I'll admit to working in the software industry; nine times out of ten, the conversation moves on from there, and I never have to say what it is I really do. If I get trapped, I give in and try to explain that my real concern is how to tell that software is correct, which usually leads to an explanation of the limits of formal verification a la _Godel, Escher, Bach_. It's only indirectly related to what I really do, but it does avoid the conversation about why there's so much lousy software in the world.

  13. # Them: "What do you do?"
    # Me: "I'm a POSTDOC (;)) at the University of Chicago, no! Geneva I mean..."
    # "That's neat. What do you teach?"
    # "Computer Science."
    # "I don't use the computers much but my kids are really into it."
    # Computer science is no more the science of computers than astronomy is the science of telescopes [citation from E. Dijkstra]

    But we should definitely stop using the term COMPUTER science. Did we discuss this lately???


  14. Olivier: Last time it came up I mentioned I liked SICP's term "procedural epistemology" instead of computer science.

  15. Last weekend I had to explain to one software engineer that I am a PhD student in computer science. He asked what I study so I replied "algorithms ... and graph theory..." which basicly resulted a blank stare as was expected. Then I hurriedly explained that in our lab we study speaker recognition as a pattern recognition problem. That saved the situation (as I happen to know a little bit of the topic B-).

    Anyway, I hate that for general public your research always has to be motivated by applications.

    By the way this is my first post in weblog and I have to say that I really enjoy reading Lances posts and comments from all of you guys.

  16. Lots of hand-waving and yelling about how encryption-breaking won the war...

    Using real-world examples, like FedEx shipping (TS), web site/email security (Alice/Bob), and coloring a map of the world, goes a long way. It's stuff that they understand and _can get excited about_.

  17. just play the Pet Shop Boys Opportunities (Let's Make Lots of Money) song as your theme when anyone asks what you do

  18. once a taxi driver after asking me what i'm doing, said that he has a broken pc at home. I said that unfortunately i cann't help. so how can you do a research? ok, i said, how can you drive a car without knowing how to fix it?