?

Log in

No account? Create an account
Dear Theory Bloggers: Please Remember that the Public is Watching - Arvind Narayanan's journal [entries|archive|friends|userinfo]

Dear Theory Bloggers: Please Remember that the Public is Watching [Nov. 11th, 2010|10:29 pm]
Arvind Narayanan
[Tags|, , , ]

Of the subfields of Computer Science research, Theory probably has the most active blog presence. This is great, but it has had some unintended consequences — while theorists (naturally) blog for an audience of their peers, the general public watches on with considerable interest and, regrettably, confusion.[1]

The interest is no doubt due to the mythical status of the P vs NP question. But we shouldn't forget that like most mathematical conjectures, the resolution of P vs NP will mean remarkably little for the everyday lives of people (unless, of course, the unthinkable happens and it is resolved negatively.)

There are two problems with the glut of media and public interest in a problem that is of great consequence to the internal progress of the field but little significance to the outside world. The first is that due to the esoteric nature of the problem and especially the methods used, there is the risk that onlookers will lose interest in the entire field.

The second, and the one I want to talk about, is that the community loses the opportunity to tell the public about theory research that is of far broader significance. Questions that combine deep theory with real-world application are everywhere in theoretical computer science, ranging from the somewhat philosophical (can random coin flips speed up a computer?) to the heavily commercial (the theory of auctions).

I'm not asking theory bloggers to write for the general public, although a post every once in a while would be fantastic. But there are two things that can be done with little extra effort:
  1. add one or two sentences to posts about new results describing why it matters, in a language that makes sense to outsiders
  2. when contacted by the media, gently nudge them to write about research that can be understood by the public, and give them pointers to do so.
This humble suggestion was brought to you by my having to explain to some confused people that the recent separation proof by Ryan Williams is being celebrated mainly for the proof techniques and has little or no relevance to non-theorists.[2] And I still remember excited commenters on web forums telling each other how PRIMES is in P was going to speed up cryptography.

[1] When I say the public, I mean the subset of the public that is interested in science and computation.
[2] Ryan's work is by all accounts awesome, and I do not mean to detract from it in any way.
LinkReply

Comments:
[User Picture]From: geomblog
2010-11-12 06:46 am (UTC)
It's a good point. although I suspect most theory bloggers (myself included) blog for the theory community without too much regard for the outside world, mainly because it's too much work to set up the background needed. Also it's more efficient, under the premise that if someone is confused, and posts a comment, I can then clarify.
(Reply) (Thread)
[User Picture]From: arvindn
2010-11-12 07:06 am (UTC)
Yes, I can definitely sympathize with that. But then again, I took a quick sample of the follower list from the Friend Connect widget on your blog, and (as I'm guessing you're already aware) most of them are outside the theory community.

May I suggest a compromise, since people are generally reluctant to post comments on blogs they don't understand — create an About page with a few paragraphs describing your field and your work, and a pointers to some introductory reading material for people who are interested in getting more out of your blog.
(Reply) (Parent) (Thread)
[User Picture]From: hukuma
2010-11-14 04:23 am (UTC)
unless, of course, the unthinkable happens and it is resolved negatively

Even so, it may not have a big effect on the real world, depending on the non-asymptotic complexity of the reduction.
(Reply) (Thread)
[User Picture]From: arvindn
2010-11-14 04:35 am (UTC)
True.
(Reply) (Parent) (Thread)