Let’s overhaul the seminar!

With all the talk in the blogosphere recently on how we should overhaul the science publishing system, I started thinking about what else might be in need of an update… and, well, isn’t it time to rethink the weekly ritual of the seminar?

The research seminar as practiced today has been around for centuries. Its format goes back to the days when the primary means of disseminating scientific research were journal publications and handwritten letters exchanged between scientists – that is, of course, unless you were lucky to attend a seminar given by a visiting scientist who would tell you all about his recent research, some of it not published yet! So now you knew about it! We don’t have go go back all the way to the 19th century for this, either. The 1960s or 70s would do quite nicely.

The seminar has survived in an almost unchanged form since then: the introduction of the speaker to the audience, the lecture, thanking the speaker for the first time (also known as “the first clap”), the question and answer period, thanking the speaker again (the second clap). There’s sometimes a dinner in the evening.

But the times, they are a changin’. The internet takes care of the dissemination well enough. The lecture has since been discredited as a teaching tool and many groups of educators seek to lose it. Do we still need the seminar, then? Or should we try to reorganize it?

Here are my modest proposals.

1. Modernize the seminar taking current pedagogical research into account. Why must we stick to the outdated lecture format? Let’s make it interactive! The audience could be asked to read the introductory materials ahead of time. Upon entrance, they will be given clickers and prepared worksheets. The speaker will then guide them through a series of hands-on activities that will enable them to discover for themselves, say, the Langlands fundamental lemma. Upon completion of seminar, the participants will be able to derive fundamental lemmas, or something like it. They will also be given Math Blaster points (displayed on an internet scoreboard) that they could elect to convert to their choice of Dungeons & Dragons or World of Warcraft points.

Let’s not forget real-world applications, either. If you proved Poincare’s conjecture, that’s nice, but we also need to know how this will influence the development of the next generation of iPods. Please try to explain it to us in about 10 minutes. A multimedia presentation would be great.

But if we’re not ready to lose the lecture altogether, then here’s another option.

2. Follow the Khan Academy model. The Khan Academy videos feature – you’d never guess! – a guy lecturing. On a blackboard. With the electronic equivalent of colored chalk. In a fairly monotone voice. The explanations of math topics are very good, but not necessarily head and shoulders above what I see around here.

The difference is that Khan has found the perfect format for the lecture: a 10-minute single-topic video clip that can be watched at home. There’s no need for the lecturer to hold your attention for 50 minutes without interruptions. If you missed some part of the explanation, just rewind the clip and watch again. The instructor does not have to vary the pace, insert jokes or resort to gimmicks, or deal with classroom discipline issues. He speaks in a normal conversational voice, without having to project across a large classroom. He does not have to struggle in class with the AV and IT equipment, generally at the development stage of MP3 players before the iPod. He does have to set up the software and recording equipment, record, edit and upload the video, but all this happens off screen. All we ever see him do is explain the math, simply and naturally.

So… why wouldn’t we follow suit? Instead of travelling to conferences and seminars (think TSA screening lines, or the middle seat with broken in-flight entertainment system on a transatlantic flight) we could simply record a series of videos explaining our work and post them on YouTube. We could even upload them to a central, Math Overflow type website that would award Math Blaster points to logged-in users.

There could be problems, naturally. For example, not everyone was born with the gift of the golden voice. Some of us have less than perfect enunciation or even a foreign accent. Would we start hiring voice-trained actors for our videos, and would that be an NSERC-eligible expense? Come to think of it, why not also hire someone better looking, given that we (women) are constantly judged on our appearance even in contexts that should have nothing to do with it? Or why not just hire a male actor to narrate my presentations, in a Remington Steele fashion? Personally, I’d love to hear James Earl Jones explain this paper to a YouTube audience.

Disclaimer. Since apparently a lot of people have trouble recognizing irony when it’s being employed by a woman: no, I’m not actually making these proposals seriously. The YouTube videos might be worth trying for those so inclined, but I’d never want it to become a de facto professional obligation. There are legitimate reasons for wanting to avoid YouTube celebrity.

I’m aware that some of our conference lectures get videotaped and then posted on institute web pages. There might even be a few of mine around. In terms of effectiveness, it’s taking a presentation in one format and converting it to another one, with a lot of compatibility issues along the way. Also, the videos don’t tend to propagate and generate notoriety in the same way as YouTube clips, or at least I have not seen that happen.

I do think that there are points to be made by juxtaposing the way we learn our own trade with the way we propose to teach it to others.



Filed under academia, mathematics: general

10 responses to “Let’s overhaul the seminar!

  1. Ray

    One huge advantage of the video lecture over the traditional format, is that you can pause it, rewind, rewatch, etc., as necessary. I find this aspect enormously helpful – if I don’t understand something, I can put the lecture on hold while I figure it out, instead of desperately trying to keep up and mindlessly copying what the speaker says to my notebook.

    I think of the videotaped lectures at the MSRI website as one of the greatest resources around, as far as learning math beyond the topics covered in well-worn textbooks goes.

  2. Agreed. That alone can make up for many deficiencies.

  3. Also, may one can ask questions and tag it to the point in the lecture where one seeks a clarification. At a later point the speaker (or some one else) can reply to the queries. Something akin to the comment feature in overflow, only thing is it would be tagged to a video.

  4. Yes, that would be neat. That way the lecture would not be interrupted, but the question would still be attached to the point in the lecture where it arose, and the person who asked could rewind to that point before listening to the response. It would be nice, too, to be able to choose to make your question public or private. (Is it just something of interest to me, or should I ask about it for everyone’s benefit?)

  5. Interesting post! Today I wrote about a prototype for a new web-based course/seminar platform developed by my colleague Dror Bar-Natan. Here is the link:
    Dror’s system allows for video playback synchronized with blackboard photographs. Comments by the remote audience can be made on every video frame.

  6. Kevin O'Bryant

    Sad to see a witty post with such humorless replies…so let me add one more.

    Israel Gelfand’s seminars were qualitatively different from the usual ones. Gelfand would insist on understanding everything, and if he didn’t he wouldn’t allow the lecture to continue. Sometimes, he would ask if anyone else understood the speaker, and if someone said they did, the speaker (even if a visiting speaker) would be deposed and the claiming-to-have-understood audience member would go to the board.

  7. I’m thinking some of the point about the conference seminar isn’t necessarily to get points A through F across. After all, everybody there has a laptop and most get journal access though their institutional affiliations. I think the real agenda is peer group identification, status and support.

    You may be the only person at your institution working on X or who even cares much about it. Maybe a few people will read your paper but not really be interested enough to go into detail. At conferences people climbing up get affirmation that this is important to some peer group and that should they succeed they get community acclaim. The person/group who busted their tails for years in turn get strokes from a community that knows what the heck they’re talking about.

  8. John — I agree with you. I’ve put a lot of effort into building up my group here and maintaining the harmonic analysis community, so I know the feeling. I was being ironic when I said we should abolish conferences. I wrote this post because I was getting tired of editorials and whatnot pointing to Perfect Solutions to all our education problems. We’ll all teach online classes! No, we’ll have interactive discussion classes! No, we need more technology! No, we need to publish teacher performance data! I probably should have linked to a couple. My actual view on the subject is that all of the above have a place, and let’s not discard old-fashioned methods just because they’re old fashioned. Some things will never be outdated, like for example telling stories, singing songs, getting together, or one person explaining something to another.

  9. Izabella_ I never actually did get the B.S. I was working on in part due to medical difficulties. Another major part however was it was near impossible to actually have a discussion about what made the material interesting. I was attending classes with a bunch of stenographers who never blinked when a professor blitted out some outrageous statement. The person at the front of the room could have been speaking in tongues for all they cared as long as they got the grade that would promote them to the next level.

    If we’re going to bring back and exalt curiosity and creativity instead of trying to internally duplicate a server stack on Google then we might want to talk to real human beings.

  10. If we’re going to bring back and exalt curiosity and creativity instead of trying to internally duplicate a server stack on Google then we might want to talk to real human beings.

    Yes, a thousand times yes.