Fortunately for Matt Yglesias, Lindsay Beyerstein only leaves him in a metaphorical smoking crater. »« Help Camp Quest Texas

Bruce Alberts, failure

This is not a very exciting video, but I might just inflict it on my cell biology students in the fall. We got a fair amount of flak from students last time around who were frustrated when labs didn’t work like a recipe from a cookbook — yet that’s how science usually proceeds, with lots of tinkering and frustration and repetition.

I also like his point about how teaching is important for science (although the students won’t really care about that.) I don’t think I really got the breadth of my discipline until I had to master it in order to teach it — there’s nothing quite like the panic behind “I’ve got to lecture for an hour on vesicle transport tomorrow!” to focus the mind wonderfully on a subject you might have found of only passing interest previously.

(via Sandwalk.)

Comments

  1. Owen says

    I still have my mid-90′s copy of that textbook. I wonder if later editions still have that photo of the authors crossing Abbey Road on the back f it?

  2. ChasCPeterson says

    We got a fair amount of flak from students last time around who were frustrated when labs didn’t work like a recipe from a cookbook — yet that’s how science usually proceeds, with lots of tinkering and frustration and repetition.

    Well, sure it does, but c’mon, that’s not one of the aspects of science that one wants to emphasize in an undergraduate lab course. Students are understandably and, in my view, rightly frustrated when lab exercises ‘don’t work’, and I doubt they learn much from it either. An opportunity to cement or extend their book-larnin’ becomes (in their eyes) a pointless exercise in bullshitting. It probably turns most students off from science, exactly the opposite of the desired result.
    so keep it simple.

  3. says

    Ah I remember when we lost 8hrs (of a 6hr lab) cause our pcr primers wouldn’t anneal and we had to come back in our spare time to redo it all 2 degrees higher the next day so we could get back on track…. good times….

  4. Arkady says

    @Owen

    The edition I had (5th I think) had a Sgt. Pepper crowd of scientists, Dolly the Sheep etc., so they’ve kept up the Beatles album theme!

  5. says

    The message isn’t “sometimes experiments don’t work, so give up.” It’s “sometimes experiments don’t work, so try again.”

  6. chigau (違う) says

    If at first you don’t succeed, try, try again. Then quit. There’s no point in being a damn fool about it.
    -W. C. Fields

  7. hyoid says

    Have you ever failed at something? Then, you are a failure. (Rays of Comforting Logic)

  8. barbara4 says

    You’re so right about the importance of teaching for learning things. The first time I actually understood meiosis was when I was a senior in college and a teaching assistant in a general biology lab. I first understood the chemistry of photosynthesis and respiration and many many math problems when preparing to teach them.

    Another way teaching is important for learning is that students often ask about things I thought I knew but obviously don’t know quite well enough. Often the question address small parts of the topic (just last night I was checking; no, tree ferns really don’t have secondary growth) but sometime I’ve come up short on basic processes of evolution and had to do some quick reading.

    Learning stuff is one of the fun things of teaching. (Watching students learn stuff is even more rewarding, but often it’s hard to tell if that is happening.)

  9. Rich Woods says

    I spent two years watching chemistry experiments fail. Whatever care and precautions myself and my two colleagues took, almost every experiment would fail. Then, eventually, we found out that two people from another group had been sabotaging our experiments as a joke. We were not happy.

    What I take away from this is a simple lesson: check that your procedure and your kit is OK, and watch the bastards around you!

  10. erick says

    I have a similar problem with the way physics problems are taught, or at least I see a gap. I remember problems that all used special parts from the physics store, like massless ropes and frictionless pullies. While it makes sense to simplify like this to make calcs manageable, there’s never any follow-through to show how the real results would be a bit different. And in labs, there wasn’t often much thought to explain why the results weren’t perfect. A minor grumble, really, but something I try to fix with my kids.

  11. erick says

    Oh yeah, and follow through on those “strange results”, don’t just disregard them as some anomaly. Sometimes that’s where the discoveries are, either in process problems or in something entirely new. Someone once said that the most prophetic words in science aren’t “Eureka”, but “hmm, that’s peculiar”.

  12. eidolon says

    I have to agree with Chas @2. One of the first things I looked at when selecting a lab for my students was did the damn thing work dependably. HS and undergrad labs are not exercises in discovery so much as they are ways for students to learn data collection and evaluation as well cementing that book stuff through experiences. In examining WHY their results do not match anticipated results, students also learn. First though, they need results that are not garbage.

  13. joe99b says

    Pretty good presentation; could have used a director to tighten things up a bit.
    Isn’t his use of theory for hypothesis non-standard? Surprised me a little.
    And why DID they release the latest edition to a puzzled General in Antarctica?

  14. Tsu Dho Nimh says

    We got a fair amount of flak from students last time around who were frustrated when labs didn’t work like a recipe from a cookbook — yet that’s how science usually proceeds, with lots of tinkering and frustration and repetition.

    ROFL: I had to take an “Incomplete” in genetics because my fruitflies had taken apparently taken a vow of celibacy … despite doing everything right, they wouldn’t get jiggy on schedule.

    The professor was cool with it – it wasn’t because I had procrastinated on the breeding crosses. the damned bugs weren’t interested in sex. So he let me repeat that lab the next semester. I think he did a short publication on the delayed sexual maturity of whatever crosses with something.

  15. yubal says

    I once spend 5 years trying to crystallize a protein. Went through > 200 purifications 50000 mg of sample 80 crystallization conditions thousands of crystals and many month of data analysis to find the one single crystal that gave us the first structure of a protein in its class. Whenever I present that people ask me “how did you do that??” And I answer “stubbornness”.

  16. says

    My junior college chemistry prof had a D.Sc. from Switzerland and years of experience in industry. He gave it all up for a slower-paced rural life, trying to teach us country kids some chem. One day a student complained about conflicting results after running an experiment twice. In response, Doc said, “We industrial chemists had a saying: You never get ambiguous results if you only run an experiment once.” When some of us were cheerfully taking him at face value, he dropped the boom on us: “Our usual standard was seven runs for each experiment. If we didn’t get five runs in close agreement, we’d re-examine our protocols and do seven runs again.” No one complained about having to redo an experiment after that.

  17. ekwhite says

    My version of Molecular Biology of the Cell in grad school was the green one in the middle of the splash page. Good times.

  18. David Marjanović says

    I had the gray edition. Haven’t watched the video.

    Then, eventually, we found out that two people from another group had been sabotaging our experiments as a joke.

    For two years!?!?!

    Christ, what assholes. I hope their asses got fired, and the rest of them too.

    I once spend 5 years trying to crystallize a protein. Went through > 200 purifications 50000 mg of sample 80 crystallization conditions thousands of crystals and many month of data analysis to find the one single crystal that gave us the first structure of a protein in its class. Whenever I present that people ask me “how did you do that??” And I answer “stubbornness”.

    Reminds me of the guy who recrystallized a thulium bromide sample 50,000 times.

    BTW, is your comma key broken? Your comment was difficult to read.