Pondering our Kochlab graduate student compass...how about "Always Contribute?"

Last weekend, my friend Richard Yeh posted a couple essays by Paul Graham onto Facebook. I loved the essays and linked to one of them on friendfeed. Michael Nielsen, in turn pointed me to another essay by Graham that he thought I'd like, "How to Do What You Love." Michael was completely right, I loved the essay. If you have not read that essay, I command you to stop reading my blog and to go read that article! You'll get much more out of his essay than this blog.

OK, now that I have you defiantly reading my blog, intent on garnering something useful from it, to spite me, let me continue. The "Do What You Love" essay resonated with me very strongly. It reminded me of the discussion of talents in "First Break All the Rules" by Buckingham and Coffman. I think Graham and Buckingham and Coffman are talking about the same thing: that finding work you love is a key to happiness (and productivity), but that finding out what you love is a very difficult task worth working very hard on. The language of Buckingham and Coffman is to talk about finding one's "talents." I've been talking with my graduate students about this a lot for the past six months. (In fact, it's time for me to have another awkward talent-finding session with them, I do believe!) I also preach to all of my undergraduate students about the importance of finding their talents and I give them an end-of-semester assignment to think about their talents. I'm delighted to have been shown the Graham essay, because I think it is yet another way of presenting this argument to students, and a very eloquent one.

Since I loved the essay so much, I sent it to the person who gave me the First Break All the Rules book. He wrote back to me and keyed in on the "always produce" part of the Graham essay:

"Always produce" is also a heuristic for finding the work you love. If you subject yourself to that constraint, it will automatically push you away from things you think you're supposed to work on, toward things you actually like. "Always produce" will discover your life's work the way water, with the aid of gravity, finds the hole in your roof.

While reading the note, the Do What You Love essay finally clicked with another I read by Graham last weekend, "How to Make Wealth."[1] It's another fantastic essay that I feel like commanding you to read. One premise in that essay is that people in start-up companies can be 20-30 times more productive than they can in an ordinary 9 to 5 job. Thus, a small group of people can create a tremendous amount of wealth by working really hard for a few years. They can also get financially rich as a reward for their production of wealth for the world. The thing that clicked for me is that you cannot make the world a better place without producing. Most people are producing at a rate at least 20 times less than they could be producing, if they found what they loved and were able to do it all the time. I garner great optimism from this fact that we're on average so incredibly inefficient. It means most people are not even close to any absolute point of diminishing returns, and with the right kinds of changes, they could easily multiply their productivity and impact on the world by manyfold.

So, then I started thinking about our research lab and the students in our lab. I thought over things that each student has done in the past year that made me profoundly happy. As I thought over all these things, I realized they had a common theme: I was recalling instances of those students being unusually productive. Furthermore, my favorite recollections involved those where the students had shared their work on our public wiki, or our blog, or in some other fashion open to the world. This made me think that it is now very easy for me to summarize my main expectation and goal for my students: "always produce," borrowed from Paul Graham, of course. My students like to make Kochlab slogans, so I thought of "Kochlab: Produce" or "Kochlab: Always Produce," but if you pronounce the "Koch" correctly ("Cook"), then it has the problem of making one think of produce the noun, e.g. apples and bananas. Thus, I am thinking something like "Kochlab: Contribute" or "Kochlab: Always Contribute."

In some sense, "always contribute" means the same thing as "always produce." The point of the producing, in regards to Graham's essays is that you're creating wealth, and therefore contributing something to society. However, I like "contribute" much better, because it has much more clarity in the science world. "Contribute" automatically points the way towards open science (aka Science 2.0). Whereas, production in the traditional scientific world ("closed science") can be done with a very limited amount of contribution.

I have mulled it over for a couple days now, and I think I really like this as the main piece of advice and constant guidance to give to our students: "always contribute." Does this work? Let me try it out in a few ways:

1. By the time the students get their Ph.D.s, I want them to have learned a tremendous amount about what their talents are. I want them to clearly see what the next step in their career should be in order to leverage those talents and help them be successful and happy. A compass of "always contribute" will lead the students towards finding ways of being productive instead of spinning their wheels. These activities will be the means by which the students and I discover what their talents are. This is the point of Graham's "always produce" advice. Check.

2. By the time the students get their Ph.D.s, I want them to have a strong and large professional network of people that know them and the work they have done. "Always contribute" tells them that Open Notebook Science is a good thing to do. Sharing code, design drawings, personal summaries of research papers, tips and tricks on protocols -- these are all ways to contribute. In our limited experience in our lab, we have received validation after validation after validation that open contributions get attention. We can see this vaguely via page views or Google search rank or quite vividly via positive feedback from people that we admire and people whom we've helped. Combined with traditional publishing (also a contribution) and attending scientific meetings (contributing), I think "always contribute" will make building a powerful professional network almost automatic. Check.

3. I want our lab to produce innovative, exciting, and high-impact scientific results. Will "always contribute" point us in the right direction for this goal? Does it point in any direction? I need to think about this one some more. I feel like it must point in the right direction--for example, innovations are contributions. But there's some risk that focusing on contributing could lead towards a lack in overall production. Basically I am thinking of the standard arguments against open science -- increased likelihood of scooping, which in turn reduces chances of funding and publishing. I fundamentally believe that those arguments are strong enough to tip the balance, but I don't think they've been proven yet. Another example of how "always contribute" may be counter to our lab's scientific productivity: some students may discover that they are wickedly talented at contributing in ways that do not advance their research projects. That's a great thing to discover! An example that hasn't happened in our lab yet would be for a student to discover that they're fantastically talented at writing popular science articles and want to do so at the expense of doing any research. I want students to discover something like this. It fits perfectly with items 1 and 2 above. However, it is clearly a problem in regards to item 3. This is not a new problem, though. My job as a research professor is to both mentor students as well as ensure production of research results. The way the system is set up those goals are not always aligned, and sometimes in conflict. It's possible to be rewarded with research grants, even by abandoning the best interests of your graduate students. Most people in the system know this and have seen the devastating results it has for too many Ph.D. students. I am absolutely against doing that and despise many people who have chosen that route. On the other hand, I don't have a good idea about what do do if "always contribute" turns into "I can't do my research." That's definitely going to happen eventually. In many cases, it will be possible for the student to discover their true calling in life, but then re-focus on making the research contributions necessary to finish the Ph.D. that they've invested so much time in. Will there come a time when the student should rightly choose abandoning the Ph.D.? Ugh, this is a tough one: Item #3 gets an: almost check / need more thinking.

I've painted myself into a corner now. If I were Paul Graham, I would figure out a way to backtrack. But I'm not, and I don't really know how to end this blog post, so I think I'm going to end it by linking to another Graham essay about writing essays. This one was linked to me by Kartik Agaram on friendfeed. It's an essay that explains why high school and college writing assignments sucked so badly. If you hated those assignments but never quite knew why, you'll love this story. Plus, you'll feel vindicated and it will give you one more reason to trust your gut in the future. For example, if your gut were telling you that "always contribute" is a fantastic compass to present your graduate student mentees.

Footnote:
[1] I'm taking a bunch of liberty here with my own story. It wasn't until I started writing this blog that I realized the two essays had clicked. But I think subconsciously this is what was happening. Also, I probably had the Gin, Television, and Social Surplus essay by Clay Shirky in my head, as Joelle Nebbe had linked to it recently.

My first PLoS comment: High rating of an article on TSLP being the cytokine link between eczema and asthma

5/27/2009 SJK Note: After I wrote this, Bora Zivkovic sent me links to the PLoS community blog where he talks about commenting and rating PLoS articles. Both are very much worth reading! Bora is the Online Discussion Expert for PLoS.

Recently, William Gunn Mr. Gunn composed an excellent article discussing online identity and the making of public comments in scientific circles. Without immediately spiraling into a stream of ridiculous conversation, I can't really comment on his post, or the ensuing friendfeed thread. Suffice to say that Mr. Gunn and others on friendfeed inspired me to be a lot bolder in commenting on PLoS articles.

So, tonight I made my first comment on a PLoS article. Previously, I had viewed commenting on the actual article site as a very formal procedure that required attaining the highest level of understanding of the article before submitting a comment. Essentially, I was viewing commenting on an online article the same way I viewed submitting an official comment to an article published in Science or Nature (or other journals). Published comments in those journals are almost always refutations of the article that seemingly without fail lead to concomitantly published rebuttals by the original article authors. Thus, the culture of commenting on articles is fraught with nastiness and putting one's scientific reputation on the line. This could be the reason that so far "official" online commenting on peer-reviewed articles has been very limited, whereas "unofficial" or off-site commenting has been more common. By "unofficial," I am loosely referring to comments made anywhere that is at least one link removed from the actual published article site. For example, an external blog, friendfeed discussion, or notes left on article managing services such as citeulike.

It occurred to me while laughing and crying my way through the recent friendfeed discussions (OK, fine, here's a link to perpetuate the madness) that this culture may be relatively easy to change. (Aside from any questions of whether it's necessary to change.) In my opinion, PLoS has already made one innovation that vastly increases the odds of a user making a public "comment." They have separated the article ratings into three categories: Insight, Reliability, and Style. From my personal experience, that opens the door almost all the way in terms of inviting some kind of reader feedback. Rating an article on "Style" does not carry much professional risk from my viewpoint. Rating on "Insight" requires understanding of the possible impact of the article, and is thus much more weighty than the "Style" rating. However, I personally feel I can rate an article on "Insight" without assessing the quality or reliability of the methods and data. I recently did this with a PLoS ONE article I saw on single-cell sequencing of uncultured organisms. To rate an article on "Reliability," I feel requires the kind of in-depth understanding that would be required for me to send a formal letter into the editor of Science or Nature that could be published. Thus, the barrier for me to rate on "Reliability" is quite high. Especially since if I'm going to put in enough effort to feel completely justified in rating, it's likely to be less than a 5-star rating. (I guess I'm feeling like I spend more time reading articles that I disbelieve than those I do believe?)

Another reason that placing online comments does not have to be as formal and negative as with traditional published comments is that the comments are published without a delay waiting for the original authors to compose a response. This then reduces the expectation that the publishing authors must respond and therefore takes the formality down a bunch of notches in my opinion. Also, in terms of PLoS the whole mission of the journal is to make research more broadly and rapidly available--and thus I think there is an expectation that the comments should also come from a broader base of readers.

So, that is what inspired me to take the time to read a PLoS Biology article and compose my first online comment tonight. I was also inspired by the belief that we're still very early in the process of dictating the culture of online discussions of peer-reviewed research--and thus a concerted effort can make impact in what ends up happening. This inspiration was combined with the coincidence that my wife sent me an article from BabyCenter today that caught my interest because it was discussing the recent PLoS Biology article. Finally, the thing that finally tipped the balance and convinced me to take the leap and make my first PLoS comment was a healthy dose of "WTF" So I stopped worrying and took the leap. :)

Time for more blogging! Warming up... You can't believe what you see...

I'm just now coming out of the end-of-semester fog.  I've been through three academic years so far as an assistant professor.  That's six semesters, five in which I've taught a course.  In all five of those semesters, I ran out of steam and could not keep up with all the things I'd have liked to have done in my areas of research, mentoring, teaching, and family.  In my opinion, 16 weeks is too long for a semester...I notice myself and my students beginning to burn out after 8 weeks. 

It's a pattern for me that I take on too much at the beginnings of semesters and then have to cast things aside as I get overwhelmed.  Before this semester I took a big leap into communicating with scientists on the internet, via blogging and discussions with a bunch of new friends on friendfeed.  As the semester engulfed me, I ended up casting aside blogging, but actually was able to maintain a lot of dialogue on friendfeed (e.g. in the Science 2.0 and The Life Scientists room).  So, even though I was disappointed to not keep up the blogging, I can look back and see that overall I made a huge amount of progress in terms of scientific communication and meeting (virtually) many great people around the globe.  I'm very happy with that, and I'm even happier that a couple of my graduate students came along with me.  They have made their own connections with other scientists and made substantial progress in open science and open scientific communication.

I've been excited for the last week or two to ramp up the blogs again.  In particular, I'm excited about a couple things.  One is to try out Research Blogging.  This was suggested to me by Michael Nielsen in a friendfeed thread in which I learned a lot, but which I started by spouting off way too ignorantly.  My apologies to Richard once again!  As I understand it, the service will allow me to write up blog reviews of specific research papers and then label my posts as suitable for listing in Research Blogging.  For me, it will be a step up from what I started doing this semester, which is trying to make a few notes on every paper that I add to citeulike.  You can see my citeulike RSS feed, with my comments added, on this Yahoo pipe.  It will be a lot more work to write what I think is a worthy blog report about a paper, but I'm excited about testing the waters.  The way citeulike is set up, I feel like my comments there are pretty much wasted as far as benefit to others goes.  In the future, I'm expecting my group to communicate more via citeulike (or another service) as a form of "journal watch."  But as it stands now, I'm pretty sure nobody reads the comments I add to my citeulike library.

The other thing that should be exciting is to write a guest blog for Lisa Green at NextBio.  We've talked about this a bit, and I think it may happen in the next few weeks.  What I will blog about that is worthy of a "guest blog," I don't know...but it should be a fun experience!

Finally, I needed some inspiration to come in here and start dusting off my blogs.  I had been a bit depressed at their dormancy, which was a postive feedback loop preventing me from blogging.  A blog warm-up idea came to me earlier tonight as I was staring at the ceiling fan.  I recalled something I'd noticed maybe 6-10 years ago in graduate school, which made me recall something else I'd noticed at the same time.  They're two "illusion" kind of things that I think are fun, and which I'm going to describe here without actually researching them scientifically.  Hopefully someone who knows something will pipe in and tell us something about them!

1. Blurry motion seems slower with peripheral vision

I first noticed this illusion when I was trying to make large graphs in Origin back in graduate school.  Whenever I made the mistaking of clicking on very large matrix plots, the graph would flash for like a minute before opening the graphing preferences window.  I would look aside in frustration each time.  And then I noticed that when I looked at the flashing graph (which was more like quick vertical scrolling of a dark patch), it seemed to scroll much slower in my peripheral vision.  A few years later, I noticed the same thing when looking at a ceiling fan, an experiment much easier to reproduce.

Here is the experiment you can try

Find a ceiling fan that's not too far away, and spinning at about 2 hertz.  When you look at it directly, if you're like me, the default image is to ignore the individual fan blades, and perceive a blur.  You can change the image dramatically by following an individual blade with your eyes.  For me, that motion seems "slower" than the blurry motion, but it's not the illusion I'm talking about here.  The illusion that I see is when you quickly switch from looking at the fan with your central vision and instead use your peripheral vision.  When doing that, the motion seems to slow down substantially.  For me, it is substantial and repeatable.

I couldn't resist checking into this a bit on wikipedia, and I found something called "flicker fusion threshold" in this wikipedia article on peripheral vision.  In the flicker fusion article, it is said, "so flicker can be sensed in peripheral vision at higher frequencies than in foveal vision."  Given this, I wonder if the effect I am seeing has to do with some part of the visual system normalizing a given flicker relative to the maximum possible perceived flicker?  The fan produces a constant rate of flicker...but it is a larger percentage of the fastest possible flicker when looked at with foveal vision?

2.  I think you can hear individual splashes in the roar of a waterfall.
(Photo by Ant J on Flickr.)
This is an experiment I can't replicate easily now that I live in Albuquerque.  We do have waterfalls in the mountains, but it's nothing like when I lived in Ithaca, NY during graduate school.  Almost every day, I would cross the bridge over the waterfall at the end of Beebe Lake.  This was on my way between Clark Hall (the physics building) and A-Lot, the parking lot the bus dropped off at. I would often stop and stare at the falling water. I noticed one day that if I followed with my eyes a particular part of the broken stream from the lake to the point at which it hit the rocks, I could audibly hear it "splash" within the "blurry" roar of the waterfall.

Here is the experiment you can try

If you have access to a waterfall, this experiment isn't too difficult to try.  As I mentioned above in item 1, you can track the blades of a ceiling fan with your eyes, and this is the same thing you need to do with the waterfall.  Track a "patch" of water all the way from the point at which it starts falling to when it hits the rocks or water below.  Keep doing this repeatedly in a cycle, and you should "hear" the individual splashes--or at least I do.  It's also fascinating to just look at the water as it breaks into pieces on the way down.

I just attempted 30 seconds of lazy google research for this phenomenon and was not successful at uncovering a wikipedia article to give insight into this effect.  I am 50% sure it's an auditory illusion, and the other 50% of me thinks, "why not...maybe the sound isn't as 'blurry' as it seems?"  It would be possible to check this with an objective audio/visual recording system.

3. The Blue Field Entoptic Effect -- Mystery Solved!

Also known as Scheerer's phenomenon.  As opposed to the above two illusions, this one I now know what it is.  I first noticed it on airplanes when flying in a really bright sky.  In Albuquerque, we have bright blue skies frequently, and I can see the effect.  According to wikipedia, most people can see what I see in these conditions: counteless point-like bright white things that travel in squiggly paths in the field of vision.  It turns out that these are white blood cells flowing through the capillaries that cover the non-foveal parts of the retina.  On a blue background (e.g. the sky), those capillaries produce very dark lines all across the field of vision--due to red blood cells absorbing blue light very well.  Somewhere in the visual processing system, these dark lines are "edited" out, so we don't perceive them.  However, when a white blood cell travels through, it is mostly transparent, and the increased light is perceived as a tiny white thing in the field of view (which it is, I guess).  One of the most fascinating things to me is that it allows you to actually visualize your own blood cells flowing.  According to the article, some doctors have tried to leverage this as a diagnostic technique.

How to try out the experiment:

This one is easy.  Wait for a nice sunny day.  Pick a big blue patch of the sky and stare at it for a couple minutes.  Keep your attention focused on looking for bright dots appearing and traveling in squiggly paths.  You won't be able to follow individual ones, but by trying to use your peripheral vision,  you can see hundreds or thousands of them.  They are quite different than "floaters."  They are smaller (point-like), quicker, and more fleeting.

4. The McCollough Effect -- An amazing optical illusion

I'm including this one because I realized that I ended up having a common theme of "can you believe your eyes?"  The answer is "no!" apparently.  So, I'm including this final, amazing illusion that is well worth trying out.

How to try out the experiment:

If you have 10 minutes, you should go try out this optical illusion: The McCollough Effect.  Spend the 5 minutes they recommend and then test it out for yourself.  You can read a description of it on wikipedia.  I found this illusion mind boggling, because for me it persisted for DAYS after I'd spent the five minutes training whatever part of my visual system that is being trained.  Just an amazing demonstration of an ability to unwillingly and semi-permanently "program" part of your brain (or visual system at least) just by staring at some images for a few minutes.