Doug Abdiel

Back onto this idea of breadth vs depth.  It has certainly not disappeared from my mind this week.  I did object lessons with my nieces about physics, taught one of them how to use a dictionary, and spent no time at all investing in areas that could make the lives of a lot of people slightly better. It was not that their plight was unknown to me.  Quite to the contrary, I have been reading the Orphan Master’s Son, about the plight of a lot of the people in North Korea, and it is heartbreaking to think what cold some of the people in labor camps there must be dealing with.  I also passed through a couple of airports which made me think about this article that I wrote a couple of months ago about airport screening, and what the experience could be like for refugees. Despite their plight being very real, and poignant, I did nothing to help the broader groups of my country, or people living in poverty.

This seems like a fitting place to pick up the last post.  While I ‘felt’ great spending time with my nieces, and teaching them little things about how the world fits together, I am not certain that this was the best way to invest my time.  While I undoubtedly brought utiles of joy and learning into their lives, I could have done the same in the lives of the nieces and nephews on my side of the family, or for any of the nameless, faceless people in the developing world.  This hits squarely on the second problematic approach of using a diminishing marginal utility model as a guide to my action.  That problem is specifically that I don’t have fungible units to move between different goods on an hourly rate.  How was I to know if the ‘learning’ utiles I created by spending an hour with my niece showing her how to use a dictionary were greater than the simple ‘joy’ utiles I created with two of my nieces and an nephew making a moat around their sandcastle?  Is a learning utile somehow more valuable than a joy utile?  If one is more valuable than the other, then what is the exchange rate?  1:2? 2:1? Do they both convert into something more fundamental?  There is no simple way around this.  My feelings about the activity often lead me astray on this.  The joy that I feel in spending time with these girls undoubtedly clouds my judgement.  Indeed, since it is the season for giving gifts it is probably worth saying that many of the gifts we get are likely clouded by what we would like to receive if we were in the other person’s shoes.  This failure of empathy results in a large deadweight loss and frustration at Christmas time.

OK, the problem is clear enough, now what of the solution?  Again, the solution here may be the same as the first, namely, diversification.  If we do our best to create different types of good outcomes, and assume that for most of them there is diminishing marginal utility, then perhaps we will be OK.  For instance, perhaps the longevity of education in the perfectly rational world makes its utility far higher on an hourly rate than the simple joy of creating a memory about a holiday by the sea, thus the exchange rate is largely in favor of education.  It seems likely that the retention rate of that education, and thus its utility would rapidly decrease, just like all other goods.  This means that each additional hour spent teaching maths, or the dictionary, or physics object lessons is less useful than the last.  Let’s just say, for argument’s sake, that education has a base utile/hour rate of 10/hour but diminishes by 80% for each additional hour in a sitting, so hour one is 10, hour two is 2, hour 3 is .4, etc. The joy utiles hour at the beach may have a lower utile/hour rate of 3 and also diminishes by 80%/hour, so the first hour is 3 and the next is .6 and so on.  Thus one hour spent at the dictionary may be much more valuable than the same hour at the beach, but if I spent an hour doing both, then the total utility may be greater than doing either for that block of time (10 of learning and 3 of joy vs 12 of learning and 3.6 of joy).  This doesn’t really solve the problem, we still can’t convert joy to education, or vice versa, but if thing decline rapidly in utility, then diversification can create net benefits.

On to the final problem of this diminishing marginal utility approach, the issue of negative upfront utility.  This is a big problem if we plan to press ahead with diversification.  Let me give a very clear example of the problem from my own life.  My mother wanted me to learn the fundamentals of music on the piano before learning the trumpet, an instrument that I probably favored because I wanted to be more like my big brother.  I only later appreciated the joy, and beauty that I could create alone and as a soloist on my trumpet.  So that’s the setup, in order to create beauty and joy for myself and others, my mother wanted me to learn the piano before the trumpet.  The problem was a woman named Mrs Wheel. She was approximately 5232 years old, and looked and smelled, to my second grade self like the criptkeeper (image here, if that’s too American-centric). Every time I went to see her, I had halitosis-induced negative utility, that didn’t improve in the year or so that I spent with here.  It didn’t help that I didn’t practice diligently (or at all), and she chided me every time I saw her.  After this year of pain, my mother allowed me to play the trumpet, and though I ultimately thoroughly enjoyed it.  I still struggled at first with headaches from the pressure, pain from an under-developed embouchure, and I certainly did not create anything that could be called beautiful in the house for others to enjoy.  Ultimately, after years of practice, I became quite good at the instrument, and played in ensembles and as a soloist.  So let’s bring this little example back to the problem of utility.  At first during both the playing of the piano and the trumpet, I created negative utility personally and for those around me through my earnest exertions, ultimately, though, I think that I created something beautiful that was recognized, amongst other ways, in a John Philip Sousa award in my final year of high school.  If I had just targeted shorter-term utiles, then I never would have created something beautiful.  Perhaps that would be sorted out by the 70 year time horizon, but perhaps not.  There were people who I ran into in my music career who were clearly not talented.  They didn’t have an ear for music, and it is entirely possible that even through years of training that they would never make a net-positive effect on the world through their exertions.  In fact, I would say this probably applies to the majority of students who have an instrument, probably a plastic recorder, placed in their hands.  They then proceed to squeak for a couple of  quarters in elementary school, decide that ‘music’ is hideous, and then move on.  Only a very few go on to make music that is heard by the masses. Let’s remove the complication that we ‘appreciate’ music more by going through this painful ritual with a recorder, and assume that people are still able to enjoy good music even if they have never created it themselves, just as I am able to enjoy sculpture even though I have never sculpted.  I say this not as a diatribe against American musical education, but to point out that my story about music is the exception and not the rule.  A great many of the things that we first experiment create negative utility at the outset, and over the long term that may never balance out.  I suspect this is especially true for non-linear systems where there may be perverse or null consequences to our actions.

OK, so I have again, probably more thoroughly than I needed to, exposed the problem.  What to do about it?  The problem here is that the only possible solution runs totally contrary to the solution of the other two problems.  If you diversify dramatically, then you might well run into a lot of negative or low utility scenarios and never make your way out of them.  If I want to protect the American and Australian publics, be a good husband who spends meaningful time with my wife and soon to be child, do my bit to continue to protect the US and Australia in the reserves, and be a contributing employee at Google, that’s already a lot to try to do well.  If I were to add to that something like caring deeply about the poor people who I run across every day, volunteering to help refugees locally, and internationally, and starting a double-bottom line business, then I run the risk of very quickly not doing anything well.  The skills that I develop in one arena may be useful in another arena, but they are not going to be a perfect match, the time that I spend switching between areas that I care about will be wholly lost time.  In general, the more things that I pile on, the more that I run the risk of doing a lot of things poorly, being stressed out of my mind trying to keep all of the plates spinning, and generally creating negative utiles in every area.

Unfortunately, this is something like the sickness of our age.  Most people I know, when asked how they are doing, say that they a ‘busy.’  This is unfortunate for a couple of reasons. First, I think that many of them equivocate ‘busy’ with ‘important.’  Second, in my experience, the more truly busy someone is the less they are actually able to meaningfully contribute to anything.

It may be possible to reconcile the initial negative utiles invested in one area with the diminishing but positive utility invested in another, but that would require mathematics.  I could, for instance compare the net utility of creating joy in my wife and child’s hearts by spending more time at home over the next reasonable planning horizon (say 2 years) with the net utility of improving the browsing experience of consumers in Japan by bringing more and better content to them more rapidly over the same time horizon. Unfortunately, given the chaotic and disparate nature of these choices, they don’t lend themselves to quantification.  I could well labor on this project in Japan for that time and make no discernable impact because of a policy or product or legal decision we make in California.

I think that I have failed resoundingly in these last two posts.  I have done so in at least two ways:

First, I don’t think that I did a good job picking apart the depth of investment in one area with the same investment more broadly (i.e. creating joy just in the heart of my future child vs the same amount of joy in the hearts of many people). My discussion often juxtaposed competing ‘why’s’ instead of breadth vs depth.

Second, I failed to come to a resolution here.  I thought I worked myself into a reasonable approach using diminishing marginal utility, but it left me with two competing tactics.  I don’t see any way to reconcile these without delving into mathematics, but the material doesn’t lend itself to that.

Frustrating…