4 ms·

Experts on the Future of Expertise

30 points by fagnerbrack 5y ago

vector_spaces 5y ago
The contributors to this piece are almost entirely (tech) entrepreneurs and content writers and editors at tech companies. Most of these people aren't known best for spending years of their lives understanding, say, combinatorial optimization, or the phylogeny of anoplogaster brachycera, or covert prestige in Kanjobalan–Chujean language speaking communities. Which isn't to say that there aren' experts of this nature among them, nor that people should be best known for this sort of niche and academic work if they've since gone on to do other things. But it makes me highly suspicious, to be frank. For instance, the first "expert" isn't even an expert -- he's an amateur data scientist who happened to build an accurate model of COVID-19. Did we all simply forget how both amateur and professional data scientists publishing bullshit COVID-19 models last year stoked the flames of misinformation for the sake of a cool portfolio project? I'm down to shit on academia. Who knows how much knowledge production have been squandered by its administrative excesses, gatekeeping and abuse? And I love that the internet has leveled the playing field in a lot of ways. But at the same time I have little faith that the private sector and tech in particular will be the great equalizers, and I'm surprised that anyone does given how much knowledge production is squandered by their gatekeeping and walled gardens. And guess what else? At the end of the day, there are lots of hard problems that can't be solved with Google, grit, and a little python -- you actually need to sit down, do the reading, do the work, listen to people who've been doing it longer than you, and do that for years.
- jrumbut 5y ago
  I want this post on a throw pillow, it's so good. I would add another criterion for picking when you want the stability of an expert over the upside/cheapness of a talented amateur: do you choose your problems or do your problem choose you? Startups are the ultimate example of choosing your problem, you try to search for one piece of low hanging fruit and build from there. If it's really low hanging fruit, a very motivated person can pick it. Government is the opposite. The Department of X has to deal with every X problem as they come up, if it's too hard they need to get better. They can't pivot (while remaining true to their mission, obviously governments fail in critical situations all the time). The stakes of a startup could be extremely high, but they're allowed to fail fast and move on with a clean conscience.
- oaxly 5y ago
  Well said. Learning takes time. The speed at which info flows between chimp brains may increase, and the quantity of accesible info may increase, the formation and rearrangement of chimp networks may happen on a button click but the chimp is still a chimp.
softwaredoug 5y ago
My hypotheses is that experts are like fast food chains. They give a reliable, mediocre experience. Whereas non experts are like local restaurants - you get much higher variance. You could find something unique and special, or you could have food poisoning. When the risks are low and upside high, experts working in isolation can indeed lead to groupthink and suboptimal problem solving. You might get gradual, incremental, boring progress. Exactly the opposite of what’s wanted! However when the downside risk is high (ER doctor, hungry kids in the back seat, legal troubles, evolving your company’s cash cow product) you will prefer the boring, consistent experience over the one that dramatically upends everything. I know I wouldn’t want someone just winging it on shift when I go in for an ER visit. The problem to me is less about expert vs non expert, but rather how we navigate our own expertise. When do you recognize the downside risk as high, so you should double down on what’s known? Or when do you realize there’s tremendous upside if we lose our attachment to prior art on a topic?
- bitschubser_ 5y ago
  I really like the Idea of Philip Tetlock of "foxes and hedgehogs" regarding expertise and decision making (in this case political decision making): https://www.youtube.com/watch?v=EeHyVauX458 https://www.youtube.com/watch?v=EeHyVauX458 This can also be applied to other fields e.g tech and somehow supports your hypotheses :)

30 points by fagnerbrack 5y ago

The contributors to this piece are almost entirely (tech) entrepreneurs and content writers and editors at tech companies. Most of these people aren't known best for spending years of their lives understanding, say, combinatorial optimization, or the phylogeny of anoplogaster brachycera, or covert prestige in Kanjobalan–Chujean language speaking communities. Which isn't to say that there aren' experts of this nature among them, nor that people should be best known for this sort of niche and academic work if they've since gone on to do other things. But it makes me highly suspicious, to be frank. For instance, the first "expert" isn't even an expert -- he's an amateur data scientist who happened to build an accurate model of COVID-19. Did we all simply forget how both amateur and professional data scientists publishing bullshit COVID-19 models last year stoked the flames of misinformation for the sake of a cool portfolio project? I'm down to shit on academia. Who knows how much knowledge production have been squandered by its administrative excesses, gatekeeping and abuse? And I love that the internet has leveled the playing field in a lot of ways. But at the same time I have little faith that the private sector and tech in particular will be the great equalizers, and I'm surprised that anyone does given how much knowledge production is squandered by their gatekeeping and walled gardens. And guess what else? At the end of the day, there are lots of hard problems that can't be solved with Google, grit, and a little python -- you actually need to sit down, do the reading, do the work, listen to people who've been doing it longer than you, and do that for years.

jrumbut 5y ago
I want this post on a throw pillow, it's so good. I would add another criterion for picking when you want the stability of an expert over the upside/cheapness of a talented amateur: do you choose your problems or do your problem choose you? Startups are the ultimate example of choosing your problem, you try to search for one piece of low hanging fruit and build from there. If it's really low hanging fruit, a very motivated person can pick it. Government is the opposite. The Department of X has to deal with every X problem as they come up, if it's too hard they need to get better. They can't pivot (while remaining true to their mission, obviously governments fail in critical situations all the time). The stakes of a startup could be extremely high, but they're allowed to fail fast and move on with a clean conscience.
oaxly 5y ago
Well said. Learning takes time. The speed at which info flows between chimp brains may increase, and the quantity of accesible info may increase, the formation and rearrangement of chimp networks may happen on a button click but the chimp is still a chimp.

softwaredoug 5y ago

My hypotheses is that experts are like fast food chains. They give a reliable, mediocre experience. Whereas non experts are like local restaurants - you get much higher variance. You could find something unique and special, or you could have food poisoning. When the risks are low and upside high, experts working in isolation can indeed lead to groupthink and suboptimal problem solving. You might get gradual, incremental, boring progress. Exactly the opposite of what’s wanted! However when the downside risk is high (ER doctor, hungry kids in the back seat, legal troubles, evolving your company’s cash cow product) you will prefer the boring, consistent experience over the one that dramatically upends everything. I know I wouldn’t want someone just winging it on shift when I go in for an ER visit. The problem to me is less about expert vs non expert, but rather how we navigate our own expertise. When do you recognize the downside risk as high, so you should double down on what’s known? Or when do you realize there’s tremendous upside if we lose our attachment to prior art on a topic?

bitschubser_ 5y ago
I really like the Idea of Philip Tetlock of "foxes and hedgehogs" regarding expertise and decision making (in this case political decision making): https://www.youtube.com/watch?v=EeHyVauX458 https://www.youtube.com/watch?v=EeHyVauX458 This can also be applied to other fields e.g tech and somehow supports your hypotheses :)