Hacker News new | past | comments | ask | show | jobs | submit login
Why the world needs deep generalists, not specialists (medium.com/swlh)
237 points by aytekin on Nov 18, 2018 | hide | past | favorite | 126 comments



Look, this essay is ridiculous. It starts by citing Aristotle, Leonardo Da Vinci, and Benjamin Franklin, and then delivers silly things like this definition of a "generalist":

> Someone who becomes competent in at least three diverse domains and integrates them into a top 1-percent skill set.

Then proceeds with:

> Despite the world’s immense need for polymaths, these individuals seem to be quite rare. That’s because society promotes specialization over generalization

No, it's because the thing you're talking about is a "genius" not a "generalist."

I agree that the world would benefit from more legendary geniuses like Aristotle, Leonardo Da Vinci, Benjamin Franklin, Francis Crick, or Richard Feynman! I do not agree that this opinion is novel, insightful, useful, or worth an essay.


> Look, this essay is ridiculous. It starts by citing Aristotle, Leonardo Da Vinci, and Benjamin Franklin, and then delivers silly things like this definition of a "generalist":

>> Someone who becomes competent in at least three diverse domains and integrates them into a top 1-percent skill set.

If you take this advice and tone it down a lot, it becomes both reasonable and good.

From Scott Adam's Career Advice [0]:

> But if you want something extraordinary, you have two paths:

> 1. Become the best at one specific thing.

> 2. Become very good (top 25%) at two or more things.

> The first strategy is difficult to the point of near impossibility. Few people will ever play in the NBA or make a platinum album. I don’t recommend anyone even try.

> The second strategy is fairly easy. Everyone has at least a few areas in which they could be in the top 25% with some effort. In my case, I can draw better than most people, but I’m hardly an artist. And I’m not any funnier than the average standup comedian who never makes it big, but I’m funnier than most people. The magic is that few people can draw well and write jokes. It’s the combination of the two that makes what I do so rare. And when you add in my business background, suddenly I had a topic that few cartoonists could hope to understand without living it.

___

0. https://dilbertblog.typepad.com/the_dilbert_blog/2007/07/car...


Thing is, it doesn't seem to pay, at least in the employment market.

Employers want specialists, and pay for them. If your CV is two or three distinct things (I don't mean a selection of languages, or front and back end dev here, I mean distinct skills like say programming and accountancy) it's very unlikely to get you a better salary.

In fact if you've diverged from a straight specialism throughout your career it's probable you'll be punished with a lower salary, and problems getting hired in the first place. Expect a lot of challenging interview questions as to why you haven't followed a straight line from junior, through senior into project management (which of course is a different skill, making this even more inconsistent).

I'm sure those same disparate skills could be hugely useful if you're starting a business or startup, but for employment it's likely to make life harder, even though many employers would benefit just as much from your increased breadth of perspective.

So outside a few rare employers, and a few lucky applicants, I don't think there's much out there for generalists, sadly.


> In fact if you've diverged from a straight specialism throughout your career it's probable you'll be punished with a lower salary, and problems getting hired in the first place.

I don't think this is true. Getting better at a thing has diminishing - often severely diminishing returns. Giving up being top 23% in programming in exchange for being top 25% in programming + top 25% in accounting doesn't seem like it would be much of a career liability.

> Employers want specialists, and pay for them. If your CV is two or three distinct things (I don't mean a selection of languages, or front and back end dev here, I mean distinct skills like say programming and accountancy) it's very unlikely to get you a better salary.

You're probably not going to get a better salary working for a company that only values one skill. You have to pick your employer more carefully. That being said, there are tons of software jobs in the banking/accounting sector ranging from product development to consulting. I'm sure that being an excellent programmer and an excellent accountant could make you much more money than just one without the other at some of those companies.


Recently I find this to be the case for me. I consider myself a generalist: engineering (a few stacks/languages), product, business, marketing, leadership and education. Recruiters tell me to focus on one thing on my CV and I feel employers don’t believe I could be even two of those things. For example with my blend of skills I was able to double the traffic, increase leads and social engagement by 5-6x at the last place I worked after my first quarter. My boss didn’t understand what happened and when I recount the steps and results to other people they don’t believe me. After winding down my own startup a few months ago I’ve struggled to get hired. People don’t advertise for generalists and as I remove certain skills and roles/duties from my CV I’ve found the hiring process gets easier. I’d love some advice, even more I’d love to put my skills to work.


A possibly silly question: did you counter with "Why don't you test my engineering skills if you don't believe me? Give me a fair chance. And then, if I'm not good, by all means reject me." That's assuming you've applied for an engineering job.

Another question: as https://news.ycombinator.com/item?id=18484883 says, would it help to rewrite your CV to highlight what the job is looking for? Again, if it's an engineering job, talk about your eng accomplishments in great detail, with perhaps only a bullet point for other skills at the bottom of the resume?

As https://www.kalzumeus.com/2012/01/23/salary-negotiation/ says:

> Did the solicitation for the job say “We are seeking someone with strong skills at scaling traffic in a fast-moving environment”? Pick out the key words. Scaling traffic. Fast-moving environment. “Scaling traffic” doesn’t sound like how I’d phrase it if I were writing or speaking for myself, but if you’ve just described your need to me as scaling traffic, by golly I will tell you how great I am at scaling traffic.


> Employers want specialists, and pay for them. If your CV is two or three distinct things (I don't mean a selection of languages, or front and back end dev here, I mean distinct skills like say programming and accountancy) it's very unlikely to get you a better salary.

This really depends on your employer and your definition of specialist. There are jobs where following a very straight career path would be a disadvantage: you might be perceived as a boring candidate who would potentially struggle if given work outside their domain. I think it's fair to say that, generally, breadth is valued more at smaller companies where people have to multi-task.

In academia it's much easier to do well as an inter-disciplinary researcher than it is to try and be the best in a single field. The obvious killer combo right now is machine learning for X, where X is a field which does not have a strong grounding in computer science. Hacker news gives a very skewed view on the world, as if everyone is up to date with the latest developments in deep learning. In reality, most people in academia are pretty unaware of what's possible, or don't have the technical skills (or grad students with them) to make use of it. This is currently true even for maths-heavy fields like physics, which you may find surprising - I certainly do. Now is a really good time to get on the ladder.

You also need to consider job satisfaction. People who are pretty good at several things may be bored with a single track career, and no amount of money will make up for that.


> I think it's fair to say that, generally, breadth is valued more at smaller companies where people have to multi-task.

IME, not at all during the hiring process. And only rarely later.


From experience at a small company? Obviously it will vary.

Your comment below gives one side, where companies want a specific thing and don't want to think beyond that. There is another side to that, where the company can't afford a dedicated engineer for X, and so being able to do some of that work is very valuable. I think this is probably more common in hardware startups more than anywhere else.

I've worked at companies (< 10 people) where we absolutely wanted different (relevant) skills, because we needed people to be reasonably autonomous in carrying out tasks. For example, we wanted people who had optical lab experience, could program and if you had mechanical engineering that was a big bonus.

My suggestion was that the odds of needing multiple skills is larger at a 5 person early stage startup than an established 200 person company. The larger you go, the more you're at the whim of HR screening. There may well have a long list of requirements, but on the job you're probably going to be more specialised.


Is your experience with companies that could make use of your other skill? If you understand accounting, and you're applying to a company that makes photography apps, that won't be given much value.


Obviously if your skills aren't relevant, they aren't going to be used.

IMHO, this is mostly due to the nature of the industry I am in (software). Hiring engineers is difficult, and only done when there are specific needs to fill - particularly at small companies, that can't afford to think too far ahead, or afield.


I was asking if you've found your other skills ignored even when you could explain to the company how they're relevant for them. For example, if you can do accounting, and you're applying to work at a company like Tally that makes accounting software, and you explain to them how your accounting skills are an added reason to hire you, have they, in your experience, still ignored those skills?


Yes.


> Hacker news gives a very skewed view on the world, as if everyone is up to date with the latest developments in deep learning. In reality, most people in academia are pretty unaware of what's possible, or don't have the technical skills (or grad students with them) to make use of it

Agreed, although I can also detect a lot of negativity from an older generation of academics in fields that classically border machine learning very closely. One remark I found funny (paraphrasing) is: "Total learning is a constant, and it consists of the sum of the machine's and the person's learning.. the deeper the learning of the machine.. the shallower the learning of the person". But I don't really have a stake in this and am not in much of a position to judge.


Do agree. I am a:

- philosopher (Bachelor's degree)

- sociologist (Master's degree with thesis on criminology)

- international relations student (another bachelor's dg in 2 years)

- C# Programmer (by a workshop)

- Java Programmer (by profession)

- Python programmer / Data scientist (by self education)

- Data Analyst with SQL(workshop & profession)

and the HR guys around here just think "I am not focused enough" and don't even get into the technical phase of the interview most of the time. Sadly, I think, if I could get into that part I'd ace it. Right now I am super demotivated though.

Maybe this is a regional thing, I can not comment about how my resumé would look to an international company but the question of "what does it mean to be an employable person" is a valid one and sadly can be answered with hyperspecialization and/or knowing the right people working in the company.


Working in a team is not only about skill, but about work-culture. If they have dismissed you because you have a broad range of interests, then that place wasn't right for you anyways.


There are very, very few places where a broad range of professional skills is considered an asset for a developer.

Most places want cogs. If you're too broad or have too many teeth, you're not going to fit in with the other cogs.

But you may be pitching at the wrong level. Developer jobs may not be the right niche for someone with a broad spectrum of skills. (By which I mean genuinely broad multidisciplinary professional skills and interests, not [three popular coding languages with some backend experience].)

Management and strategy may be a better fit.


Or analysis/statistics field?


I have a startup in healthcare (health.ai) looking in San Diego for broad interest developer... mike@health.ai


"Full Stack Developer" is pretty much a generalist, and they are as in demand as ever.


It is a generalist within the "computer programming" skill, so still a specialist. It all comes down to how narrow we define "specialisation".


on the contrary many job postings advertise for two or more core skills. Searching for any Finance or Advertising software position is likely to turn up something like the following

https://www.spotifyjobs.com/job/senior-ad-tech-engineer-back...

In addition to the standard Backend swe callouts the job desires ad-tech knowledge and advertising business knowledge. I'd be willing to bet they have a strong inclination towards anyone who's in the top 25% of both, but would consider cross-training for a top 1% in either.

As you go up the IC ladder the demand for cross-functional skills rises.


Is your experience with companies that could make use of your other skill? If you understand accounting, and you're applying to a company that makes photography apps, that won't be given much value.

Are you also good in your primary skill? Is it like 100% of engineering (say) + 20% of UX design, or 60% each of engineering and UX design? In the former case, one will clear the bar for an engineer job post, with your UX design skills helping to position you above other qualified engineers. In the latter case, you may be judged unqualified and rejected.

And what do you think about related skills, like UX design or marketing in addition to programming.


Having domain knowledge in accounting would certainly help you if you were writing accounting software. And similarly, having programming skills as an accountant could be quite useful for creating custom tooling.


Sure this is useful, but the vast majority of my jobs have involved picking up the domain knowledge in the first couple of months.


exactly. I throw all art out of my resume and the best offers come from companies in the same market as my current employer.

I'm a generalist, but for each application I have to mold it in to a fitting expertise. Even when applying for product and project management.

even within companies, no matter what you actually do, your job title will haunt you. doing 90% frontend dev and being the designer, doing 60% product and being the project manager.


but that is exactly what the OP is trying to change because the current status quo can not be upheld.

that it doesn't pay now is only because not enough people are aware of the benefits.

it's similar to the problem of hiring for a specific tools as opposed to hiring people who are good at learning new tools. if the new tools are similar to the ones they already know they will be productive in no time.

greetings, eMBee.


Project management requires a more generalist skill set compared a senior programmer. Specialization is good to a point, but then it can be more valuable to branch into other areas.

After you learn how to program, you will need to learn what to program and that requires more diverse skills and knowledge to make those decisions.


Is that actually true? I've met very few project managers who know any discipline other than project management with anything I would call depth. I can't even imagine talking to a PM about any serious technical question anymore. Their answers in the past have always been a unique type of unhelpfully vague but hard to argue with which puts all the effort back on you, but not in a way which adds anything to solving the problem.

Things like "what's best practice" when I've just exhaustively explained how we're in the precise corner case to which the normal best practice doesn't apply are, for me, the hallmark of PM ideas. Deep technical insight or domain knowledge are not.


Yeah, I have found a few PM's who were "used to be technical". To be honest I found it a bit frustrating at times as they would think they knew enough to make (some of the) technical decisions, when I felt that I should actually be making those decisions as was the one actually getting my hands dirty with the code. Plus I had more experience as a developer than them, as they had transitioned to PM and hadn't actually been building stuff for a while.


I would encourage people to find a better career coach than Scott Adams. Like many people who achieve remarkable cultural success, he doesn't actually know why it happened. For instance: does anyone think Dilbert succeeded because Scott is good at drawing? But it's first on his list of self-superlatives.

Edit to add: Great drawing skill does not matter for most comic strips. Look at Peanuts, or Garfield, or more recently Pearls Before Swine or XKCD. Ryan North has been posting Dinosaur Comics 3 times a week for 15 years using the same art for every single one.


Apart from him saying that his drawing skills aren't that great (which was the exact point above, it's better to be slightly above average in multiple domains than trying to be the best in one, which is almost impossible), I disagree with you that the mentioned comic artists don't have drawing skills.

XKCD looks simple, but the recognizability of these simple characters isn't that easy to achieve.


+1, I like your catch that the article is similar to Scott Adam’s Book’s advice. I liked the article, but I am biased since in my career I have tried very hard to be an expert at three things: classic symbolic AI, building large scale distributed systems, and Machine Learning.

I am in my mid 60s so I have had the benefit of working with many people who were individually very skilled in more that one field, and these people were the highest producers not just for the direct work they did but also for influencing other developers. These 7 or 8 people have had the greatest influence on my career.


Be part of the top of 25% in three independent fields and there is your (0.25^3 ~= 0.016) top 1-2% skill set.


It's hard to reach that level of skill in two (let alone three) fields in a reasonable amount of time.

It may also take your focus away from becoming good enough in your primary skill to pass the hiring bar. If you can do engineering, marketing and UX design, but you're not good enough in any of these fields to get hired as an engineer, marketer or designer, you're in trouble.

I think it's better to pick a primary field, get to the top 10% in that field, while being good enough at a secondary skill. For example, "I'm a great engineer, but I can also get you your first 10,000 users on the marketing side. I won't be as good as a dedicated marketing person, of course, but I can get this job done." Then you can apply for an engineer job, convince them how good an engineer you're, and then convince them that you also have some design and marketing skills.

(10% was an arbitrary number. The point is that your skills in your primary field should be good enough to clear the bar.)


Indeed. There is a fair bit of incoherent conflation going on in this article. I'm definitely a generalist, yet somehow this has, to my dismay, failed to result in me being hailed as either polymath or genius.


> then delivers silly things like this definition of a "generalist":

> No, it's because the thing you're talking about is a "genius" not a "generalist."

I think you may be misunderstanding that definition. It's not saying that you have to be the top 1% of practitioners of those domains. It's saying that by gaining general competence in at least three disparate domains and identifying ways to integrate those into each other, you inherently create a top 1% skillset for that amalgamation of skills. You might only have an average level of competence in each one individually, but drop into a situation that involves more than one of your competencies, and you've suddenly become far more valuable in that environment than more competent specialists that don't have the cross-domain knowledge that you do.

I have the type of "deep generalist" skillset that the article references. I'm competent enough to get by in about a half dozen competencies and a career that has hopped between technical, business, and consultancy roles in a variety of functional disciplines and industries. If I tried to pivot to a role focused on any one of those skillsets, I'd be hard-pressed to pass an interview for a role at the appropriate seniority level to be commensurate with my current salary.

But in the right environment and right organizational structure, I'm incredibly effective. I currently work for a consulting agency, and in the past year I've mainly worked on four different accounts. For two of those accounts, their needs were such that I was able to leverage a combination of my skillsets for them. The results were uniquely impactful enough that the client teams for both know me by name and both have tried to poach me from my agency, along with one vendor I worked with (on a client's behalf). For the other two accounts, I was brought in to fill a singular need that happened to be one of my competencies. While I filled that need adequately, the results weren't noteworthy enough to warrant me even having any direct contact with the clients beyond being one of the disembodied voices that pipe up sometimes on a conference call.


I do think there's still the point that we're wasting our geniuses by making them specialists.


Also nowadays all domains have advanced so much that it's far more difficult to become highly proficient in more than a portion of a single one of them.

Still, generalist-like integrators are needed, but not at the cost of specialists.


Feynman is kinda the counter argument though — I think his IQ was 120ish. Obviously very smart, but his unique insights seem to have come more from a broad base of knowledge and an intensely curious personality vs raw brain power. I don’t think you need to be a genius to make an impact in a lot of fields.


I think it's a huge mistake to compare the life of Richard Feynman and the results of an IQ test and think "huh, maybe he wasn't that smart."

Richard Feynman was obviously brilliant, based on any measure of real life impact you can imagine. Meanwhile, an IQ test is just something we made up.

The only reasonable approach to this set of data is to conclude that IQ is not as good as we wish at actually measuring intelligence.


I see IQ measurements as a rough estimate of certain cognitive skills that have some predictive ability of how well you will fare in solving classical school problems.

The pattern recognition skills that most IQ tests ask are fairly similar to solving a lot of exam questions. But they're only a part of the skillset that you need to become a great physicist.

If Feynman had prepared for an IQ test I bet he could have made it to 160. The kinds of patterns you see on IQ tests can be trained. But it's not a particularly interesting pursuit. I've personally done a professional IQ test and reached 130 while in university. Because in school you're training similar kind of problems regularly. I'm sure if I did an IQ test now I wouldn't even get close to 130, because I don't do exam questions regularly anymore. I also can't quote any successes similar to Mr. Feynman, so I doubt this is meaningful.

What IQ doesn't measure is curiosity. It doesn't measure how well you connect disparate disciplines. It doesn't measure how ready you are to upset the status quo. And if I read Feynman's book "Surely you're joking Mr. Feynman" correctly (great read btw), it was these things that actually made him think about physics problems from a different perspective.


Well first off, 120 is very smart. I don't see how saying that suggests he wasn't highly intelligent. My only point is that there's a reasonable case to be made that his contributions had more to do with how he acted, his personality, his drive, etc., than just with being born smarter than everyone. I really dislike the notion of "genius"; because I think it gives people an excuse to never develop themselves or to be happy with whatever intellectual plateau they've reached. Whenever I hear someone say "well so and so was a genius" I always feel like there's an unstated but implicit message of "so why bother?"


Usage of Feynman's official IQ score as some sort of 'proof' of anything is often overused.

I don't have the exact citation but it is fairly well documented that he scored well on other very challenging tests that correlate strongly with IQ (not to mention his actual career accomplishments).

Needless to say, people bandying around his score on this one test as evidence that he was 'bright but not in a high IQ' way is disingenuous or ill-informed, to say the least.

I could be wrong but I believe the common consensus is that his 120ish score on that test was not representative of his actual IQ, and was maybe the result of some other circumstance.


I like how when Richard Feynman's IQ test score doesn't conform to some narrative, HN is in contortions with excuses why the test doesn't matter, but when we come around to the topic of white IQ supremacy, suddenly HN is armed to the teeth with factoids about how g-loaded and circumstantially neutral the mainstream IQ tests are.


This seems like a reach, especially considering I don't personally believe the latter point you're making about IQ tests being neutral.

I'm not an expert on psychometrics, I think all I'm trying to point out is that Feynman performed highly on other tests that likely correlate highly with IQ, so I believe it more likely that there is some confounding factor at play other than this being representative of his actual IQ.

Or maybe that his IQ averaging here doesn't tell us much about which domains he scored highest in (some posit that he may have scored very poorly in a verbal section vs. mathematical etc.).


Or perhaps there is a difference between one-off individual measurements and group statistics, but please feel free to feel outraged...

Sorry for the snark but do you honestly not see that there is a difference between the two cases?


Thomas, were the same folks making these comments?


In that case, the question is, why those two groups of people never interact and never discuss with each other? You would expect a discussion or fights between these groups once in while.


Yes, and one might find them in the "comments" section of a technology news aggregator.


No, I am just sort of throwing up my hands and howling into the void here.


Maybe that's because a single data point [1] of a single person doesn't refute decades of research and heaps of evidence?

[1] An unreliable/uncertain one, even.


I don't think it's disingenuous or ill-informed at all. Maybe people get sensitive when it comes to intelligence, but say I said something like "Drew Brees is one of the greatest quarterbacks of all time despite being short for the position". It doesn't mean that height doesn't matter for the position (short stature is a disadvantage as a quarterback), it just means that he found other ways to succeed. I think you could make the same argument with intelligence: once you reach a baseline of "good enough", what you do with that intelligence is more important than just having a lot of it. Obviously Feynman applied his intelligence really well.

I think the argument of "so and so was a genius so don't bother you can't do what they did" is a really defeatist attitude.


I would have to go back and read my old copy of "Surely You're Joking Mr. Feynman" but I recall the story about his participation in the Manhattan Project. The way the story gets told, he quickly gained a reputation as someone who was unafraid to tell the older physicists when their ideas might have flawed thinking in them, and in return they would seek him out as someone fresh out of Cal Tech, trained in all the latest thinking, who could help them stay on track. So he quickly gained a reputation as a brilliant young hotshot, not to be taken lightly, even among the most esteemed Physics minds of the time.


that aside I think the question whether we need more generalists (of the non-genius sort) or specialists is a very important one.

And I would say that this is probably a good idea because it seems like most valuable knowledge (and with the most benefit) is increasingly found between disciplines rather than in any single given discipline. There is huge potential in bioinformatics, applying quantitative methods to urban planning, technology assisted design and so forth, all of which requires combination of very different sets of traits.


I'd argue none of these individual were geniuses though. By all means they were men that made considerable contributions to society but I think they were simply men that had opportunity and means to pursue their passions.

Aristotle almost certainly never worked a day in his life. He likely benefited from patroi-cliens arrangements. His job was to think, to learn, to observe the world around him. We don't really have anything like what he and his contemporaries had today, the closest we have is career academics.

Leonardo (his name is NOT Leonardo da Vinci, that would be like calling me Ryan Speedway) was again someone who mostly lived off of patrons/benefactors after his youth, which we know virtually nothing of, and had a lot of free time. He went into an apprenticeship at 14 and by 28 he was working for, arguably, the richest family in the world as an artist. Again, he simply had a lot of free time to contemplate the world around him and tinker.

Benjamin Franklin read. A lot. Like really a LOT. By 21 he was hardcore networking by creating Junto and later as a Mason. Combine his reading with his mechanical knowledge by working with printing presses and being generally successful financially again gave him a lot of free time for the era, time to spend contemplating existence, observing the world around him, talking to experts in other fields, sitting around Lodge discussing various topics.

I wouldn't say they were geniuses, I would say they were fortunate. They mad means of support and had free time. They had the freedom to pursue things that interested them and support systems in place to allow them to worry little about surviving and focus more on tinkering.


There’s probably been hundreds of thousands if not millions of people who had free time since Leonardo.

Leisure time is a necessity, but it’s not the difference maker.


>There’s probably been hundreds of thousands if not millions of people who had free time since Leonardo.

But did they have wealthy benefactors underwriting their life?

Aristotle thought for a living. He almost certainly never had to worry about money, ever.

Leonardo had to churn out works of art occasionally to keep money coming in. In the case of Leonardo he could pursue an odd project and in some cases get someone wealthy to finance it and as a student of art he studied human form, studied the mechanical nature of living creatures. He naturally applied that artist's eye to form and function to doodling inventions, some of which he made and some of which stayed on paper. I wouldn't be surprised if there have been hundreds of thousands of people that have doodled such machines and contraptions and simply not had the wealthiest family in the world to pay them to make some of them as novelties for court.

Franklin learned about printing, ran away, hustled and started his own printing operation becoming quite a name in the printing industry then used that success to throw money at many other industries. When he died his estate was worth tens of millions of inflation-adjusted dollars. It's a hell of a lot easier to be successful at a myriad of things when you actually have the funds to tinker.

Look at Silicon Valley, you see founders with multiple successful exits. It's a lot easier to get funding after you've already had a success, it's a lot easier to get press after you've already had success, it's a lot easier to get customers and attract good talent when you've already had success.


> He almost certainly never had to worry about money, ever.

Neither has every wealthy housewife, ever.

Nor every senator’s son, etc.

If that was the determining factor, every young adult rolling around in old money should be pumping out sistine chapels and eradicating polio. It’s necessary, but it’s not the thing that makes the difference.


> No, it's because the thing you're talking about is a "genius" not a "generalist."

I would disagree. A polymath is an expert / master of multiple fields (let’s say 3 or more). A genius does not necessarily excel in multiple fields, but he will certainly produce novel, groundbreaking contributions to a specific field. A genius tends to be a specialist — at least where his major contributions and legacy lies.

A generalist is not necessarily an expert or master of any field. He could just know enough in each field to be competent. So I think we’re talking about 3 different people: geniuses, generalists, and polymaths.

Anyone with a liberal arts degree can operate as a generalist. Leonardo Da Vinci and Ben Franklin were polymaths. Einstein and Feynman had genius.


We could use more unicorns. Unicorns for everyone. They are more common than ever!


looking at population increase - if unicorns exist at all - big if :) - then there should be more in absolute terms.


it's true and with the increasing connectness then that means they can contribute more readily as well! Gotta get all those unicorns hiding in east asia to save the planet!


I don't know why you have Francis Crick on that list. Franklin being on the same list as the others is also pretty weird.


You don't think that Franklin was a polymath? That's a pretty jarring take, can you expand on it?


The list in the parent comment was "legendary geniuses." Crick was certainly not that. He did a physics degree then did a bunch of half-assed work in biology. He stole some data, confounded the idea of gene and open reading frame so thoroughly that we're still trying to undo the damage decades later, and summarized a lot of early molecular biology results in an oversimplified way (the "central dogma") that we're still trying to teach people the real nuances of. On the whole he and Watson were probably a net negative for science.

As for polymath, he certainly wasn't that. I've done physics, biology, statistics, software engineering, violinist, composer, dance historian, learned a bunch of languages, and written a novel. I'm a pretty pedestrian level polymath, and Crick was nowhere near me.


Which has what to do with Franklin, exactly?


Oh, sorry. I misread. Franklin was certainly a polymath. But putting him in as a legendary genius next to da Vinci seems off.


“A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects.”

--Robert Heinlein


Have you ever tried to plan an invasion? Like, actually done a net assessment of a scenario in detail accounting for terrain, equipment, training, intelligence, logistics, and all the other things that go into a military encounter? It is time-consuming and hard. So is efficiently butchering a hog. So is learning to program well.

Changing diapers and pitching manure? Not so much.

Dying gallantly? Only get to practice that once, so how would you get good at it?

People throw this quote around because it is fun to imagine yourself as a supremely Competent Man in all these domains. Indeed, it makes for excellent fiction—-but it is just that, fiction. When you actually start trying to seriously think about investing in specialized skills like military science, this quote quickly reveals itself to have not made any distinction between life skills that the average person should have and things which require years of study.


My father started butchering lambs based on reading a book, and using tools already around the house (he might have bought a cleaver, but pretty sure that was after he'd done the first few). The first one took quite a while, but worked out fine. Did he ever get as efficient as someone who does it every day? Of course not. But it's very definitely a learnable skill.

I do agree that the invasion is a bit of an outlier, although even in that case there are questions of scale.

I've always loved that quote, but my reading is: on top of the everyday things, it's worth picking up several skills which require some deep study. No, you won't get them all, but having several -- and being open to learning something new if a good opportunity arises -- is valuable.


Sure, and its useful as an opener to the discussion of when to specialize and when not to. But it isn't an answer to the discussion.


The quote is tongue in cheek, you should have probably gotten that on plan an invasion.


It's simply not to be taken that literally but that doesn't make it tongue in cheek. Write some poetry, it helps understanding what others write.


In times of Aristotle, Leonardo Da Vinci, and (to a lesser extent) Benjamin Franklin one person could learn a significant portion of total scientific knowledge. That's no longer the case. There could be no more deep generalists.


So true. There's two types of research; one that builds upon prior knowledge to new conclusions, and another that undermines everything previous to be incorrect or incomplete. The first is done by your average PhD in Physics, for example, while the latter was Einstein and then Heisenberg, Planck, and Born.

While someone can be successful in the latter kind (undermining previous knowledge) through a fresh new hypothesis, the former kind of research, and the more common kind, requires understanding everything prior and reaching the current point of knowledge in that field. It's become increasingly hard, as evidenced by the increasing number of years of education it takes to master a subject.


To undermine previos knowledge you need to understand everything prior and reaching the current point of knowledge in that field, even more than if you want to make some incremental discovery. Einstein, Heisenberg, Planck, and Born had a deep knowledge of the current state of Physics at their time.


While it's necessary to possess a substantial amount of knowledge, I would argue that people who are older and have even more knowledge in the field are usually too accepting of the current information to question it enough to change it.

Einstein started working on general relativity at 28, and Heisenberg published the uncertainty principle at 26. While Planck and Born were older for their most famed work (late 30s?), they were still relatively young.

I would argue that this seems to show that there is some higher likelihood in a scientific epiphany at a young age, when you've yet to accept information as undoubtedly true, and instead are questioning it as you're learning.


You can still master 2-3 fields, if not to an international level. There are just far more fields to choose from now.


But can someone who knows only two or three fields out of so many that exist now be really called generalist?


I think you’re kinda stuck on scientific fields though. I imagine it’s hard to know all of physics and chemistry for example; but I imagine you could simultaneously be say an excellent painter and also an excellent fiction writer, etc.


Where by 'the world' he means 'the economy', whose neurotic demands for ever greater consumption are precisely what 'the world' does not need.


I mean... good luck finding a system of self organization for seven billion people which doesn't involve people trying to better their situation in some way?

I mean I appreciate the sentiment, but 'the economy' might also be what's needed to tip us towards renewables (as people will want cheaper energy, and jobs making it possible).

And can it really be neurotic when it's the culmination of seven billion individuals actions? Doesn't neurotic imply an inherit interest in self?

And can you blame the poor for wishing for a better life?

I'm just not sure there's much to this comment besides a "holier than thou" vibe.


I'm quite happy to admit (it's kind of self-evident) that criticism is easier than building. It's a known known now that our economy does end in catastrophe - the collapse of ecosystems (on which all life, ours included, depend) is happening apace, is entirely well-documented, and barely starts to be addressed by shifts to renewables. That's mostly limited to power generation, and in any case climate collapse is only one amongst thousands of ecosystem challenges.

Knowing, as we do, that an ever-expanding economy leads to global collapse doesn't imply that we know the fix. I don't. Personally I doubt there is one. I think nothing in H. Sapiens' evolutionary inheritance empowers it with the attributes necessary to manage a global civilisation.

> besides a "holier than thou" vibe.

If you get that vibe from me, that's just my inadequate communication. I'm neither claiming to live better personally (which is anyway just more consumerist narcissism), nor to have a blueprint for a sustainable global civilisation (which would make me a towering genius).

The only claim I'd make here is that I'm not in denial about physical reality (which economists and most business people are).

> And can you blame the poor for wishing for a better life?

The poor aren't the problem. The rich are. Adequate food and shelter could be achieved without trashing our home. Endless expansion of consumption cannot.

> can it really be neurotic when it's the culmination of seven billion individuals actions

It's not. There's such a thing as culture - human 'individual' actions have systemic causes. A culture of ever-increasing consumption is neurotic. It replaces the essential (functioning ecosystems) for the trivial (expanding consumer consumption).


> Knowing, as we do, that an ever-expanding economy leads to global collapse

what?


Our world economy can't keep growing. Resources must be extracted and consumed to facilitate the movement of wealth. We must eventually stop, even the gray goo replicators run out of resources once they convert the planet to themselves.


I agree your conclusion is undeniable, but I don't think conceptualising endless growth as mere 'resource' consumption quite captures the danger & tragedy of the physical processes involved.

'Economic'(1) growth, in real physical terms (forget the fake virtual world of the economist), is the transfer of matter and energy from our world's self-sustaining complex systems to crude, temporary, early-stage technological ones. It is the transfer of entropy to the complex evolved systems on which we depend, causing their dynamically stable states, developed over millenia, to break down. These living systems are dying, everywhere on the planet. Our extremely fragile global civilisation will not last long as crises of our home's destruction start to roll in.

Look at the convulsions caused in Europe by the feeblest of mini-crises in 2015. Far more is to come, and the consequences will be expanding waves of ever more intense wars and collapses of what passes for cooperation between nations.

We've allowed the propaganda of endless economic growth, fictionally severed from physics, chemistry, and biology, to be broadcast without challenge by corporate capitalists for too long to pull back now. People believe in it. It is religion (and by far the most dangerous fundamentalism in history). I think it's a done deal now.

(1) scare quotes intended here because the 'economy' is a Platonic fiction cooked up by fluffy-minded business folks & economists to make it seem like money floats, untethered, beyond physical reality. But it doesn't. Reality is coming back to bite them, and, unfortunately, all of us (where "us" includes the more-than-human world)


I believe the key is to never specialise in one thing for too long. You can always come back to something but you need to diversify your interests over time.

First I can't imagine any of the polymaths simultanenously working day to day at a deep level across multiple fields. We don't seem to be able to multitask that well.

However, I have a theory (so take with a grain of salt) that when we focus strongly on a task our brains rewire our neurons to optimise to use particular patterns for that situation. When you switch to a different field or spend time on a hobby your brain switches again as the previous patterns may not apply. It seems the breakthroughs come when the patterns from one field do apply to the new one. You may have been working on it forever but it took that change and the rewire to properly see it differently.


I like to think of myself as somewhat of a polymath in tech - enjoying and doing decent work in a variety of types of software at all different layers. I also enjoy other things that I would put myself squarely in the hobbyist side: small engine repair, metal and woodwork, plumbing etc.

So how would one determine if someone has 3 "top 1%" skill level (per the article)?

I suppose the answer is in 2nd half of the article?

Two types of more reasonably defined polymaths emerge - pick three skills and achieve a certain level of output in a reasonable time:

apprentice level 4% input, 64% output, 10 months

journeyman level 10% input, 80% output, 48 months.

thoughts??


"Polymath in tech" is by nature not in line with the article IMHO. Let's assume you are a tech expert, now try to be a master chef or PhD in sociology or Olympic gold medalist in rowing etc

I think the article mentions people who are able to master at complete different fields.


you are right, but there is a paralell. when it comes to hiring programmers i prefer people who are good at programming in general and at least familiar with two or more languages. because those who are very good at only one language and have never touched another will have a hard time learning a second language, and, will also have a hard time getting to expert level because they will miss out on that additional pwrspective that other ways of doing things provide.

same goes for knowing multiple web frameworks like angular and react for example, to be even more specific.

greetings, eMBee.


good points - completely different fields is sort of key to the article's points

edit: btw - when I read your first sentence - I was like "ah shit...duh" :)


My first thought on the requirement of reaching the top 1% was that it's unrealistic to achieve in three very different fields. But after some careful consideration it might actually be possible for quite a lot of people. What does it take to be a top 1% programmer? If you're smarter than most people, probably less than 20 years. Do that three times in different fields (and keep up with the other fields while learning a new one) and you're a polymath before 80. Be still a lot smarter, you can do it before 50.

It's just that only some people are capable of reaching top 1% in their field to begin with. And of those capable, most won't because they change fields before they reach it or they have other priorities, etc. And the few people that do reach top 1% tend to be passionate enough to keep at it for the rest of their lives, and it's usually lucrative enough to try reaching an even higher level at that point.

I like your definition of a reasonable polymath and it's my goal to be one. Most of us can't be da Vinci. Still, I hope we will see more of the true, strict, history-changing polymaths.


I think being a good polymath is always seeking knowledge and taking on a vast variety of small jobs.

Nothing is hard if you get going in it. And the best way to learn something is to have a problem to solve (that isn't too far fetched).


Being a generalist by nature makes it very hard to build a successful career. I'm a generalist and what I see is the society is built around specialization.


Personally I think being a generalist is ironically is effectively the same niche as yet another specialization. It sounds nonsensical and contradictory at first glance but makes some sense. If you are able to do everything equally well you will likely be out-competed in any given field by specialists (barring say a generally applicable approach flat out being better like an engineer designing tools which can outperform master craftsmen in both volume and precision).

However if there is a task which requires broad knowledge (say integration) or it is just more efficient or otherwise better for the current scaling due to the costs of trying to maintain a stable of appropriate specialists it is a niche where a generalist will win out. We usually see a general practitioner first instead of heading straight for a dermatologist about a rash.


> We usually see a general practitioner first instead of heading straight for a dermatologist about a rash.

Which is why, sadly, most polymaths within tech end up getting stuck at tech support.


Quite right, an area where you also see this is consulting.


Consulting is not an "area". You are supposed to specialize in some area and accumulate specialized expertize in it first before you can become a consultant anybody would be interested to consult to. Isn't this so?


In consulting it pays off to be a generalist.

On my case it helped quite a few times to jump between projects, or to be person that customer would like to ask to jump at an existing issue.


IMOP the best profession for a generalist is entrepreneurship. Not that entrepreneurship doesn't entail its own hardships though.


Perhaps but entrepreneurship sounds quite fearsome for people that neither feel really sociable (and I doubt many generalists do) nor have a serious financial capital they can risk safely.


Testify, brother!


The author focuses on the generalists who did great things, but ignores the specialists who do important work, and speaks nothing of the many jack-of-all-trades who can’t find a job because their knowledge isn’t deep enough in any valuable area. If anything, most of the people I’ve seen not hired and let go has been due to not enough specialized knowledge, and this preventing them from being effective.

The software engineer unemployment rate in the US is something like 1.6%. Engineers who specialize in self-driving cars get paid absurd amounts because there aren’t enough of them. We need more specialists, not generalists. If they’re specialists in more than one thing, that’s even better, but it’s not the main problem.


Eric Weinstein just needs to teach the world how to self-teach yourself anything. He's given plenty proof he's managed to do it himself, and he says he knows the secret. But I still eagerly await him to explicitly describe the polymath recipe.


I think there is an issue of power/money. If you have power or money you can call yourself a generalist and hire the experts in the various fields you want to assemble.

For the generalist not in a position of power, the dynamic is really different, people won’t hire you to read a book or increase your knowledge on something else during office hours, either you have to be independently wealthy or use family time.


The article mentions Google's Code University. According to its web page "The University Consortium is no longer actively maintained."


Generalists are the first ones to be laid off. Eventually there is cost cutting and management says, "what does that guy do again?".


It takes 50,000 hours of productive practice to master a trade or study. Who has the time to accomplish three of these?


Some people learn fast, very very fast, and they might get so fascinated by something that they have no choice but to try to learn certain things. It can feel like suffocating not to try to learn that amazing looking something, and it can be so much fun when the learning is so fast. So for some people its less about time and more about accidentally learning multiple subjects.


A category theory studies the general form of many familiar structures. A category theorist's publications are however usually starkly specific.

What is the point of this article then? Maybe it is popular, cheap reading like Malcolm Gladwell and something in the line of "career advice".


I'm not sure how useful this article is, but I think the idea that combining multiple disciplines can lead to breakthroughs is hardly controversial. It's certainly great for startups and for the whole "software is eating the world" thing.


I don't see how someone conciously chooses different domains and becomes good at them. this isint a strategy that can be followed. what ends up happenning with people who are pretty good at a lot of things is that - they start doing something because they find it interesting, then they get bored and move onto something else that catches their interest.


I've never seen a school that teaches generalizing skills. To find one, that's a mystery.


It's limited to the general concept of the computer industry, but I did a degree in IT and several professors saw the focus as being "integration". You basically did a CS minor, plus classes in electronics, signals, web dev, databases, operating systems, etc. Projects have a heavy emphasis on communication, project management, etc.

My peers ended up very quickly becoming programmers along-side CS grads, managers of various kinds, founders, IT personnel, tech support, hardware designers, etc.


in my university, we were allowed to pick any minor with a computer science major because they believed that computer science can be applied to everywhere.

this was in the 90s when many universities required.that compter science be paired with math or a few other selected fields.

turning that into a CS minor with the same sentiment would be an evolution of that approach.

greetings, eMBee.


A lot of the data science community was pushing statistics minors with some other major (or vice versa) for a while. I know that field is generally a bit over-hyped from an employment perspective to begin with, but I think that makes a lot of sense.


Try Scotland. It used to be a thing there.


College ?


Don’t we need both?


Good luck selling yourself if you go this route.


Technically better than betting on a specialty that becomes reduced to uselessness such that professing knowledge in it is practically shameful but it still sucks.


I know this is true from my own experience.


me too, so I just tell people that I know node.js :/


Which is true regardless of your current knowledge of the subject, when you compare against the hordes of people who claim the same.

You likely can pick up node.js baseline knowledge in a week.


you should know this was kind of a lightning-bolt comment for me. thanks!




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: