Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

I strive to be a 0.1x Engineer (2016)

benjiweber.co.uk

Post reply on HN
Print view View original

121–130 of 144 posts

Re: I strive to be a 0.1x Engineer (2016)

#121

Post by throwawaylinux »

trashtester wrote: post #116
throwawaylinux wrote: post #111

Earlier quoted context omitted.

Yes it is absolutely possible, for particular problems . There are problems where the average or even majority of engineers would create exactly zero value over time.

> Yes it is absolutely possible, for particular problems. When I said infinite is impossible, I meant in absolute terms, not relative to somebody else. n-x-engineer means that: productivity_of_n-x_engineer / productivity_of_reference_engineer >= n (assuming the reference isn't negative) etc So, I'm arguing that the numerator cannot be infinite, so if n is infinite, the denominator has to be 0. In that case _everyone_…

> When I said infinite is impossible, I meant in absolute terms, not relative to somebody else.

The whole thread is about relative terms, including when I said infinite-x. And the context is clearly about certain worthwhile problems, nobody talks about a 10x engineer as someone able to make a trivial phone apps or web pages 10x faster than than average.

Re: I strive to be a 0.1x Engineer (2016)

#122

Post by eloff »

nerdponx wrote: post #36
eloff wrote: post #30

Earlier quoted context omitted.

Do you guys work on the same industry as me? There's easily a 10x difference between the best guy on the team and the worst guy who's about to be on the performance plan. It's a real thing. A lot of programmers really suck. I interview them and wonder why they're in this industry to begin with. I'm easily 10x those guys, they struggle with for loops. Then there are some superstars like Evan Wallace who created esbuil…

So you're 1x, Evan Wallace is 10x, and the pour souls in question are 0.1x.

It's all relative

Re: I strive to be a 0.1x Engineer (2016)

#123

Post by eloff »

xeromal wrote: post #34
eloff wrote: post #28

Earlier quoted context omitted.

That's no leverage in running. There's a lot of leverage in software engineering. That's what allows such a wide diversity of productivity.

What is leverage? VIM?

In this context, the ability to have an outsized impact in relation to your input. Think of a physical lever.

Re: I strive to be a 0.1x Engineer (2016)

#124

Post by badpun »

qikInNdOutReply wrote: post #105
badpun wrote: post #90

Earlier quoted context omitted.

> I wonder why this has not yet happened with open source. If a sufficient enough part of the public uses your software, the state offers you a maintainer role. Nothing big, but enough to life on and do work. It didn't happen because people already do it for free, so why overpay?

Because softwares ability to change are a living thing and it becomes stagnant and brittle once its left with no humans to keep it alive in symbiosis?

Is there any crucial project left without any maintainers though?

Re: I strive to be a 0.1x Engineer (2016)

#125

Post by selfhoster11 »

happymellon wrote: post #96
Aleksdev wrote: post #82

Earlier quoted context omitted.

You do realize Einstein was always studying? Sergei Brin worked like a maniac. Struggle breeds success. Not everyone successfully picks the right thing to struggle on.

None of those things were products of their work environment. Einstein was able to always study and Sergei Brin was able to work like a maniac precisely because they were able to remove the restrictions of paid technical labour that drains your brain before you even get to the interesting stuff.

Exactly. Even if I want to do something in my own time that's similar to my work after hours, I have no personal energy to devote to that unless I'm rested or very desperate. It's rare that I actually boot up an IDE on my personal machine.

Re: I strive to be a 0.1x Engineer (2016)

#126

Post by badpun »

Demonsult wrote: post #119
coding123 wrote: post #63

So here's a weird question. Are 0.1x developers getting paid the $400k, 500k salaries or is that just 10x'ers?

I've been doing this for more than 25 years and the income gains for being productive are miniscule. Strategizing your job hops carefully is what gets you the 1/2 million.

There's probably an economic opportunity for a tech company which only hires true 10x-ers, and pays them $500k-$1m.

Re: I strive to be a 0.1x Engineer (2016)

#127

Post by trashtester »

throwawaylinux wrote: post #121
trashtester wrote: post #116

Earlier quoted context omitted.

> Yes it is absolutely possible, for particular problems. When I said infinite is impossible, I meant in absolute terms, not relative to somebody else. n-x-engineer means that: productivity_of_n-x_engineer / productivity_of_reference_engineer >= n (assuming the reference isn't negative) etc So, I'm arguing that the numerator cannot be infinite, so if n is infinite, the denominator has to be 0. In that case _everyone_…

> When I said infinite is impossible, I meant in absolute terms, not relative to somebody else. The whole thread is about relative terms, including when I said infinite-x. And the context is clearly about certain worthwhile problems, nobody talks about a 10x engineer as someone able to make a trivial phone apps or web pages 10x faster than than average.

> The whole thread is about relative terms, including when I said infinite-x.

Relative means numerator/denominator. My first reply was simply breaking that down into parts, aruing that the numerator could not be infinite, and that the denominator being zero is not a very interesting case.

> nobody talks about a 10x engineer as someone able to make a trivial phone apps or web pages 10x faster than than average.

Actually, that's precisely what I think a lot of people who hear the term interpret it as. Especially from those who reject that the 10x engineer exists.

And there are quite a lot of engineers out there that believe that all any software engineer is doing, is to follow detailed instructions/specifications for some product, whether that is to make trivial web page or improve how gcc optimized code on a new cpu architecture.

While in fact, the most productive engineers out there do NOT follow such specifications from others, but rather develop something revolutionary based on their own ideas. You're not likely to find 10000-x or better software developers that don't do it like that (I would argue that people like Linus Torvalds or Elon Musk would rate at around the million-x level, but not infinite-x )

Re: I strive to be a 0.1x Engineer (2016)

#128

Post by citizenpaul »

Man that applying the 0.1x part really reminds me of being in mgmt. I felt like my job was more like chief disappointer and garbage tosser of ideas. People forget that its easy to sit and spit out 100 ideas at a session. The hard part is throwing away the stuff that is good but just doesn't need to happen for various reasons.

Re: I strive to be a 0.1x Engineer (2016)

#129

Post by tasuki »

shepherdjerred wrote: post #13
devmor wrote: post #10

Earlier quoted context omitted.

Unfortunately even the most prolific and very successful CEOs and management in our industry often fail to understand this. A couple years ago I was rejected from a 3rd round interview for a very large company that should have known better for the primary reason that my "coding challenge" exercise didn't meet an arbitrary lines of code minimum, despite meeting all of their requirements. And of course we have the rece…

> an arbitrary lines of code minimum, despite meeting all of their requirements. There's no way this is true right? Why would anyone care about LoC for a coding challenge?

If fizz buzz took you ten thousand lines, I'd care!

Re: I strive to be a 0.1x Engineer (2016)

#130

Post by hoosieree »

kqr wrote: post #112
ColonelPhantom wrote: post #62

Earlier quoted context omitted.

Especially as compiling often and proper indentation are often taken care of by running a linter and an autoformatter. I'll add "learning how to use a debugger", though.

In my interactions with beginners, adding more tools has never helped them learn faster. That goes for linters, autoformatters, debuggers, and version control. Manual formatting and printf debugging goes a long way, and takes only foundational understanding of the problem, not the manual for another tool. (Don't get me wrong: I love those tools. For a person who already knows them, they're a productivity multiplier.…

It's good to have a teacher who can recognize the moment when print-debugging is no longer sufficient and a real debugger will solve the problem immediately.

Introducing debuggers too early leads to lots of "why do we need this? I can just print..."

It's a catch-22 though, because ideally you'd learn the debugger in an intro course, but intro course problems can usually be debugged with simple print statements.

Post reply on HN