I think this is the closest I've seen to a concise answer to the 10X programmer's advantage. Earlier in my career, I noticed a fellow programmer who was quite noticeably faster at completing tasks than everyone else including myself. We also had a coding competition and I 'programmed' as fast as he did in direct comparison on the same task. So it turns out, the performance boost was not in raw coding speed but a lot of other things like not being distracted, being ok with a great but not the greatest design, getting to work and avoiding analysis paralysis. A sense of urgency from day one on a project.
When competing head to head for a fun competition, many of these things come naturally but when given 'time' it can be easy to fall back to less productive habits when a deadline is months away.
Like my mentor Steve Hayes once told me 'hands on keyboard'
>The programming community is extremely polarized about the existence or not of such a beast: who says there is no such a thing as the 10x programmer, who says it actually does not just exist, but there are even 100x programmers if you know where to look for.
I think it depends on your definition. If you compare with the average programmer in your company then you can easily be a 10x if you're working in a big company where the average abilities are low. But it becomes very hard in a smaller company where most of your colleagues are competent.
If you define it relative to a more global average I think it's too difficult to know what an "average" programmer even is.
Now if in that big company you have to work only on small parts of the whole it often becomes harder to be fast again, since your colleagues might be writing APIs with a lot of hidden complexity.
If however your task is more isolated it becomes easier to do good design.
> Teams that work well together can outperform groups of people who are collectively more talented because there is a multiplication factor for teamwork.
There's some great advice here about building a team culture that allows everyone to level-up and contribute their best. Some comments here have suggested that this isn't what 10x means but that's the point! What if we value the wrong kinds of people in our teams - the talented jerks? We should instead call out the value of those that help the whole team improve.
On top of the productivity gains, the value of building a culture like this is invaluable for a startup or team with a long-term vision. It builds a sense of camaraderie and environment for learning and growth that draws people in and keeps them around for a long time.
Jessica Livingston shares on the role she played in founding YC and building the culture. She points out some of the same kinds of ideas in one of her talks[1] showing that some of the soft skills and the social environment that founders build are way more important than they often realise.
> Invite discussion with your language by using words like “I think” and “maybe”. (I call this conversational style: "Maybe you should talk more like a woman?")
Aside from the fact that I did not understand that "talk like a woman" thing, is the author
advising us to stop the long-running advice of "speak precisely" with "say anything, even if you are not sure' with the only caveat being to let people know when you aren't certain of something?
That's like a public speaker telling me to say "uuhh..." in between my phrases!
> is the author advising us to stop the long-running advice of "speak precisely" with "say anything, even if you are not sure' with the only caveat being to let people know when you aren't certain of something?
You got it all wrong. The key point is that by adopting a conversation style where you cease to impose your beliefs as unquestionable facts and start to leave the door open for others to provide their observations and insight, you start to get observations and insights from others.
It's as simple as that.
And by the way, your personal opinion is not precision or an expression of a fact. Your personal opinion is just the conclusion you've made based on the information and experience you had up to that point. If you cease to be defensive about your opinions when sharing them with others, others will follow suit.
I don't think the advice entails lower quality discourse. The prescription here, read literally, is to say the same thing, phrased in a way that invites relevant discussion.
Language emphasising subjectivity can do several things - a) you signal uncertainty so that the group has a better idea of what's really known, b) you explicitly distinguish opinion from fact, c) you invite people to share their understanding so you can get on the same page.
The emphasis seems to be less the informational (a) and (b) and more the cultural (c). It's tricky - in some sense if you're certain, you undermine the force of your opinion this way. Comparing the two communication styles there's a tradeoff between reducing friction (unnecessary? examination) and encouraging communication in general, with possible benefits for cohesion, shared understanding and expertise but also the possible detriment of communal navel gazing. If you trust the studies this refers to, encouraging equal turn-taking overall has long-term benefits for productivity.
I think both approaches have failure cases where communication becomes ineffective. Individuals will adapt to each style to different degrees, and there's no total substitute for savvy folks who are alert to social forces at play, and guide discussions when team members over- or under-play their hands.
To share the generalisation used by this piece, the suggested conversational style is more hospitable to women. Where this article talks about psychological safety, equal participation etc., I think a lot of male engineers feel safe in much more confrontational and argumentative situations than women - they can even find a kind of fun in conflict, or regard it as totally impersonal, where many women would find it deeply unpleasant and to be avoided at all costs.
Maybe framing it different might help. “Be honest with yourself and others about how certain you are about your facts, and never state opinions as facts.”
If you do this, people will more readily challenge the ideas they have evidence against and trust you when they don’t. It also allows you to throw more ideas into the pot since you don’t always have to know the perfect answer before saying something.
When working as a part of a group on a complicated problem, it is valuable to throw in information of which you only have say 70% confidence, provided you qualify it with an explicit statement of such confidence ("I think", "maybe"). This is especially relevant in problems involving people/teams in large organizations - there it's often hard to know much for sure and planning becomes an exercise in probabilities and speculation.
Here are some serious outstanding problems in software:
* Considering how fast modern computers are, software is so damn slow.
* Software frequently stops working. When it does, the process to diagnose the bug and fix it conclusively is not a straightforward one.
* Software is poorly understood. Most developers today work with other people’s abstractions. It’s rare to find a software system that someone can describe the inner workings of end to end.
* Software is excessively complex. To understand a codebase, you might have to crawl through ten layers of dependencies, imports, and scaffolding before finding code that actually does anything.
These are just some obvious ones off the top of my head. Does following the advice on this list make me any better at tackling any of these problems?
Could someone conceivably do none of the things on this list, and still be a very skilled developer who makes strong strides toward solving these problems?
> These are just some obvious ones off the top of my head.
But they are all the same one (the last form is the clearest statement), restated in different ways or viewed through effects that the same source causes.
Psychological safety, I think, comes with not being afraid of unexpected actions or consequences of actions, and from knowing that not everything is on the line all the time, ie it is ok to make mistakes. Safe = not having to fear being fired or worse due to either mistakes or unexpected reactions of bosses and peers.
You can do accountability while having that safety by (kindly) informing people that they will be expected to do a task that they accept/are given, and being clear about other expectations like deadlines, BEFORE they start the task. Give people the informantion needed to predict the result of their actions. And by growing a culture where people know that it is bad not to uphold an agreement (and that may include some kind of “punishment”, which ofc is known to everyone beforehand). It is not bad to do you best to solve a task but make a mistake doing it. That should not have consequences.
Every developer should do all of these things, the behavior and norms described in this blog post seem like the absolute minimum to have a healthy work environment.
I know a >250x programmer. That is strictly objective: on a 500-engineer 6-month death-march project, fully half the code delivered was his. He knows somebody who he estimates at 10x his own rate, and who wears out 2 keyboards a year (or did, 20 years ago).
(You might object that this makes him a 500x programmer, but he doesn't agree: a fair bit of code was written that did not end up in the product, an unknown fraction of which was good.)
I'm curious about what was the technology stack for that project. Also, it would be very interesting to learn what those ">250x" programmers think on various languages / technology stacks (that they use) from the developer productivity perspective ...
[+] [-] reese_john|6 years ago|reply
I think Antirez[1] has more practical advice on how to actually become a "10x programmer"
Key points:
* Experience: pattern matching
* Knowledge: some theory is going to help
* Low level: understanding the machine
* Debugging skills
* Perfectionism, or how to kill your productivity and bias your designs
* Design sacrifice: killing 5% to get 90%
* Simplicity
* Focus: actual time VS hypothetical time
[1] http://antirez.com/news/112
[+] [-] djmips|6 years ago|reply
When competing head to head for a fun competition, many of these things come naturally but when given 'time' it can be easy to fall back to less productive habits when a deadline is months away.
Like my mentor Steve Hayes once told me 'hands on keyboard'
[+] [-] carlmr|6 years ago|reply
I think it depends on your definition. If you compare with the average programmer in your company then you can easily be a 10x if you're working in a big company where the average abilities are low. But it becomes very hard in a smaller company where most of your colleagues are competent.
If you define it relative to a more global average I think it's too difficult to know what an "average" programmer even is.
Now if in that big company you have to work only on small parts of the whole it often becomes harder to be fast again, since your colleagues might be writing APIs with a lot of hidden complexity.
If however your task is more isolated it becomes easier to do good design.
[+] [-] benjaminjosephw|6 years ago|reply
There's some great advice here about building a team culture that allows everyone to level-up and contribute their best. Some comments here have suggested that this isn't what 10x means but that's the point! What if we value the wrong kinds of people in our teams - the talented jerks? We should instead call out the value of those that help the whole team improve.
On top of the productivity gains, the value of building a culture like this is invaluable for a startup or team with a long-term vision. It builds a sense of camaraderie and environment for learning and growth that draws people in and keeps them around for a long time.
Jessica Livingston shares on the role she played in founding YC and building the culture. She points out some of the same kinds of ideas in one of her talks[1] showing that some of the soft skills and the social environment that founders build are way more important than they often realise.
[1] - https://www.youtube.com/watch?v=8d-cApFHjeY
[+] [-] Gravyness|6 years ago|reply
Aside from the fact that I did not understand that "talk like a woman" thing, is the author advising us to stop the long-running advice of "speak precisely" with "say anything, even if you are not sure' with the only caveat being to let people know when you aren't certain of something?
That's like a public speaker telling me to say "uuhh..." in between my phrases!
[+] [-] rumanator|6 years ago|reply
You got it all wrong. The key point is that by adopting a conversation style where you cease to impose your beliefs as unquestionable facts and start to leave the door open for others to provide their observations and insight, you start to get observations and insights from others.
It's as simple as that.
And by the way, your personal opinion is not precision or an expression of a fact. Your personal opinion is just the conclusion you've made based on the information and experience you had up to that point. If you cease to be defensive about your opinions when sharing them with others, others will follow suit.
[+] [-] al-king|6 years ago|reply
Language emphasising subjectivity can do several things - a) you signal uncertainty so that the group has a better idea of what's really known, b) you explicitly distinguish opinion from fact, c) you invite people to share their understanding so you can get on the same page.
The emphasis seems to be less the informational (a) and (b) and more the cultural (c). It's tricky - in some sense if you're certain, you undermine the force of your opinion this way. Comparing the two communication styles there's a tradeoff between reducing friction (unnecessary? examination) and encouraging communication in general, with possible benefits for cohesion, shared understanding and expertise but also the possible detriment of communal navel gazing. If you trust the studies this refers to, encouraging equal turn-taking overall has long-term benefits for productivity.
I think both approaches have failure cases where communication becomes ineffective. Individuals will adapt to each style to different degrees, and there's no total substitute for savvy folks who are alert to social forces at play, and guide discussions when team members over- or under-play their hands.
To share the generalisation used by this piece, the suggested conversational style is more hospitable to women. Where this article talks about psychological safety, equal participation etc., I think a lot of male engineers feel safe in much more confrontational and argumentative situations than women - they can even find a kind of fun in conflict, or regard it as totally impersonal, where many women would find it deeply unpleasant and to be avoided at all costs.
[+] [-] Olreich|6 years ago|reply
If you do this, people will more readily challenge the ideas they have evidence against and trust you when they don’t. It also allows you to throw more ideas into the pot since you don’t always have to know the perfect answer before saying something.
[+] [-] badpun|6 years ago|reply
[+] [-] ravenide|6 years ago|reply
* Considering how fast modern computers are, software is so damn slow.
* Software frequently stops working. When it does, the process to diagnose the bug and fix it conclusively is not a straightforward one.
* Software is poorly understood. Most developers today work with other people’s abstractions. It’s rare to find a software system that someone can describe the inner workings of end to end.
* Software is excessively complex. To understand a codebase, you might have to crawl through ten layers of dependencies, imports, and scaffolding before finding code that actually does anything.
These are just some obvious ones off the top of my head. Does following the advice on this list make me any better at tackling any of these problems?
Could someone conceivably do none of the things on this list, and still be a very skilled developer who makes strong strides toward solving these problems?
[+] [-] dragonwriter|6 years ago|reply
But they are all the same one (the last form is the clearest statement), restated in different ways or viewed through effects that the same source causes.
[+] [-] commandlinefan|6 years ago|reply
Conversely, you could do everything on this list, and still be a lousy programmer. Or not even a programmer at all.
[+] [-] peter_d_sherman|6 years ago|reply
"1. Create an environment of psychological safety"
"6. Hold yourself and others accountable"
Now, not to get all lawyerly (I am not a lawyer) or anything, but...
How exactly do you do #1 if it conflicts with #6?
?
And...
How exactly do you do #6 if it conflicts with #1?
?
[+] [-] Msurrow|6 years ago|reply
You can do accountability while having that safety by (kindly) informing people that they will be expected to do a task that they accept/are given, and being clear about other expectations like deadlines, BEFORE they start the task. Give people the informantion needed to predict the result of their actions. And by growing a culture where people know that it is bad not to uphold an agreement (and that may include some kind of “punishment”, which ofc is known to everyone beforehand). It is not bad to do you best to solve a task but make a mistake doing it. That should not have consequences.
[+] [-] downerending|6 years ago|reply
[+] [-] al-king|6 years ago|reply
That said, it's certainly a bait-and-switch headline. Kind of par for the course.
[+] [-] dang|6 years ago|reply
[+] [-] mpoteat|6 years ago|reply
[+] [-] ncmncm|6 years ago|reply
(You might object that this makes him a 500x programmer, but he doesn't agree: a fair bit of code was written that did not end up in the product, an unknown fraction of which was good.)
[+] [-] ablekh|6 years ago|reply
[+] [-] smabie|6 years ago|reply
[+] [-] commandlinefan|6 years ago|reply
[+] [-] lonelappde|6 years ago|reply