11 ms·

> good junior devs can code not only "just as fast", but probably faster too! Pfft. I'm about as "senior" as it gets and I'm generally the fastest developer i

> good junior devs can code not only "just as fast", but probably faster too! Pfft. I'm about as "senior" as it gets and I'm generally the fastest developer in any team I'm on. I've never met a junior dev that could hold a candle to me, and most senior devs I know are the same...we're coding monsters.

darkerside 9y ago
> I'm about as "senior" as it gets Senior developers make sure to question their assumptions on a regular basis, FYI
moreless 9y ago
True. But the team as a whole would probably still achieve much more if you were mostly guiding other (less experienced) devs.
- g051051 9y ago
  I do. But I'm still faster than them in head to head coding throughput (code produced, error counts, etc). I've got decades of experience in developing software. No way someone fresh out of college is going to equal, let alone out-code me. That's not to say they're not good, but the assertion: > can code not only "just as fast", but probably faster too! is pure BS.
  - virgilp 9y ago
    I am flabbergasted that you're senior and measure your productivity in "code produced". What is that, LOC? Can type faster than a recent graduate? Good junior devs will be more skilled in some areas than you are. Maybe it's stuff you're not interested in; or just stuff they're really interested in. The only way a good junior dev doesn't "out-code" you, ever - is if you're only skilled in a very narrow niche (bonus points if only few people are interested in it, at all). Junior devs absolutely can, and often do, build "wrong things" faster than senior devs. The measure of seniority is in my mind about knowing what to NOT build, in the first place.
    - g051051 9y ago
      > I am flabbergasted that you're senior and measure your productivity in "code produced". What is that, LOC? Can type faster than a recent graduate? Who uses LOC? I'm talking about completed, tested, accepted features, as defined by our project teams. > Good junior devs will be more skilled in some areas than you are. Well, yeah...of course. I'm not comparing myself to someone who works in an entirely different field. A junior front-end dev will be better than me at front-end stuff. I'm talking about a junior in my area, who I'd be in a team with or would mentor. > The measure of seniority is in my mind about knowing what to NOT build, in the first place. I agree with that statement.
      - virgilp 9y ago
        > I'm talking about completed, tested, accepted features, as defined by our project teams. Still rather strange metrics for productivity in a senior engineer. This is more what I'm taking about: https://zef.me/the-100x-engineer-6d50a690a866 https://zef.me/the-100x-engineer-6d50a690a866 If you're really senior, you shouldn't be working on the kind of features that get delivered at a rate of "5 per sprint". More like on stuff that gets delivered once every year. The junior SHOULD outperform you in "code produced" - they just shouldn't outperform you in dollars produced (or saved).
        g051051 9y ago
        You do what your team and company needs. If they need me to work on a big re-engineering project, or build a core framework feature, I do it. If they need me to get onto a regular sprint team for a while and churn out the backlog, I do it.
  - milesvp 9y ago
    I believe it. Though my experience is that junior devs can be much faster at producing lines of code, and most of it's junk. Senior devs will always be faster at producing bug free lines of code, and certainly at catching more of the edge cases. Moreover, the best devs I've worked with know when not to code at all to solve a problem, which is a totally foreign concept to junior devs who seem to prefer coding to thinking. When I look at code written by junior devs I often think 'dear god, how is there so much code that does nothing, and how did it all get witten since I last had a chance to look'. Makes me wonder if there are parallels in writing. My understanding is that the most prolific authors can only write 8 publishable pages a day. I'd bet that amateur writers can produce more pages than that, but no page could be published without at least as much time spent by an editor.
    - moreless 9y ago
      > dear god, how is there so much code that does nothing, and how did it all get witten since I last had a chance to look This, 100 times over! It is simply incredible how much code and how complex systems a junior dev (junior by knowledge, not by years of coding) can produce in a short time while you were looking the other way... :-D
purerandomness 9y ago
Dunning-Kruger called, they want their overblown ego steamroller back.
- musage 9y ago
  The Dunning-Kruger effect is a bias and as such deals with probabilities. It doesn't mean any given individual says they're the best (in their neck of the woods) at something is definitely wrong.
- Retric 9y ago
  People get better with practice. So, yes senior developers tend to be significantly faster coders. It's not that they do everything faster, but by getting better at the hard parts they finish complex tasks vastly more quickly. PS: Most people have never worked closely with someone with 25+ years of experience, but the difference is staggering.
  - virgilp 9y ago
    Well, I am someone with 25 years of experience, does that count as "worked closely with"? Good junior devs are... well, by definition, good. They're just younger, and less experienced - but every bit as smart as you are, and pretty skilled in the art of coding. If anything, they have more time than I do and are a lot more eager to "prove themselves". I don't view the value I add as "can code features faster"...
    - g051051 9y ago
      > I don't view the value I add as "can code features faster" I do, at least partly. I can code so much faster that my cost per feature is lower than a junior dev.
      - ranci 9y ago
        Your cost is less by being paid more? Seems legit
ataturk 9y ago
Welp, get back to me when you're 50 Mr. Senior and we'll see how you fared over the long haul. Because "coding monster" or not, when you hit 40, companies will be less enthusiastic about hiring you, and by the time you're 50, you'll be out, so I hope you're making bank right now and saving as much of it as you possibly can because there are no do-overs.
- g051051 9y ago
  Really? I'm 54, and constantly have to turn away recruiters who pester me. Long haul is looking pretty good, right now anyway. Even if you're a junior dev, outlooks can change and shift...it's all about staying relevant and proving your worth.
corpMaverick 9y ago
Sr Dev with 26 years of experience. I get my edge not by coding fast. I maximize the amount of work that we don't have to do. - Better (usually simpler) designs. - Breaking the problem and choosing the order in which things are implemented. - Setting up usefull feedback loops. - etc, etc. It is a lot about judgment.
- g051051 9y ago
  A deep understanding of your tool set is a huge advantage.
- nawitus 9y ago
  That is "coding fast". Coding fast doesn't mean literally churning out a lot of code and pull requests.
  - TheCoelacanth 9y ago
    Managers often have trouble recognizing the difference, though. Even if they know the difference, measuring actual productivity is difficult and subjective, measuring number of pull requests or number of JIRA tickets closed is easy and objective.
    - TimTheTinker 9y ago
      > easy and objective. that’s the worst kind of data - gives a meaningless feeling of objectivity.
xigency 9y ago
What about cross-discipline? What if a problem comes up that would be better solved with technology Y instead of X and you are an X master but random intern happens to know Y. Are you still the fastest? What if the intern is better at making slides for code reviews that show important trends? Or they can write fantastic documentation? Or tools? I'm sure you're a quick learner but at least in scenarios I've worked in there are often many, many possible solutions and no specific "right answer." After all, it's hard for anyone to be right all of the time.
- g051051 9y ago
  None of that has anything to do with the original blanket assertion of "good junior devs can code not only "just as fast", but probably faster too!".
  - xigency 9y ago
    Aye, but you probably can't code faster if you have to read a manual first? It is easy to construct scenarios where someone from a bootcamp could code "faster" than a principal, senior lead, or CTO of all things. It's just a matter of picking the problem. A better point is that this drought of new talent is just the other side of the pendulum. In the odd years, the problem is layoffs and difficulty finding work without being a new-grad or knowing the latest tech fad. You've seen some of those summers, too?
    - g051051 9y ago
      > Aye, but you probably can't code faster if you have to read a manual first? If we both had to read the manual? Yeah, I'd still be faster. It's just as easy to cherry-pick for one side or another.
    - virgilp 9y ago
      The general point he makes is that "if I define what speed means, I always code faster", so yeah, he's right. Sort of :)

darkerside 9y ago

> I'm about as "senior" as it gets Senior developers make sure to question their assumptions on a regular basis, FYI

moreless 9y ago

True. But the team as a whole would probably still achieve much more if you were mostly guiding other (less experienced) devs.

g051051 9y ago
I do. But I'm still faster than them in head to head coding throughput (code produced, error counts, etc). I've got decades of experience in developing software. No way someone fresh out of college is going to equal, let alone out-code me. That's not to say they're not good, but the assertion: > can code not only "just as fast", but probably faster too! is pure BS.
- virgilp 9y ago
  I am flabbergasted that you're senior and measure your productivity in "code produced". What is that, LOC? Can type faster than a recent graduate? Good junior devs will be more skilled in some areas than you are. Maybe it's stuff you're not interested in; or just stuff they're really interested in. The only way a good junior dev doesn't "out-code" you, ever - is if you're only skilled in a very narrow niche (bonus points if only few people are interested in it, at all). Junior devs absolutely can, and often do, build "wrong things" faster than senior devs. The measure of seniority is in my mind about knowing what to NOT build, in the first place.
  - g051051 9y ago
    > I am flabbergasted that you're senior and measure your productivity in "code produced". What is that, LOC? Can type faster than a recent graduate? Who uses LOC? I'm talking about completed, tested, accepted features, as defined by our project teams. > Good junior devs will be more skilled in some areas than you are. Well, yeah...of course. I'm not comparing myself to someone who works in an entirely different field. A junior front-end dev will be better than me at front-end stuff. I'm talking about a junior in my area, who I'd be in a team with or would mentor. > The measure of seniority is in my mind about knowing what to NOT build, in the first place. I agree with that statement.
    - virgilp 9y ago
      > I'm talking about completed, tested, accepted features, as defined by our project teams. Still rather strange metrics for productivity in a senior engineer. This is more what I'm taking about: https://zef.me/the-100x-engineer-6d50a690a866 https://zef.me/the-100x-engineer-6d50a690a866 If you're really senior, you shouldn't be working on the kind of features that get delivered at a rate of "5 per sprint". More like on stuff that gets delivered once every year. The junior SHOULD outperform you in "code produced" - they just shouldn't outperform you in dollars produced (or saved).
      - g051051 9y ago
        You do what your team and company needs. If they need me to work on a big re-engineering project, or build a core framework feature, I do it. If they need me to get onto a regular sprint team for a while and churn out the backlog, I do it.
- milesvp 9y ago
  I believe it. Though my experience is that junior devs can be much faster at producing lines of code, and most of it's junk. Senior devs will always be faster at producing bug free lines of code, and certainly at catching more of the edge cases. Moreover, the best devs I've worked with know when not to code at all to solve a problem, which is a totally foreign concept to junior devs who seem to prefer coding to thinking. When I look at code written by junior devs I often think 'dear god, how is there so much code that does nothing, and how did it all get witten since I last had a chance to look'. Makes me wonder if there are parallels in writing. My understanding is that the most prolific authors can only write 8 publishable pages a day. I'd bet that amateur writers can produce more pages than that, but no page could be published without at least as much time spent by an editor.
  - moreless 9y ago
    > dear god, how is there so much code that does nothing, and how did it all get witten since I last had a chance to look This, 100 times over! It is simply incredible how much code and how complex systems a junior dev (junior by knowledge, not by years of coding) can produce in a short time while you were looking the other way... :-D

purerandomness 9y ago

Dunning-Kruger called, they want their overblown ego steamroller back.

musage 9y ago
The Dunning-Kruger effect is a bias and as such deals with probabilities. It doesn't mean any given individual says they're the best (in their neck of the woods) at something is definitely wrong.
Retric 9y ago
People get better with practice. So, yes senior developers tend to be significantly faster coders. It's not that they do everything faster, but by getting better at the hard parts they finish complex tasks vastly more quickly. PS: Most people have never worked closely with someone with 25+ years of experience, but the difference is staggering.
- virgilp 9y ago
  Well, I am someone with 25 years of experience, does that count as "worked closely with"? Good junior devs are... well, by definition, good. They're just younger, and less experienced - but every bit as smart as you are, and pretty skilled in the art of coding. If anything, they have more time than I do and are a lot more eager to "prove themselves". I don't view the value I add as "can code features faster"...
  - g051051 9y ago
    > I don't view the value I add as "can code features faster" I do, at least partly. I can code so much faster that my cost per feature is lower than a junior dev.
    - ranci 9y ago
      Your cost is less by being paid more? Seems legit

ataturk 9y ago

Welp, get back to me when you're 50 Mr. Senior and we'll see how you fared over the long haul. Because "coding monster" or not, when you hit 40, companies will be less enthusiastic about hiring you, and by the time you're 50, you'll be out, so I hope you're making bank right now and saving as much of it as you possibly can because there are no do-overs.

g051051 9y ago
Really? I'm 54, and constantly have to turn away recruiters who pester me. Long haul is looking pretty good, right now anyway. Even if you're a junior dev, outlooks can change and shift...it's all about staying relevant and proving your worth.

corpMaverick 9y ago

Sr Dev with 26 years of experience. I get my edge not by coding fast. I maximize the amount of work that we don't have to do. - Better (usually simpler) designs. - Breaking the problem and choosing the order in which things are implemented. - Setting up usefull feedback loops. - etc, etc. It is a lot about judgment.

g051051 9y ago
A deep understanding of your tool set is a huge advantage.
nawitus 9y ago
That is "coding fast". Coding fast doesn't mean literally churning out a lot of code and pull requests.
- TheCoelacanth 9y ago
  Managers often have trouble recognizing the difference, though. Even if they know the difference, measuring actual productivity is difficult and subjective, measuring number of pull requests or number of JIRA tickets closed is easy and objective.
  - TimTheTinker 9y ago
    > easy and objective. that’s the worst kind of data - gives a meaningless feeling of objectivity.

xigency 9y ago

What about cross-discipline? What if a problem comes up that would be better solved with technology Y instead of X and you are an X master but random intern happens to know Y. Are you still the fastest? What if the intern is better at making slides for code reviews that show important trends? Or they can write fantastic documentation? Or tools? I'm sure you're a quick learner but at least in scenarios I've worked in there are often many, many possible solutions and no specific "right answer." After all, it's hard for anyone to be right all of the time.

g051051 9y ago
None of that has anything to do with the original blanket assertion of "good junior devs can code not only "just as fast", but probably faster too!".
- xigency 9y ago
  Aye, but you probably can't code faster if you have to read a manual first? It is easy to construct scenarios where someone from a bootcamp could code "faster" than a principal, senior lead, or CTO of all things. It's just a matter of picking the problem. A better point is that this drought of new talent is just the other side of the pendulum. In the odd years, the problem is layoffs and difficulty finding work without being a new-grad or knowing the latest tech fad. You've seen some of those summers, too?
  - g051051 9y ago
    > Aye, but you probably can't code faster if you have to read a manual first? If we both had to read the manual? Yeah, I'd still be faster. It's just as easy to cherry-pick for one side or another.
  - virgilp 9y ago
    The general point he makes is that "if I define what speed means, I always code faster", so yeah, he's right. Sort of :)