Hacker News Reader

Comment by moreless | Hacker News Reader

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

View on HN

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.

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.

> 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.

> 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

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).

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.

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.

> 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

Reply on news.ycombinator.com