Hacker News Reader

Comment by eropple | Hacker News Reader

> frequent job hops can improve your salary, but they reduce your ability to stick with a long project and become good at anything

This is a major reason why I went into platform engineering and devops. Projects are very accelerated (usually because you end up starting too late, relative to the best time to change a dev team, its culture, and its practices) and you really can learn quickly and get better at a rapid pace. It helps, too, that the state of the art is changing quickly enough that effectively nobody is an expert at everything.

> Java engineers can set up multilateral bidding wars every 2 years and get huge pay bumps

...and this is the other reason. Best of both worlds, in my view.

View on HN

This is a major reason why I went into platform engineering and devops.

I can see the appeal of that. It also makes you less exposed to changes in window dressing and to political changes.

..and this is the other reason.

Sure, but it's also bad for the world that "market signals" are encouraging engineers to use bad languages for reasons having nothing to do with productivity. We know the real reason why there's so much Java in the software world: middle managers want headcount (they'd rather manage a team of 50 mediocrities than 5 elite engineers doing just enough work) and Java is such a low-productivity language that it forces teams to be big if anything is to be accomplished.

If an engineer who'd be nothing special in the Haskell community can take home $400k per year slinging Java, then what is a half-decent Haskell engineer (who, most likely, makes a lot less than $400k) really worth? $700k? $3 million?

> If an engineer who'd be nothing special in the Haskell community can take home $400k per year slinging Java, then what is a half-decent Haskell engineer (who, most likely, makes a lot less than $400k) really worth? $700k? $3 million?

The actual language isn't worth anything. Knowing the language that my business application is written in, is what counts. COBOL is the best known example - outdated, hard to use language with no features. Yet COBOL programmers are probably, on average, the best paid programmers out there, because they work on systems that the businesses care about.

The reason a 'nothing special Java dev' is worth $400k is because there are systems that make businesses crap-tons of money out there, written in Java. Not because Java is a good/bad language.

Is Haskell really that much more productive for a team? What are some large-scale projects written in it? It may make recruiting harder and take new joiners longer to get up to speed. You may not have easy access to as many libraries. A lot of "PL enthusiast" types would rather show off how clever they are than solve real problems. There are many valid concerns a manager could have that aren't headcount-related.

Actually, it makes recruiting easier. The ratio of good people eager for Haskell to Haskell jobs looks pretty favourable to employers.

You are right about some of the other concerns.

Sorry - I meant that in devops roles, those bidding wars exist too and I don't use Java. (Scala for web stuff, Ruby for plumbing.) Everyone needs devops, especially the people who don't know what it is. =)

Reply on news.ycombinator.com