Hacker News Reader

Comment by jonesb6 | Hacker News Reader

Energetic. Productive. Cheap. They want all three. Easier to get that when you target college students.

Indeed. The cheapness comes from inexperience, though. These fresh graduates have also not learned the dangers of overworking themselves (both to themselves and their employers), so everyone is very willing to continue this grand delusion.

As someone who developed a mild anxiety disorder after ~6-8 months of starting work as a SWE, this hits home pretty hard. Luckily I've been able to overcome with help from my family, friends, and coworkers, but whenever I try to warn others I fear I come off only as a doomsayer and immediately bounce off the "That won't happen to me" shield.

Seriously. If you bet on one company with a dozen kids writing in whatever framework has the most blog posts this week, vs three seasoned lispers, I'm betting on the lispers every time.

With experience, you also learn better ways to express your intent in the languages and frameworks you know well. Inexperienced, high energy devs will just be fast at writing boilerplate.

With experience, you learn how to be able to spot the latest fads that will likely be waning or gone in a few years once they're no longer shiny things. This can make you very unpopular with those people who take it for granted that shiny things and new things are almost certainly better than non-shiny, non-new things.

Remember when MongoDB, for example, was widely assumed (by the inexperienced, mostly) to be the obvious data storage tech for building virtually any web-facing application? Now, after wide experience with formerly-shiny Mongo, it's viewed as just one among many alternatives for reaching "web scale".

Productive? On the surface maybe, but any product with any lifespan will trend towards ball of mud very quickly if you only have greenhorn devs.

Exactly! Or even people who just haven't learned a whole lot in terms of engineering practices. This seems to be especially true in Ruby, where "it works" is good enough to go out into production.

There is no programming, reliability, stability, or scale. Most greener engineers program mostly for the sunny day scenario, which is the crux of the problem.

They can't see race conditions and timing problems like someone how have spent years doing that. I see other people's code and quickly understand that they're doing really cost prohibitive queries while others who are green will never notice it.

That's what I am thinking. Its taken me my fair share of making crappy choices in the past to know how to make better ones now.

> Cheap

For some definition of cheap. If you look at $/hour, no question they are cheaper. If you look at $/amount of value created (which is far harder to capture, but worth it), I don't think it is a slam dunk.

The difficulty is the financiers; they want visual bang for their buck, and they want to be told "yes, I'll make it happen" even if the person talking doesn't yet know how. If that means a bunch of young people looking busy and working themselves into the ground, versus a small team of professionals that will calmly and realistically tell the financiers what can't be done in a given time frame and budget... Realistic with older professional is boring; fantasy with young people is exciting! Guess what happens every time?

Collage graduates even at big name schools are rarely productive.

Reply on news.ycombinator.com