Hacker News Reader

Comment by 3np | Hacker News Reader

> Node.js is too easy to get started. This means that the pool of available programmers is not the highest quality. Runtimes like Go or Deno are still havens for the ‘connoisseur’ programmer.

wat

(Deno seems worth checking out though!)

View on HN

"Too easy to get started" is just gate keeping.

On one hand, JavaScript being easy makes programming accessible. On the other hand, the state of programming is terrible and getting worse. So you can't say if it's bad or not.

> On the other hand, the state of programming is terrible and getting worse.

What does this mean? It's an order of magnitude easier to build an app/service/whatever today than it was a decade ago. Having to maintain separate code paths for IE because it doesn't support many of the APIs and CSS features you need was "terrible"; by comparison, engineering today is heavenly.

Engineering might be easier, the user experience is magnitudes worse than a decade ago, pretty much universally.

it's not getting worse. you're just getting older

100%. If people want to still use Wordperfect 5.1, they still can.

It’s not getting worse. Go ahead and make a relatively complex app with jQuery 2.0 and let me know how it compares to spinning up a Next.js app.

Or better yet, you could use neither! If it helps you remember, JavaScript is bad, so more JavaScript is more bad.

Reverse gate keeping, yeah. You built a fence around your safe zone, and locked yourself in.

I've been playing with Deno lately and I can assure the author, I'm no connoisseur

This means at the level of their ability to draw in candidates, Deno interest is signal to their hiring pipeline of more quality. The consequence is their hiring pool is also that much smaller. One might say the same thing about Java v Go.

> The consequence is their hiring pool is also that much smaller.

In the current market, unless you're a big company that gets flooded with applications on a daily basis, why would you ever reduce your hiring pool arbitrarily? If you're a 13-person startup with good funding, you want all the candidates you can possibly get. Excluding potentially great engineers because they've never worked with Deno doesn't make any sense.

It does make sense, because you filter out the non enthusiasts, a proxy for great programmers. When Rust was still nascent, the people interested in it were likely to be of a higher quality bar than any regular old Javascript dev. That's not to say that JS is bad necessarily, just that enthusiasts correlate to great programming skills.

Especially if you're a product-focused startup, the last thing you may actually need is an enthusiast for a language. 90% of the time you want people that just want to use the right tool for the job.

We're contending with the reasoning behind an observation, but the observation is already useable to the author of the post — if true. Is the author of the post seeing improvements in candidate quality because Deno is actually a proxy for burning the midnight oil? Who knows. If we generalize to Java vs Go will the observation hold? Who knows.

But as chrisco255 pointed out, neither usage of Deno nor applying a weight to some observation will reduce your hiring pool.

I would also say that a technological perspective of "right tool for the right job" is somewhat independent to the expensiveness of your hiring pipeline. Sometimes Erlang is the right tool for the right job, but that undoubtedly changes the experience of hiring.

My goodness, if you're excluding any hire because they've never worked with (insert new hotness here) you are simply making a big mistake in hiring, in general.

Deno is still JS based at the end of the day and attempts to conform to web standards in its design. I don't know why in this particular case it would limit your pool.

I could see how Golang vs Deno (JS) would impact your hiring pool dramatically though.

As chrisco255 pointed out downthread, I made a mistake in saying that Deno would reduce your hiring pool. That Deno is a signal in hiring does not mean you've lost out on Node talent, it just means that you've found a net gain in hiring by weighing Deno more strongly (assuming the blog post is correct).

However, Java vs Go would probably make a big difference in terms of hiring pool size.

Yea, this is one of the most garbage takes I've ever heard. Requiring the use of something esoteric is not suddenly going to allow you access to a pool of higher quality candidates. Instead you will still see the same pool, but now the average candidate will be even less knowledgeable about the technology.

I do like Deno personally, but this reason is not a great one to do so.

Same is true about Lisp and was true about Python in early 00s. Rare tech brings out enthusiasts.

Really? I never noticed any particularly high barrier of entry with either.

Pygame for example was very popular with hobbyist game programmers in the early 2000s as it provided a much easier way to get started with game programming than C, which was the most popular alternative back then (Unity 3d only released in 2005 and Unreal Engine only became free in 2015).

What I do remember from the early 2000s, though, was legions of aspiring game programmers struggling with C and C++.

I've tried Deno. The author might rethink that line after interviewing me.

Really though this seems almost like resume keyword checking level of a candidate quality check. Actually it sounds exactly like that...

Reply on news.ycombinator.com