logoalt Hacker News

tavavex • today at 7:38 PM • 3 replies • view on HN

A 1.25x improvement would already make it a no-brainer for a new project. I'm not saying it wasn't better for your use case, I'm saying we should be careful with tossing around words like 2000%. Everywhere you look in the software field today, you see people talk about 5x, 10x, 100x, never some 'normal' +20%.


Replies

brightball • today at 7:50 PM

There were a lot of reasons for it.

Java itself was a much slower standard workflow. Dynamic languages in general, like Ruby, Python, Perl or PHP were much faster to work with than the burdensome recompile workflows. Think instant feedback vs waiting for your CI to finish after every single save. The speed of iteration was critical.

Beyond that, when the speed of development is slower, more people get hired to do the work. When more people get hired to do the work, there's more communication relay as the team size grows. Teams coordinating between each other. Decisions going to committees. Middle management layers.

You'd routinely see companies with 3-5 Rails devs outperforming shops with 50 Java devs. Communication and iteration speed was huge. The other piece of it was the convention over configuration aspect because it removed so many things you'd have to discuss and make decisions about as a group which further accelerated communications.

10-20x multiplier isn't an exaggeration at all. If anything I could be underselling it.

buu700 • today at 8:12 PM

I see both types of claims a lot on HN, and find both credible.

On one extreme, a solo dev could prompt a frontier SOTA agent with not much more than "Build Uber", give it the necessary hardware/software/permissions to build itself a useful feedback loop and act autonomously, put it in goal mode, and let it cook, and as long as the tokens kept flowing they'd eventually end up with something reasonable in actual orders of magnitude less time than it would have taken that dev to build the same thing — especially if you're counting calendar days rather than hours worked, and extra especially if they wouldn't have had the luxury of investing full-time hours into that particular thing. At the opposite extreme, BigCo full-time SWE #5000 tasked with an extremely specific ticket for a critical revenue-generating brownfield system might experience even "20%" as liberal, and correctly observe that writing code was never the bottleneck for their job. And then there's the massive spectrum of circumstances in between those two ends.

None of that means that the gains aren't real, or that anyone in particular is wildly exaggerating in either direction. It just means the gains aren't evenly distributed.

IsTom • today at 8:12 PM

It might be different from Java shop to Java shop, but I'll never forget that time I changed jobs from something more productive to Java and my first day way sprint planning. It all made sense until the end when they told me that we were estimating in days, not hours and it turned out that it really took that much longer to do things there.

➕ show 1 reply