Hacker News Reader

Comment by dan-robertson | Hacker News Reader

Google is famous for having amazing dev tools. If any (large) project is going to have this most-important thing, it is likely to be best able to get it at Google. And yet Google also has a reputation for low developer productivity. So, what gives? Are some of my premises wrong, is the most important thing not close to sufficient, is Google actually productive?

View on HN

Google had amazingly below-average dev tools when I was there. It was standard in my org to have multi-minute builds, with input/output streamed to/from a remote server. Yes, every keystroke had a small but perceptible delay. I was working on an outlier project that took over 30 minutes to build for a single-character change. I never learned as slowly as I did during my time there. Startups generally have faster tech, bigcos generally have more customized tech with more options and that can operate ‘at scale.’

Funny story. Sourcegraph itself (the company the author belongs to and cofounded) started to bring the awesome code search internal tool at Google to the world as a product.

If productivity could be broken down to an equation, it would certainly have more than one explanatory variable.

If the relative value of things tends to follow a power law (this seems to be true in many cases) then one would expect the best few things to be much more important than the rest. So even though many things contribute to the result, the most important things may still be very important.

Reply on news.ycombinator.com