Hacker News Reader

Comment by retSava | Hacker News Reader

I haven't seen that particular stack and build setup to dispute or support that claim. Have you?

What I have seen in my part of the developer world, is very brittle stacks/setups where minor changes have butterfly effects on completely unrelated parts. Pushing straight to important branches, build depending on several pretty unrelated repos, unclear and messy dependencies, etc etc. I can totally understand such a claim.

View on HN

I don't remember Twitter being known for downtime before he bought it and the claim it needed a complete rewrite indicates a lack of experience in software development. Rewriting everything rarely ends up in better software than you had before.

Google ‘fail whale’

I wonder if Twitter is still running on Ruby on Rails. Those were the days...

In case this isn’t sarcasm: it isn’t.

For those who don’t remember that era: https://www.theregister.com/2012/11/08/twitter_epic_traffic_...

Shopify does somehow to this day.

The “fail whale” was a thing years before Elon’s acquisition.

The fail whale was a thing in the late 2000s, when Twitter was famously still on Ruby on Rails. After the early 2010s rewrite in Scala (?) it was pretty much stable.

I used to be friends with a group of developers at Twitter after their rewrite. They had no complaints about “brittleness” or stability generally.

We do know Elon isn’t anywhere near as smart as he likes to project in areas of modern computing, so I tend to not believe his claims.

This was the man who publicly called a senior developer at Twitter an idiot for saying something about the government using SQL.

Reply on news.ycombinator.com