Hacker News Reader

So if the bookstore doesn't think it's worth it to spend more than 2 hours on deploying to the edge, maybe it's not that important to them, or :) ?

View on HN

I think fly.io believes that they can allow the bookstore to do so in less than two hours for a price where it's still important enough for the bookstore and profitable - in aggregate - for fly.io.

I personally believe that could be true, but their investors going to find out ;)

So change your infra to save a few hours :D I guess they have other compelling features.

Tell me you've never without telling me etc...

Roughly this, yes. It's easy to run boring apps close to your users. Thus, people buy computer time from us instead of a place that runs in one city.

I've long suspected this is what you guys would end up focusing on (and applaud it) -almost like an OS for edge compute that abstracts away its details while allowing boring architectures built for single-point to 'just run' on it.

But is there a limit to how boring you can go? Can you truly, fully abstract over distributed systems stuff like eventual consistency and its basically random effects on UX in certain usage edge cases?

Or do you think 'boring' will have to meet you in the middle with frameworks and developer mindset also shifting over time to a more edge-first world?

Reply on news.ycombinator.com