zlacker

Yes, yes, unified billing and security generally make up for how much Amazon's technical offerings suck donkey balls and those network effects are why we let them get away with pushing such god awful product. We all know how this works. We don't have to like it.

>>schmid+(OP)
> Amazon's technical offerings suck donkey balls and those network effects are why we let them get away with pushing such god awful product.

What are we comparing to here?

Google Cloud? Azure? Linode? Openstack? OVH? Oracle Cloud? Hetzner?

They all suck too, just in different ways.

>>schmid+(OP)
I know we'll never know the answer, but I'd love to see a break down of AWS billings from expensive misconfiguration of cloud settings. After the recent Accident Forgiveness post from an AWS competitor it does seem like an area that AWS happily accepts. Small bootstrapped start up hires a 3rd party "expert" suggested directly by AWS to be experts. The 3rd party doesn't actually know everything and causes a 3000x increase in monthly cloud expense. Small startup uses their paid AWS support which resolves the expensive issue in under 10 minutes. The 3000x increased bill is still due. </rant> That's a story as old as time (of cloud), but it's gotta be a nice way to earn bonus for product managers

>>michae+j1
Individually developed domain-specific products. Even ones that long predate the hyperscaler offerings are often much better... except for billing and security and integration. Not through any fault of the developers, but integrating those things takes work and the ability to reduce that work is a killer advantage of the hyperscalers. But the hyperscalers know this so they invest less in making the core product any good, so it isn't. Reduction in billing/security/integration suck -> increase in technical suck. That's my whole point.