Unfortunately you can't always keep your bucket name a secret, if you use presigned URLs to let your users directly download objects...

It’s a good idea to use a reverse proxy like Nginx to handle the transfer through an internal redirect, which will hide the bucket name and authentication credentials.

But this means data transiting your network instead of being directly fetched from AWS infrastructure, which for me mostly defeats the purpose of presigned S3 URLs.

If your server is on AWS, then the egress cost should be identical whether the transfer originates from S3 or EC2. But if your server is on premise and your host charges for egress (or has impactful bandwidth restrictions), I can understand where you're coming from, as you could end up paying twice.

Introducing complexity, failure point, scalability issue just to work around broken billing. You are probably right but I feel it’s wrong architecturally. I always liked S3 precisely because I could offload file serving to them.

The presentation in the article is actually top-down, contrary to the title.

An alternative title could be "all the hoops we have to jump through because we don't have lightweight threads and Loom yet".

Your first point is incorrect, inline classes can have fields that are regular reference types.

Interesting; I must have “over-read” the restriction. You are correct.

Re-reading the doc, the restriction is that all fields of a value type must be final, because value objects do not have identity.

Make sure to read the second part, Language Model (http://cr.openjdk.java.net/~briangoetz/valhalla/sov/02-objec...).

Still no official Java driver? How unfortunate...

I know its not an official driver, but I maintain the Community Java driver here[1]. If its missing something you would need feel free to open an issue and I can add it for you. I have been using RethinkDB with java for a few projects and it works well enough for me.

https://github.com/dkhenry/rethinkjava