I really want to write a post on why I think Kafka, Nginx, and Centrifuge all make a lot of sense to me, and I wish we had lib versions of all of them. But too tired to do that rn lol.
-
Show this thread
-
Okay fine, I think I can do a lazy tweet version of this. I want to summarize some features I like about all of the above, and what that might look like in lib form.
1 reply 0 retweets 0 likesShow this thread -
So Nginx is pretty well known — reverse proxy. Basically what it does is provide a single entry point into your system. A lib version of this would mostly be about SSL resolution, and forwarding connections (including wss://). Extra points for letsencrypt + live route updates.
1 reply 0 retweets 0 likesShow this thread -
For me Kafka is about removing the need for caches (like ActiveRecord/Redid) and migrations. Instead you have an immutable event log that you can process differently per use-case. The lib version of this would probably need 1. a log (hypercore maybe), 2. materialized views.
1 reply 1 retweet 2 likesShow this thread -
Finally Centrifuge. It's a new project by Segment which handles retries to services outside your system. Think email providers, or analytics processors. This is probably some form of smart queue + good retry strategies + VCR-style testing.
2 replies 0 retweets 0 likesShow this thread
And that's most of it. Thanks for tuning in to our brain dump special. Our regular programming will resume shortly.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.