Hey everyone, Ben here.
If you’ve been following the journey of DEV and our open source project Forem, you know we’ve always been obsessed with we...
For further actions, you may consider blocking this person and/or reporting abuse
Good stuff :D
I've always wondered. when I upload an image to the thumbnail or body and delete it do y'all clean it up?
I have a feeling so many services end up storing images that no longer get served, I could be wrong though.
Really fascinating to see this from the infra side! 😊 As someone who mostly thinks about images in terms of "does the asset show up on screen," this was a good reminder of how much invisible cost and complexity sits behind something as simple as a post cover image.
The edge scripting fallback for broken images is such a nice detail too—turning a jarring "broken image" icon into something on-brand feels like exactly the kind of polish that's easy to overlook until you've been burned by it at scale. Appreciate you sharing the actual numbers and architecture behind the decision, not just the "we switched and it's better" version! 🌸
This is the kind of infra simplification more teams should talk about. Image pipelines quietly become cost and reliability problems: origin load, resize latency, cache fragmentation, storage duplication, and 2am operational ownership all add up.
You broke external images in the process and could save even more by unbreaking them.
Right now they're pinging the CDN which obviously doesn't work because they're external.
github.com/forem/forem/issues/22908
Smooth Sailing, this should be a new standard to get out of sluggish sites and high storage costs.
The Perma-Cache strategy is a massive win. Offloading egress and origin traffic to the edge like this is the smartest way to scale while keeping costs sustainable.
Can we bring in our CDN? so use Bunny APIs to optimise images and use CloudFront to distribute?