Glad to be helpful! Good luck :-)

I concur. The first two points you mention in my opinion balance each other out: even if Cloudformation fails "hard," as you say, at least you have the support that can help you (with Terraform you don't have any--and even it sometimes runs aground). As for transforming an organization's entire code base--it's probably not worth it. Also, I work with organizations that are just now starting to approach IaC, and they still choose Terraform because it's very simple and doesn't require developer skills, and it's easier to find staff (especially from Operations) who are able to use it.

It's funny because is true... 🫠

From a purely philosophical point of view, I could agree... but in the end, code execution still has to be launched from a CI pipeline and not from one's own PC. So instrumenting a docker with Node or a docker with Terraform makes no difference.

Thank you for sharing such outstanding insight! I like it very much and it will be super useful for me!!

Love the waterfall image! :-D

I started to hear something around on the IfC topic, but not being a developer I haven't tried any experiments. It was very interesting to read this post of yours, thank you for sharing it - it will certainly be useful to me!

Aaaaand just when we were talking about it... https://aws.amazon.com/blogs/aws/introducing-aws-resource-explorer-quickly-find-resources-in-your-aws-account/ Ta-daa 🚀

I've had some cases where Nuke has been helpful: https://github.com/rebuy-de/aws-nuke You can run it in dry-run mode and it is very comprehensive. Obviously it's just another DIY tool. I agree that AWS could be more helpful in this department.

Brilliant article. I was a bit sceptical about Pulumi/CDK but I'll definitely give it a try now :)