Hacker News Reader

Comment by bovermyer | Hacker News Reader

I'm curious if Pulumi is interested in this.

they might well be! As well as all other TACO providers eg Spacelift, Scalr, Env0

Yes we are competing in the same market; customers using Spacelift probably wont be using Digger and vice versa. But we feel like having a community-friendly fork would benefit everyone, without harming anyone.

Why would they be? Pulumi talks to providers at the protocol level - contrary to the belief of many, Pulumi is not a Terraform wrapper.

I thought they used portions of Terraform's code.

Clearly, based on my downvotes, I'm mistaken and should be ashamed.

I think it was still a good question. Pulumi is relevant if we talk about cloud orchestration in general, and is a Terraform alternative. Terraform is an abstraction over cloud APIs and there isn't anything which would stop the development of a HCL -> Pulumi compiler for example.

One actually exists already, built into the CLI [1]!

[1]: https://www.pulumi.com/docs/using-pulumi/adopting-pulumi/mig...

> Clearly, based on my downvotes, I'm mistaken and should be ashamed.

Downvotes are nothing more than an indicator of how many people clicked the downvote button.

Last time I looked at their class docs for some of their infra resources, they very very very closely resembled their Terraform counterparts...

That was several years ago, though; might be stale.

It uses the _providers_ for Terraform, but not Terraform itself, and the licenses for those are unchanged.

There is an unfortunate meme that this makes it a “Terraform wrapper” like CDK for TF, which it does not.

Thanks for the clarification; I knew I was missing something.

Isn't that to be expected? I only used Terraform for AWS, but the Terraform resources correspond almost directly to those of AWS CloudFormation and AWS's API.

A sort of franken-Terraform that treats (some dialect of) HCL as yet another Pulumi frontend language would be interesting.

Reply on news.ycombinator.com