noyaml VS json2jsii

Most commonly written in YAML, these files are large and complex to read and understand. And being written in YAML comes with its challenges (and quirks) since it is an additional programming language that devs need to learn.

YAML is kind of like C++:

> You like C++ because you're only using 20% of it. And that's fine, everyone only uses 20% of C++, the problem is that everyone uses a different 20% :)

https://eli.thegreenplace.net/2009/10/17/the-c-bashing-seaso...

The YAML footguns are too numerous to reproduce here, so here are some sources:

https://stackoverflow.com/questions/3790454/how-do-i-break-a...

https://www.arp242.net/yaml-config.html

https://noyaml.com/

Relevant: https://noyaml.com/

YAML and its ecosystem is full of footguns and ergonomics problems, especially when the length of the document extends beyond the height of a user's editor or viewport. Loss of context with indentation, non-compliant or unsafe parsers, and strange boolean handling to name a few.

It becomes even worse when people decide that static YAML data files should have variable substitution or control flow via templating. "Stringly-typed programming" if you will. If we all started writing JSON text templates I think a lot of people would rightly argue we should write small stdlib-only programs in Python, Typescript, or Ruby to emit this JSON instead of using templated text files. Then it becomes apparent that the YAML template isn't a static data file at all, but part of a program which emits YAML as output. We're already exposing people to basic programming if we're using YAML templates. People brew a special kind of YAML-templated devops hell using tools like Kustomize and Helm, each of which are "just YAML" but are full of idiosyncracies and tool-specific behaviour which make the use of YAML almost coincidental rather than a necessity.

Yes, sometimes people would prefer to look at YAML instead of JSON, in which case I suggest you use a YAML serialization library, or pipe output into a tool like `yq` so you can view the pretty output. In a pinch you could even output JSON and then feed it through a YAML formatter.

The Kubernetes community seems to have this penetrating "oh, it's just YAML" philosophy which means we get mediocre DSLs in "just YAML" which actually encode a lot of nuanced and unintuitive behaviour which varies from tool to tool.

Look at kyverno, for examle: it uses _parentheses_ in YAML key names to change the semantics of security policies! https://kyverno.io/docs/writing-policies/validate/ . This is different to what I think is the (much better ideas of) something like kubewarden, gatekeeper, or jspolicy, which allow engineers to write their policies in anything that compiles to WASM, OPA, and Typescript/Javascript respectively.

We engineers, as a discipline, have decades of know-how building and using general purpose programming languages with type checkers, linters, packaging systems, and other tools, but we throw them all away as soon as YAML comes along. It's time to put the stringified YAML templates away and engage in the ecosystem of mature tools we already to use to perform one simple task they are already good at: dumping JSON on stdout.

Let's move the control flow back into the tool and out of the YAML.

The webpage documenting some of the sharp edges of yaml is also displayed as an editable yaml document

https://noyaml.com/

It’s the least secure config format, even worse than XML IMO since it’s unsafe even with trusted inputs. https://noyaml.com/

The more I look at it, the more I think that Deno is actually a perfect fit for infrastructure as code.

The first reaction I get is usually "What, JavaScript!?". Here is why its better than it looks like.

Deno easily runs TypeScript without a compile step. TypeScript is a very mature, developer friendly language that was designed to model the super complicated types often seen in JS. This includes unions and intersections which help you model complex rules between optional properties, as well as template literal types which can help you restrict string constants. As it so happens, those same types of objects are present in our configurations.

TypeScript is also flexible about how much you'd like to model. Do you want only properties with really nice autocomplete with docs from the language server? Sure we can do just that. Do you want to only allow strings that look like a time duration for that property? Can be done too with template literal types. You choose where to invest your energy.

Building abstractions on top of IaC is seen as painful and obscure. TypeScript has tools to help you avoid that, such as api-extractor: https://api-extractor.com/ - it can enforce that all your function abstractions are documented and it can generate that documentation for you.

The larger ecosystem also comes with a lot of ready-made codegen tools that can help, from small modules such as json-schema-to-typescript (https://www.npmjs.com/package/json-schema-to-typescript) to larger mature projects such as jsii and https://github.com/cdklabs/json2jsii . They can be used to build the tooling to import things like CRDs and other external schemas.

What about running arbitrary code, launching rockets, turing completness? This is where Deno comes in, with its permissions model (https://docs.deno.com/runtime/manual/basics/permissions). You can allow only a subset of commands (lets say, `helm` and `kustomize`, while you're migrating away from them), a subset of (writable) directories, accessible network hosts - or none at all.

What about the pains of managing JS modules? Deno lets you import files directly from URLs. It also allows you to set private tokens in env vars to ensure the imports succeed: https://docs.deno.com/runtime/manual/basics/modules/private - as a result, you can manage your IaC libraries the way that makes sense to you, without dealing with the unwieldy package managers of the node ecosystem.