Add basic instructions to setup a Rome project

.github/CONTRIBUTING.md

@@ -16,6 +16,8 @@ $ scripts/dev-rome --help
 
 No dependency installation step is required as we check in our `node_modules` folder that contains only a copy of TypeScript and some definitions.
 
+Refer to [Getting Started](../docs/getting-started.md) for more usage documentation.
+
 ## Testing
 
 You can run the test suite with the following command:

README.md

@@ -38,6 +38,20 @@ See [CONTRIBUTING](.github/CONTRIBUTING.md) for more information.
 
 The current area of focus is **linting**. See the umbrella task [#20](https://github.com/facebookexperimental/rome/issues/20) for tracking.
 
+## Getting Started
+
+To setup Rome in a project, all you need is a `rome.json` file.
+
+```bash
+$ mkdir hello-world
+$ cd hello-world
+$ echo '{}' >rome.json
+```
+
+This file is used to configure Rome and indicates the boundaries of your project.
+
+See [Getting Started](docs/getting-started.md) for more usage instructions.
+
 ## Philosophy
 
 This list includes general ethos the project should abide by. This list is not comprehensive. Some of these are obvious but are stated for completeness.
@@ -62,4 +76,6 @@ This list includes general ethos the project should abide by. This list is not c
 
 ## Community
 
+Contribution and development instructions can be found in [CONTRIBUTING](.github/CONTRIBUTING.md).
+
 Additional project coordination and realtime discussion happens on our [Discord server](https://discord.gg/9WxHa5d). Remember that all activity on the Discord is still moderated and will be strictly enforced under the project's [Code of Conduct](.github/CODE_OF_CONDUCT.md).

docs/getting-started.md

@@ -0,0 +1,55 @@
+# Getting Started
+
+Rome requires a project configuration in order to operate. This can take three possible forms.
+
+- A `rome.json` file
+- A `rome.rjson` file (What is RJSON? See [#13](https://github.com/facebookexperimental/rome/issues/13))
+- A `rome` field on `package.json`
+
+This can just be an empty file. It's required in order for Rome to determine all the files in a project. This is important as when running the CLI, we build an in-memory file system listing in order to perform operations like module resolution.
+
+```bash
+$ mkdir hello-world
+$ cd hello-world
+$ echo '{}' >rome.json
+```
+
+## Commands
+
+Rome has a dozen different commands. Documented below are some more useful ones when testing functionality.
+
+### `lint`
+
+This command will lint a file with a set of default lints and display the produced diagnostics.
+
+When ran with no arguments, all JavaScript files in a project are linted.
+
+```
+$ rome lint file.js
+```
+
+### `compile`
+
+This command will compile a file with a set of default transforms. There is currently no options for this command to specify a subset of transforms.
+
+```
+$ rome compile file.js
+```
+
+### `parse`
+
+This command will parse a file and output a pretty formatted AST.
+
+```
+$ rome parse file.js
+```
+
+## Daemon
+
+Rome has an optional daemon. When starting the CLI, we'll check if there's a server running, and if there is, we'll connect to and that's where the request will be processed.
+
+You can run the daemon with `rome start`, and stop it with `rome stop`.
+
+The daemon allows Rome to maintain long lived memory caches which can drastically speed up operations. We intend to utilize this server for any LSP integration.
+
+When the CLI is ran without a running server, then we initialize a server inside the CLI process that's only used for the lifetime of the command.