Update contributors.md

docs/contributors.md

@@ -3,20 +3,25 @@
 ### Checkin
 
 - Do checkin source (src)
-- Do checkin build output (lib)
-- Do checkin runtime node_modules
-- Do not checkin devDependency node_modules (husky can help see below)
+- Do checkin a single index.js file after running `ncc`
+- Do not checkin node_modules
 
-### devDependencies
+### NCC
 
-In order to handle correctly checking in node_modules without devDependencies, we run [Husky](https://github.com/typicode/husky) before each commit.
-This step ensures that formatting and checkin rules are followed and that devDependencies are excluded. To make sure Husky runs correctly, please use the following workflow:
+In order to avoid uploading `node_modules` to the repository, we use [zeit/ncc](https://github.com/zeit/ncc) to create a single `index.js` file that gets saved in `dist/`.
 
+### Developing
+
+If you're developing locally you can run
 ```
-npm install                                 # installs all devDependencies including Husky
-git add abc.ext                             # Add the files you've changed. This should include files in src, lib, and node_modules (see above)
-git commit -m "Informative commit message"  # Commit. This will run Husky
+npm install
+tsc
+ncc build
 ```
+An files generated using `tsc` will be added to `lib/` however those files also are not uploaded to the repository and are exluded using `.gitignore`.
 
-During the commit step, Husky will take care of formatting all files with [Prettier](https://github.com/prettier/prettier) as well as pruning out devDependencies using `npm prune --production`.
-It will also make sure these changes are appropriately included in your commit (no further work is needed)
+During the commit step, Husky will take care of formatting all files with [Prettier](https://github.com/prettier/prettier)
+
+### Testing
+
+We ask that you include a link to a succesfull run that utilizes the changes you are working on. For example, if your changes are in the branch `newAwesomeFeature`, then show an example run that uses `setup-python@newAwesomeFeature`. This will help speed up testing and help us confirm that there are no breaking changes or bugs.