Add comments to build.sh file for easy memory jogging

bin/build.sh

@@ -1,13 +1,27 @@
 #!/bin/bash
 set -ex
 
+# change working directory to project root directory
+# (this file lives in /bin)
 cd "$(dirname ${BASH_SOURCE[0]})"/..
+
+# make a new directory to hold the build
 mkdir -p ./release
 
 # adapted from https://lebenplusplus.de/2018/03/15/how-to-run-npm-install-as-non-root-from-a-docker-container/
+# creates a temporary password for a one-time, non-root user (to avoid file overwrite permissions issues)
 echo "node:x:$(id -u):$(id -g)::/home/node:/bin/bash" > /tmp/fake-passwd
 
 # let gulp build the assets
+
+# -rm : remove the docker file system after this command exits
+# --label=gulp : meta data, label for the docker container while running
+# --volume=$(pwd):/srv : mount the current directory (which is the root of the project due to the `cd` command above) to the `srv` directory in the docker container
+# --volume=/tmp/fake-passwd:/etc/passwd : mount the temporary user password created above to `/etc/passwd` in the docker container
+# --volume=/tmp:/home/node : mount local file system's `tmp` directory to `/home/node` in the docker container
+# -u $(id -u):$(id -g) : use the current user and group inside the docker container
+# huli/gulp : name of the docker image to run
+# build --production : command sent to gulp inside the docker container
 docker run --rm --label=gulp \
        --volume=$(pwd):/srv \
        --volume=/tmp/fake-passwd:/etc/passwd \
@@ -17,6 +31,9 @@ docker run --rm --label=gulp \
        build --production
 
 # and then let hugo build the site
+
+# -e HUGO_ENV=production : tell hugo to build a production-ready version of the site
+# ash -c "rm -rf..." : remove any old build artifacts and rebuild the site using the config matching the command line argument ($1)
 docker run --rm --label=hugo \
     --volume=$(pwd):/src \
     --volume=$(pwd)/release:/tmp/build \