- Take some inputs
- work out what images are wanted; thus what URLs
- return urls with appropriate metadata
- Take inputs as command line opts and/or config file (and/or guesswork from location?)
- work out file paths
- filter according to needed or not
- Pass to lib for urls
- download from urls (give output while doing so)
- write to file paths
- Take inputs from web interface (ELM?)
- Get image(s) or pass back to browser/ELM?
- ...