Scrape and Parse Trademark & Patent Data into a DB

General information for the business: We collect data
Description of requirements/functionality: IMPORTANT! I will not respond to any form letters or lists of projects that you've done in the past. What I specifically want to see from any proposals is your approach to the problem. Anything other than that will be ignored.

I'm looking for someone to build a scraper to pull down and parse all of the trademark and patent data from this Google Books repository http://www.google.com/googlebooks/uspto.html into a database.

The scraper should crawl and download all of the files, extract them and parse them into the database. Every subsequent crawl of the database should only crawl, download, extract and parse the new files.

I'm open to whatever database you'd like to use although MySQL or MongoDB are preferred.

The deliverables should be the database, an outline of the database schema and the script.

The script should automatically run hourly on a cron job, but should feature a frontend that shows the progress of the crawl and parse job and previous jobs and allows for a forced crawl and parse at any time.
Specific technologies required: python, php, mysql, mongodb
Extra notes: