What is web scraping?
A way to extract information from websites
Acquire non-tabular or poorly structured data from a site and convert it to a structured format (.csv, spreadsheet)
Source: https://xkcd.com/974/
Automate web data retrieval
- identify sites to visit
- define information to look for
- set how far to go
- set scraping frequency
Scraping vs crawling
Crawling. What Google does to index the web - systematically "crawling" through all content on specified sites.
Scraping. More targeted than crawling - identifies and extracts specific content from pages it accesses.
Some sites disallow web scraping with a robots.txt file.
Is scraping the best option?
- is it easy to copy/paste?
- does the site provide an export option?
- is there an API?
Ethics and considerations
Am I allowed to take this data?
Check the website for terms of use that affect web scraping.
Are there restrictions on what I can do with this data?
Making a local copy of publicly available data is usually OK, but there may be limits on use and redistribution (check for copyright statements).
Am I overloading the website's servers?
Scraping practice should respect the website's access rules, often encoded in robots.txt files.
When in doubt ask a librarian or contact UBC's Copyright Office
https://copyright.ubc.ca/support/contact-us/