This project uses intellectual property information (copyright and patents) to identify the universe of software development.

Copyright is a form of protection provided by U.S. law to authors of "original works of authorship" from the time the works are created in a fixed form.

• Copyright Basics (Circ. 1)

U.S. Copyright Office. n.d. “Copyright Public Records Catalog Online. 1978 to Present.” US Library of Congress. Public Catalog. https://cocatalog.loc.gov/.

• Relevant Records

  • Type: Computer Files
  • Date: Last 20 years

• Access

  • Public Domain
  • Access through government portal
  • Available for harvest
  • No bulk download function
  • Relevant documents

• Data harvest will be performed using a script in the server during least disruptive hours and the data will be saved to the database.

To understand the OSS universe, traditional methods is not enough because open source developers are not filing copyright for their code. GitHub the best potential source as all OSS codes are attached with OSI-approved licenses.
We aim to collect the following information from GitHub repositories:

• Names of repositories that have OSI-approved licenses
• owners of repositories
• contributors of repositories, including their locations and organizations
• contribution to each repository, including additions, deletions and commits

• Rest API
  Easy to access but provides no way to directly query for particular licenses.
• GraphQL
  Offers direct query, but has page limitation and API calls timeout issues.
• Web Scraping
  Can be combined with GraphQL to obtain information that is difficult to query.
• BigQuery
  A public dataset provided through Google which is infrequently updated and thus not a reliable data source for our purpose.
• GHTorrent
  An open source project updated multiple times a day that reflects the data on Github, but did not offer information on repository licenses or additions and deletions.

• poster_graphs
  This folder contains scripts that are used to produce graphs for the 2019 poster. The "poster_graphs.md" file is a readable document.

• additional_scripts
  This file contains some products that are not included in the 2019 work product.