Week 9 Dig: Preservation

Conservation — the difference between preserving code and preserving the ability to run it, and the institutions that try to do both.

19 April 2027

Before the dig#

Try to build and run your dig site on your current machine. Document what happens:

1. Can you install the dependencies? Do they still resolve? Are any yanked or missing?
2. Does the build succeed? Does it need a Node version that no longer exists? A Python 2 interpreter?
3. If it runs, does it do what the README says it does?

If it fails, document exactly where. The failure itself is a finding.

Also: check whether your dig site is archived anywhere — Software Heritage, the Internet Archive’s Wayback Machine, GitHub’s Arctic Code Vault. Is the archive a snapshot of the same state you cloned?

In the session#

Share build-and-run results. The class compiles an executability report: what proportion of dig sites can still be built? What is the most common point of failure? npm’s node-gyp and Python 2’s print statement are frequent offenders.

Afterwards#

Write 200 words on what would be lost if your dig site’s repository were deleted tomorrow. Not the code — that is in your clone and maybe an archive. What else? The issues, the pull requests, the discussion threads, the contributor graph, the stars-over-time chart. Which of those are artefacts and which are metadata?