Step 2: Post a comment on the check-in issue

What you are doing

A GitHub repository holds the files of a project. The repository of your workshop holds the check-in issue, and every data package gets a repository of its own. An issue is a public discussion thread attached to a repository. People open issues to report a problem, propose a change or ask a question. Every issue has a number and a page of its own, and anyone with a GitHub account can comment on it.

A comment on an issue is the smallest contribution GitHub knows. It carries your username and a time stamp, and it stays on the page. You use it three times in this workshop. In step 3 the form for your dataset opens an issue. The review of your data package happens in comments on that issue. And on the day you post one comment on the issue of a published package, on what a human reader still misses. Each change that follows names the person who proposed it, so the record shows who contributed what and what the AI did.

This step is the same action, done once in advance. Afterwards you know that your account works and where to type.

Steps

  1. Sign in to GitHub.
  2. Open the check-in issue.
  3. Scroll to the box at the bottom of the page, under the last comment.
  4. Write your first name, your institution and one line on the data you work with.
  5. Click “Comment”.

Your comment appears in the thread with your username. That is the whole test.

What happens next

GitHub sends you an email whenever someone else comments on the issue. If that is too much, click “Unsubscribe” in the right hand column of the issue page. You still take part on the day.

If something goes wrong

  • The comment box is missing: you are not signed in. Sign in and reload the page.
  • You cannot find the issue: it is in the “Issues” tab of the workshop repository, titled “Check in” and pinned at the top.
  • Anything else: write to lars.schoebitz@ethz.ch before the due date, and we sort it out.

Next

Go on to Step 3: send a dataset.