Ep. 1 Introduction to Twitter

Ep. 2 Getting familiar with JupyterLab

We harvest the bjules timeline in order to test the installation
We do first and last with bjules and hashtag gas.
Generally, this is too long. This isn’t a Jupyter lesson (or could it be?)
After configuring twarc in jupyterlab, helpers can help struggling learners on the twarc2 configuration while the instructor may go over the remaining episode content.
We’ve written the lesson using @ecodatascience as the solution to the ‘2 Timelines’ so if you do that as an instructor, subsequent code should run as is. We plan to replace @ecodatascience with @bergisjules in subsequent episodes in order to simplify things.

The key point is to show a little bit of json and talk people into immediately converting to csv and dataframe
The difference between timelines and search/filter datasets needs to be made explicit here.

@bjules vs #catsofinstagram. What’s the difference difference between a timeline and a sample/search capture?
@ucsblibrary is going to be edited out
The final challenge in episode 4 is to live harvest a dataset using a ‘cute’ hashtag. You may wish to pre-harvest a dataset. Beware of improvising: you may get results that you don’t want on your screen.

This Jupyter Notebook has most lines that use quota or take a long time to run
are commented out.
This notebook will run the workshop from start to finish AND consume quota, assuming you have downloaded the appropriate data from the setup page.
Halfway notebook is a natural starting point for day two if you split this lesson in half. It runs and leaves you with three dataframes in memory.