A basic analysis of comments from Slow Boring between January and April of 2022.
Some high-level findings
- 212 articles were posted in this time period of four months
- 29,900 comments were posted on these articles
- 2,040 unique users posted comments
- 2 comments were posted by the median commenter
- 1,508 comments were made by the most prolific user
- 56 users each posted at least 100 comments
- 12.4% of users accounted for 80% of all comments
- 2.35% of users accounted for 50% of all comments
- the Gini coefficient of comments per a user is 0.819
The descriptive stats for comments per a user are:
| stat | mean | min | 25% | 50% | 75% | 90% | 95% | 99% | max |
|---|---|---|---|---|---|---|---|---|---|
| comments / user | 14.7 | 1 | 1 | 2 | 7 | 23 | 55 | 243.7 | 1508 |
In terms of likes (i.e., ❤'s)
- 95,423 likes were given in total
- 229 likes were given to the single most-liked comment
- 4,545 likes were awarded to the most liked user summed across all of their comments
- the median commentator received 5 total likes
- the Gini coefficient of total likes per a user is 0.774
The descriptive stats for total likes per a user are:
| stat | mean | min | 25% | 50% | 75% | 90% | 95% | 99% | max |
|---|---|---|---|---|---|---|---|---|---|
| likes / user | 46.8 | 0 | 1 | 5 | 23 | 78 | 183 | 841 | 4545 |
The details of the analysis are in the Jupyter notebook analysis.ipynb.
Comments are fetched using substack_client.