These are some services, cron jobs, automated scripts, etc., I run.
Thanks to Heroku for hosting these services for free.
Wayback Machine is a digital archive of the web. This script is scheduled as a cron job and takes a digital backup of my personal website by parsing its sitemap daily.
This script returns a PNG containing the current date and time on the
/now.png endpoint. This is useful to test HTTP proxies for caching. I'm using it to see how long GitHub caches images in the README for and how frequently it updates those images. Here's a demo:
Result: If you set the right
Cache-Control: no-cache and
eTag headers (the latter Express generates for you), GitHub's camo service will also respect them and not cache your images.
I have an idea: I have a personally hosted instance of GitLab, and I make frequent contributions to it. Those are, unfortunately, not listed in my GitHub contributions, so I lose out of potential green blocks. I'll make a CRON which takes my commits from all repos on GitLab and does the same commits (with empty content) in a GitHub repo. It seems fair to me.
Instead of crontab on Heroku, I'm using IFTTT Date and Time to Webhook as a CRON, because Heroku's free servers don't run jobs when they're idle (which is most of the time). An external request wakes them up.