It happened to me one in the desert, at night. I have blindly followed Google Maps, and I ended up without a signal, in a middle of no where. Luckily, I was not very far from the city.
Just to add to this comment. If users can't watch/read, you can listen to podcasts while you run or drive a car, it can help you to understand how people try to build a business. There was a podcast (How to Start a Startup | Podcast on Spotify)
. On that note, I enjoyed also listening to those who are building projects, and share their journeys e.g. Nathan's interviews with SaaS founders (https://nathanlatka.com/podcast-thetop/) and The Startup podcast(https://gimletmedia.com/shows/startup)
What are the main advantages compared to Airflow. We use Airflow to orchestrate ML jobs/tasks, and I found it to be more flexible comparing other tools we tested.
So the whole idea here is to analyze the data dump from fb. I have checked the website, but it seems that you put efforts to reverse-engineer the logic in the dump and how it is stored.
A follow up question. Other than data dump, is there any other sources you use?
I always thought that recommendation engines should be customized and adjusted by the user. It seems that social platforms are forcing what they think is relevant on me, and I have stopped using some social platforms for this reason. In some cases, I wrote scripts to fetch content (e.g. tweets, rss feeds) and apply my own methods to track topics, or bust the bubble. When the timeline started to show irrelevant topics to my interest (that changed over time), or non-eye opener topics, I usually start leaving the platform as a user.