HN user
PDF is an open format. Scanning PDF for the links is probably the least you could assume.
You gotta assume they will reconstruct generations of your family tree if they can from your chat history.
TensorFlow & PyTorch, Angular & React,
Or even Karma & Jest,
Facebook seems to be late to market, but learns from Google's mistakes, to create simpler and more elegant tools.
Clickable links:
https://promesante.github.io/2019/08/14/clojure_graphql_full...
My coworker (who uses Windows 95) was having trouble downloading a self-extracting archive off the net. In an attempt to make it easier to open the file with WinZIP, he associated *.EXE with WinZIP.
Nothing worked after that. Every program he tried to run would load WinZIP first. He couldn't even run REGEDIT to delete the association.
Happened to me.. :) One of those Windows 95 days!
If built correctly it's genuinely difficult to mess things up.
Indian Space Research Organization (ISRO) [0], Indian Institue of Management - Ahmedabad [1], and (the not so known) Physical Research Laboratory [3], premier institutions in India, all of them founded by the same person - Vikram Sarabhai. Each of them stood the test of time.
Given the situation of all other government run institutions in India, ISRO is truly a marvel. Maybe "it's genuinely difficult to mess things up"!
[0] https://en.wikipedia.org/wiki/Indian_Space_Research_Organisa...
[1] https://en.wikipedia.org/wiki/Indian_Institute_of_Management...
[2] https://en.wikipedia.org/wiki/Physical_Research_Laboratory
Firefox uses PDF.js[1] to render pdf. This is a standalone project to render pdf in the browser using html and js. If I remember correctly, you can even render pdf in chrome using this.
A non-techy person who has heard of emacs!
Essentially with all that data, YouTube still hold a bad recommendation engine.