Good highlights in the screenshot regarding the benefits of code-as-text we take for granted: https://twitter.com/gordonbrander/status/1191947925871628289?s=21

I find the bit that’s not highlighted in between even more important. That somebody/humanity spend a lot of time/effort figuring something out isn’t really a good quality of that thing. He describes some qualities that make text interesting though, and that’s useful — they help us understand why text is useful. Unfortunately, arguments are only entertaining, if both sides represent oversimplified extremes. And so we end up with nonsense either-or discussions all the time. The really interesting solutions are somewhere in between the extremes, using the puristic arguments from the extremes to distill which of the properties need to be combined for a truly better solution. Text is too important to leave behind completely. Arranging glyphs in a grid with interactions invented before screens existed is not.

i think there’s a bit more to this that’s remained unstated by all parties: the graphical relations of text are semi-unstated, fairly flexible, and extremely modular. you can pretty easily initiate a pattern like repeated capitalization (eg camelCase) and get meaningful results out of text without breaking its understandability