His observation that the rules should fit on 1 page is interesting, though I am not sure I agree.
I think that's more about having a rules summary/cheat sheet that fits on one page and not all rules. From the site: "It should be possible to summarise the key rules
on one side of one sheet of A4. I don't want to have to turn my fast play sheet over. "
I also disagree with his statement on abstraction, though I do agree that D&D models combat completely wrong (though hit points management can be made into a fun game).
I think it makes sense in context. Armour class works by providing protection against hits but not against damage in D&D and that's kinda odd when you actually have hit and damage in two different rolls and it feels like heavy armour affects the wrong one. His point is about making good abstractions and not arbitrary ones. If you combine that with the aim of finding good rules that give you interesting choices that matter (instead of just more or very detailed ones) you should end up with a fun game even if it's not realistic and some rules break the rule of "wrong abstraction". And I think he's talking about D&D-like level in the context of a wargame (and not in D&D itself) although a simple levelling system is fun in campaigns. Maybe he's against levels in general and wants a skill based system that's more fluid and not as stratified as a levelling system is where you get better in big chunks every level but have nothing in between?
The bit about D&D hit also depends on which D&D you are talking about because sometimes you die at zero, then there were rules where you are just unconscious at zero but lose 1HP but really die at -10. I think I remember three variations of how HP
work in 2nd edition (and with extra rules for bleeding) that were made to fit into certain expansion books (like the one books that split your attributes into two: If I remember correctly "Strength" became "Muscle" and "Stamina", or something like that).
What are some core concepts that you feel are essential to any design? What are some that vary by project?
I think overall all those points are quite good but I would also say that realism/authenticity is not that important if you have a fun game (despite what I wrote above) because his ideas seem to be more about historical games. It needs to be fun and your decisions need to have an impact. There's not much to add as the post seems to already cover quite a lot.