This is the /first/ L&L where I didn't once find myself shaking my head at something he said. I'm pleasantly surprised.
I disagree with the "1 feat = +1 to ability scores" bit. If not for legacy reasons, I'd get rid of ability scores and switch to modifiers only. +1 to Strength Mod is easier to design around than +1 to Strength Score. Otherwise I found the feat section well thought out.
(As an aside; I've considered doing this in my 4e games, after a fashion. Dropping feats and replacing them with a +1/2 level bonus to damage, like Gamma World. I've instead opted for a "restricted format" where a large chunk of feats are simply disallowed, but it's an option I've had sitting at the back of my head for a while.)
With the skill DC issue, they /really/ need to make that clear in the 'Skill Module', but otherwise I'm okay with it. I like the "Areas/Proficiencies/Benefits" set up; I might co-opt this for a modified skill system in my 4e games.
I like the fighter "traditions". Built right, that could add a lot of customizability without adding a lot of rules overhead.
I'm still not sold on 5e - the math's still too out-of-whack at the moment - but this is the first thing in over a year that's made me feel positive about the system. Kudos.
EDIT: I think I figured out what made me like this article as much as I did. Previous articles left me feeling like they didn't have a handle on the math behind the game, or the real consequences of design choices they're making. The feat section of that article, though, was very clear about the results of various design choices they might make. It was positively /lucid/ . That's what I expect to hear from Mearls. That's one of the main things I liked about 4e - how mechanically transparent and /solid/ it is - and if they can bring that forward to 5e, they'll have a much easier time selling me.