Man, I really like this comment.
And I realize that sentence alone is pretty pointless, so I'll expand on it a little:
One of the huge sources of cognitive dissonance I have in HN and startups in general (and this is coming from someone who is explicitly not a startup/SV person, though I find it a great spectator sport and a wonderful font of knowledge) is the clash of lean against polish. I felt like I've spent the majority of my life indoctrinated to believe that polish above all else is valuable: to spend hour after hour poring over iteration after iteration, to settle with nothing except perfection. This manifests in a few ways: Pixar, for example, calls it something along the lines of 'dusting under the cabinets': making sure each aspect of each scene is absolutely pristine.
And then there's sort of the 'great is the enemy of good' / 'fail fast' / etc line of thinking: get something out the door, optimize for value over time, focus on efficacy. Which has seemingly become trendier than ever -- partly because it clearly works, and because of the undeniable numbers that people rarely buy goods and services because of, say, the fonts -- but an annoying little part of me cringes when I see a new service that doesn't even bother to customize the default Bootstrap theme.
Most times, Ii's hard for me to reconcile these two seemingly distinct philosophies because they seem at so odds with each other. I'm beginning to suspect that the truth lies somewhere in the middle: that there's a time and a place for prioritizing growth and validation above everything, and that there's a time and a place for prioritizing perfection over everything.
Threes succeeded because of a relentless pursuit of quality -- and because of an addicting core concept that arose from that pursuit.
2048 succeeded because it of its price and accessibility, because it was easily hackable, and because it could spread like wildfire. (But then, it also succeeded because of Threes.)