Interesting analysis. I think there are a few design factors which drive Steem away from "quality of post" to "quality of network."

1. Short payout period--- a quality informational article can be found useful by many people over the course of years. A week captures little of that long tail of potential value. Great for blogging, not so good for encyclopedias.

2. No way to capture the value of non-network participants. Anybody can read Steemit.com for free. But a blog which attracts 10,000 outside eyeballs will be rewarded only based on the number who decide to sign up (and they probably won't.) Ads are one way to monetize those outside viewers, but Steemit itself probably won't run them and individuals will have difficulty attaching them one discussion at a time.

3. Poor discovery mechanisms. Some curation is done in an ad-hoc fashion by individuals, but tags are frankly awful. Finding content of interest to you requires a lot of upfront work to figure out what communities exist.