Thanks so much for for your (relative) newbie perspective of just how important DNS is.
When I think back on how many resources in the past year have been considered more important than DNS I get a very bad feeling in my stomach. DNS is one of the foundational elements that brought me into this community, and like you kenCode I feel very passionate about it.
I don't think the primary obstacle to rolling out a solution is technological, but is rather methodological. The heart of the problem is how to provide open access the the namespace in a manor that best serves all parties. Since some of these parties have conflicting interests the solution is a mater of vetting proposals and building consensus to implement the one chosen. It is largely an issue of what does consensus look like and how is it measured.
In the early days before DNS disappeared from focus there were a few proposals made, but none that seemed overwhelmingly satisfying to me personally. If I recall correctly the auction proposal was the predominate front runner. It is probably the best choice, but my fear is getting a simple name (without conflict with other) could take time to go thru a bidding process and I would need to expose my choice to others potentially taking it away from me to use. If I came up with a catchy, clever name that others find appealing and someone else who had no direct interest in it (other than to resell it), someone could come along and take that away from me.
Some say, well, that's the free market. Others might say, "but I invented that name" and feel it thus belongs to them. The former might say those with the most money will always win, and poor clever wordsmiths never do, is a potential outcome. And then there is the issue of migration & name ownership. Should somebody be able to squat on names like IBM, Microsoft, TrumpEnterprises etc. in the new DNS namespace and devise schemes to hold on to such names (in an auction system) long enough to sell them at huge profits to those who own those names in the existing DNS?
I believe it would be trivial to implement DNS on the blockchain compared to many of the features BitShares has already implemented. I see these issues as the reason DNS lost steam, the community (including the dev team) just couldn't come to a consensus. Where is Brent Allsop (canonizer.com) to help with this specific issue when you need him?