I mean, on the one hand, it's clear SB DF-9 didn't sell enough to justify continuing to have 4 people working on it full time. So I can't blame them. If Minecraft hadn't taken off, Notch would have dropped it at some point. No one is going to work on things indefinitely while losing money. And it's pretty clear SB DF-9 isn't going to happen. They've tried promotions and discounts and there's not much interest and no one who is playing seems very enthused. It seems like an appropriate project to fold.
But on the other hand, the resulting game is clearly a disappointment to maybe 90% of buyers and there's no way that's what they set out to achieve and it doesn't bode well for their reputation or brand to leave a project in that kind of state. It also suggests that perhaps they should have shouldered additional risk iterating on the game BEFORE consumers got their hands on it. The initial Early Access was clearly too early. I don't think Early Access should be for the first build, I think it should be for something that's really quite playable on its own and just requires further iteration. I've played a lot of great early access stuff and even as a Double Fine fan, I didn't pick this up because it never seemed like it was very far along. And the initial pricepoint was bananas; even if they finished the game it's not clear its scope or production values would have been in the $25-30 bracket rather than the $15-20 type bracket.
Finally, their rollout of this announcements has been really poor. Their "well the game is done, enjoy" statement is transparently false; everyone immediately concluded the game was being abandoned. So the attempt at framing did not work. As a result, would it have hurt them to be honest? By saying that the game's sales and development haven't lived up to expectations, and while they're happy with what they've accomplished, they know it could have been more but they won't be able to take it there. I guess this is made more difficult by the fact that Indie-Fund had bragged about the game's runaway success. I would have also expected them to compensate people for project failure in a way beyond releasing the source; for example, by offering backers codes for other DF games, DF swag, partial refunds, transparency about the budget and sales figures, etc. I have no idea how doable these things are and I recognize that it's difficult to offer a refund after spending the money, but I feel like the rollout of this process has been pretty poor. They disappeared for a while, updated to say that major stuff was coming, and then updated to say the game was done. That to me suggests they either mislead people to begin with, or that the cancellation of the project was sudden for even the team; if it's the latter, hey, it happens, but again honesty would be valuable.
WRT Kickstarter, which has nothing to do with Early Access, the "Broken Age was a failure because it took longer and I don't like the game and also how do you run out of 3 million when you only asked for fifteen bucks and a case of beer? Also I heard that Tim Schafer is a feminist and mumble mumble cheated on her ex and Leigh Alexander who wrote an article about gamers was drunk on Giant Bombcast and Giant Bomb formed from ex-Gamespot refugees and Gamespot gave Twilight Princess an 8.8 and it's clearly at least a 9.1 and in summary FUCK KICKSTARTER AND IOS" angle isn't really worth dignifying with a response. Major, $75k+ Kickstarters are by and large quite successful, most release in a reasonable timeframe, and most of those that are delayed are fairly communicative and show signs of progress. Failures are rare and when they occur tend to be from projects with major warning signs at the beginning, and successes are common and notable.
On a broader level, it seems to me like SBDF-9, Hack and Slash, and Autonomous have not panned out the way they hoped. I'm not sure if it's a case of resource starvation coming from smaller teams or too many projects and not enough management bandwidth, or something about their iterative process that doesn't separate interesting experiments from worthwhile full games at an early enough stage. In contrast to Costume Quest, Stacking, Iron Brigade, and Sesame Street, all of which ended up quite fully formed and workable as games.