Would DAO/SPS proposals based on an "Advance" funding model be supported?

One of the challenges of the DAO/SPS proposal model is that proposal voters need to have a high degree of confidence that the project is worthwhile before voting yes on it. That is doubly complicated because the voter is evaluating both: 1) Would the project be valuable if it was done, and 2) is the proposer going to deliver the project. Plus, with the current distribution of Steem Power, some large voters will be needed to get any proposal over the "return proposal" level, but the large accounts may be reluctant to support proposals out of a desire to not be seen as "playing favorites" by picking and choosing who to support.

I was thinking about this and wondering if there might be an alternative model that might be more palatable: the "advance" model. In conventional book publishing publishers often give authors an "advance against royalties", so the author technically gets some small amount for each copy of the book sold, but is given a large payment up front and only starts getting paid the per-book rate after enough books have been sold that it equals the amount of the advance they've already been given. Because book publishing is a hit-based industry only very successful books tend to earn out their advance and get to the stage where per-book payments matter, but being able to get a nice check from the publisher helps the writer to be able to write for a living rather than relying on the fickle tastes of the book-buying public to be able to put food on their tables. Functionally it's kind of like transferring some money from the possible-future-world where the author is wildly successful to the version of themselves who needs money in the present.

What if there were DAO/SPS proposals that worked similarly: the creator has a proposal that gives them a steady income in SBDs, with the expectation that some fraction of SBDs that they earn via their activity on the chain (maybe even 100%) get paid back to the DAO until they've met the amount they got. Since at least some of the SBDs would be put back in the DAO there's less need to judge whether the project itself is valuable enough to be paid for with the DAO's funds (the SBDs that are paid back offset how much is being paid out), and it would be easier to police whether the proposal creator is doing what they said rather than needing to wait for some big deliverable (for example, if they get some big rewards on a post but don't send them to the DAO account that's a signal that they're not living up to their commitments and should have support withdrawn).

What do people think? Is this a good idea? Would the community support proposals structured like this? How about the big accounts that are currently a big factor in whether a proposal can get funding?