Personally if I were going to write the code for such a proposal system I would use a TCR (token curated registry) design pattern. I would say it probably shouldn't cost that much money to write this kind of code as it's not as complex as people make it sound.

In fact I would say because it's so simple there should be competing models and multiple fundraising infrastructure code being written. Blocktrades can write theirs and I can support funding it if it's cheap enough, but I do not think Blocktrades version should get exclusive rights to blockchain funding. BAD IDEA.

Let multiple attempts be made to do it. Let multiple different proposal systems compete for funding from the blockchain if we are going to let the blockchain fund any of them. Let the most efficient proposal system be the most used. I'm not convinced that using dilution/inflation is how to fund it. I understand when SMTs don't exist it seems like a desperation method of funding but it's inferior to using SMTs and just doing ICOs or token offerings or regular old fundraisers.