If you could manage to build a reddit-style community behind a pay-wall, the pay-wall could be used to generate the 'finder's fee' money for posts and you could allot every customer x number of 'coins' that they could upvote with--needs to be tied to what they pay to prevent scamming. The end product would need to be vastly superior to HN to have a hope at attracting the initial community, but that's basically what Bloomberg machines do for finance, right? I'd easily pay $100/mo if not more for the best hacker news and commentary on the web (you could have the same pay-to-post, get-paid-for-upvotes model for comments to keep them quality).
The pay wall was exactly what I was thinking. It would get users used to paying for upvotes, and then when that money runs out, their credit card is already linked, and makes it easier to fill up.
The main goal is to create a market which supports high quality/high cost content.
That's all above the question of 'how much will it improve the site?' and 'will it prevent people from giving good, interesting links?'
There's what, a few hundred new links per day in 'newest'? That's ~$200 minus credit card fees. Is that enough to manage the headache of dealing with the issues surrounding money?
My answer to these is "no." It's not worthwhile. What are the problems you're trying to solve?
the idea behind charging to post/upvote is to generate enough of a return to warrant higher value content creation.
This would get in the way of innovation.