Score voting, the remedy for Bipolarchy recommended by Broken Ladder (see comments on "The American Bipolarchy: Power Without Discipline") is synonymous with "range voting," as described in this sympathetic Wikipedia entry. It works in a way similar to polls designed to determine the most popular movies, record albums, etc. Voters would rate each candidate for an office on a numerical scale. The winner can be the candidate with the best cumulative score, the best average score or the best median score. Political scientists and range-voting advocates have suggested further tabulation methods to minimize the impact of voters giving some candidates exaggeratedly low scores or neglecting to score other candidates. The anti-bipolarchic aspect of score voting is its presumed tendency to encourage voters to rank sincerely preferred candidates (i.e. independents) higher than major-party candidates, while still allowing them to rate one major candidate higher than another. Score-voting advocates insist that some system along these lines is necessary to discourage voters from thinking that votes for independents are "wasted," as they're alleged to be when each person has to give 100% of his vote to one candidate.
At first glance, score voting seems to introduce a questionable element of subjectivity into voting, depending on the scoring range permitted. It'd be minimal if the scoring range is determined by the actual number of candidates (i.e. if you have six candidates and must rank them in order from 1 to 6), but allowing a range from 1-100 would permit people to exaggerate the actual differences in quality or desirability among candidates, though remedies to this tendency, as I said, have been proposed. Any such remedy must take into account the likelihood of conservatives minimizing scores for candidates to the left, and vice versa, or else ideology might distort the process more than it does now. Apart from that caveat, I have no major objection to score voting as a concept. I do question the argument that score or range voting alone can prevent the consolidation of a bipolarchy. There are many possible voting systems in which all it would take to elect independent candidates would be independent thinking and will power on the part of voters. Duverger's Law and other arguments for the necessity of alternate voting systems have an air of capitulation about them. They concede something about the habits of voters that I would rather not take for granted.