>>57
You havn't seen or visualized a good weighting algorithm then. It would BBCodify text in a manner similar to a human poster, with pleasing patterns visible i.e, nesting subs to depth 5 and then unnesting while alternating bold and italicized words. I would try, but for the algorithm to be good enough to meet my expectations it would be good enough that I could write a research paper on it, and that would just be a large waste of time when I could design better useful algorithms I will actually get paid to submit to journal instead.