New Feature: Message Rating System

I don't share your cynicism, and I certainly haven't been using the current system that way. For me, a post that is nothing but bare assertions, unsupported denials and flawed logic gets a low score, while one that provides sound argument backed (where appropriate) by evidence with links to relevant sources gets a high score (I almost never use anything other than "5" or "1"). I think that's meaningful. Apart from being too bulky, too complicated, and way too much information, the worst problem with this idea is that voting should be anonymous. There should be a lot more people who are willing to cast a vote than are willing to commit to going on record with a specific response. (And lots of people will vote a particular way for the same reasons, so why have that repeated by every voter?)I'm a regular in a programmers forum where posts are voted on by members (perlmonks.org). The system there is rather intricate, but strangely effective. I'm really not proposing anything like this for EvC -- it's too complicated, and is probably way beyond the scope of what is possible to implement -- but I mention it because some of the ideas are interesting:

• all members get a limited number of votes they may cast on nodes within a 24-hour period
• the number of votes you can cast is low for newcomers, and increases as your "experience/reputation" increases
• a vote is either "up" (good node) or "down" (bad node), but of course, many readers of many nodes cast no vote at all -- it's optional
• every vote on a given node counts toward the "reputation" of that node
• votes by others on a node you post will affect your "experience/reputation" standing, but not all node votes count toward your personal ranking (some random and non-random factors affect the counting)
• just the act of casting votes on other peoples' nodes can increase your "experience/reputation" standing (but again, not every vote you cast will be counted toward your own standing)
• in a normal thread display, you don't see the "reputation" of a node until after you vote on it, but there are ways to find nodes on the basis of "best score" or "worst score", as well as being able to view a list of nodes sorted by their relative rankings.
• in a normal thread display, you don't see the ranking or score of a given member, but this can be found on the member's "home node", and it's possible to get a ranked list of the "best" members.

It certainly isn't "perfect" (no voting system is), and most members acknowledge that the scores are "meaningless" -- nonetheless, people use the system, and it tends to reflect actual quality when viewed in the aggregate.In terms of how this translates into ranking members: some people get high rank simply because they cast a lot of votes over a long attendance with no posts of their own; people who generally write very helpful, informative, sensible and inspiring posts get a strong positive standing (and get to cast more votes, but never an unlimited number); people who routinely behave like trolls tend to get a firm negative standing (which limits their ability to cast votes) -- of course, you generally don't need a scoring system to know who the trolls are...Every now and then you get the "dedicated troll", who views the negative-score standing as a goal to be maximized. It can get ugly, but safeguards (clear criteria for hiding or deleting nodes, such as those already in place at EvC) take care of the worst cases.There is a likely downside of community ratings here at EvC, owing to what appears to be a population imbalance between those who accept evolution and those who reject it, the former being a decisive majority. This seems to reflect the trend in internet communities generally, based on what I've seen elsewhere (but admittedly, my viewing patterns are not a "balanced sample", and I don't have a sense of supportive hit counts at anti-evolution web sites).Anyway, I would expect that anti-evolutionists here would have a hard time building up high reputations by whatever metric is chosen, unless/until they really work harder on making cogent arguments. Does this automatically make any ranking system wrong or unfair?Perhaps people are prone to confuse recognized quality with "mere popularity", but it shouldn't be surprising to find that the two can actually be correlated to an appreciable extent.Edited by Otto Tellick, : No reason given.Edited by Otto Tellick, : added a point to the list to clarify a detail

---

autotelic adj. (of an entity or event) having within itself the purpose of its existence or happening.