Main Area and Open Discussion > Living Room

FARK creator doesn't believe in the wisdom of crowds

(1/3) > >>

Lashiec:
It's pretty telling when the founder of one of the pioneers of news aggregation says the model doesn't work at all. Perhaps that's why FARK has an editorial team that promotes links to the frontpage, while the rest remain in the back.


Of course, the statement has to be taken with a grain of salt, as he's talking about comments, which are personal opinions on a matter, not a combined effort of a group of people in order to reach a single answer. Plus, this is FARK we're talking about.

40hz:
I'm curious, has anybody ever done a hard scientific study to determine if there's any truth to the claims being made for the "wisdom of crowds?" Most of what I've heard advanced as 'evidence' supporting this idea amounts to little more than parables, assertions, and anecdotes.

I can understand how it could work for something like an open software development project where "many eyes make for quick debugging." But in this case, the many eyes are programmers (most with formal training) who are focusing on something very specific and well within their area of expertise.

I wonder how well that same crowd would perform if you turned them loose on a problem that was outside their scope of knowledge and experience?

Increasing the number of something applied to something else doesn't automatically accomplish a goal. Very often, simply increasing the amount of input (money, ideas, time, voices, nuclear warheads) exacerbates the original problem and further clouds the issue.  Our government's tendency to go out and "throw big money at the problem" repeatedly demonstrates just how ineffective applying 'more' can be.

And not everything benefits from a group approach. Some issues are far better handled by small, tightly focused and highly trained individuals or small teams. Especially in critical areas where consistency, speed, and accountability are important.

From my perspective, I can't really see how a group one hundred idiots automatically becomes smarter than a group of ten idiots. I'd expect you'd more likely see an increase in erroneous thinking and logical error as the crowd got bigger.

When you consider that all distributions tend towards the mean, increasing the number in the group should have the effect of dragging the overall level of the group towards the middle rather than boosting it on the high end. We've seen this happen in committee meetings where the dumbest members tend to drag the overall performance of the group down to their level far more often than the brighter members succeed in pulling it up to theirs.

I'm not saying there's no truth to the 'wisdom of crowds.' All I'm saying is that there's a big difference between asserting something is true and proving it is.

So I think we need to be careful about too easily accepting something as valid before we find out if it actually is. Especially when dealing with something that "sounds so right" as this notion.

While group wisdom may have the capability of bringing about major social and information breakthroughs, it also has a dark side. Ideas such as ethnic cleansing, master race, apartheid, burn the witch, and thought crime all came out of crowd thinking.

And sometimes, there's very little real difference between a crowd - and a mob.

JavaJones:
I think, like most things, it's a matter of applying it properly. Crowds are apparently (according to research I've read - can't find good links at the moment though) good at predicting outcomes, especially when they are representative of the opinions that will influence the outcome (e.g. movie box office results). There are other things they're good at too. What they're not good at is true *intelligence*, and I don't think any intelligent person is really making that assertion. There are certain problems that can *benefit* from that averaging effect you describe - think about those times you're trying to *find* the average, for example. But when you're talking about intelligently designing something, or coming up with a clever solution to a common or age-old problem, the crowd is much less useful, and really only of value in that it may connect you with that 1 individual who can actually answer.

I think the problem is people see crowd sourcing as a panacea, a fix-all. Or at least some people do. And that's just ridiculous. Even Wikipiedia, while it acts on a massive scale as a "crowd" in overall concept, is actually more like tons of much smaller micro-groups, in some cases highly intelligent and specific subject matter experts. So really there's a spectrum of "crowd sourcing" that needs to be understood to really evaluate efficacy and existing and future system design with that concept properly.

- Oshyan

mouser:
Apparently, neither does the FARR creator.

Deozaan:
I thought it was common knowledge that crowds are stupid and irrational. People do things in crowds they wouldn't do by themselves or with just a couple more people.

Navigation

[0] Message Index

[#] Next page