# 2.0

## Unwisdom of Crowds

In Uncategorized on July 7, 2006 at 8:15 am

Author: Marc Fawzi

~~

A Crowd Has No Wisdom

Before we make this argument, let’s define the types of crowds.

{The designations of ‘condensed’ and ‘dispersed’ given below for crowds are relative to the ability of the members of the crowd to communicate with each other and affect each other’s judgment.

The word “crowd” is used here to mean a large group of people, not 5 or 10 people but thousands or millions of people.}

A dispersed crowd (without a formal hierarchy) will produce averaged judgment. For example, asking each of 200 people (not at the same time or place) how many jelly beans are in a jar would result in an averaged judgment, which would eliminate values that are too high or two low, resulting in an estimate of the number of jelly beans in the jar (which is a measurable value) that is close to the actual value. In this case the crowd is nothing more than a decent statistical calculator. It has not exhibited any more wisdom than the tool it is being used as.

A condensed crowd (without a formal hierarchy) may produce averaged or lowest-common-denominator judgment, depending on whether or not its judgment is rationally or psychologically driven. In case the judgment is about a measurable value it would most likely be rationally driven, and, thus, be an averaged judgment. In case the judgment is about a quality it would most likely be psychologically driven, and thus, be a lowest-common-denominator judgment. In the rational case, the assumption is that, even though the crowd’s members can communicate with and affect each others judgment, if each member is rational enough and the judgment to be made concerns a measurable value then the crowd will likely produce an averaged judgment (i.e. the average of independent judgments.) If, however, the crowd members can communicate and affect each others judgment and the judgment to be made is qualitative not quantitative then the crowd’s judgment should tend toward the lowest common denominator.

A typical crowd is a mix of both the dispersed and condensed crowds. Thus, its range of judgment with respect to both measurable value and quality include both averaged as well as lowest-common-denominator judgments.

The problem with averaged judgment when it’s applied to quality (rather than measurable value), which can happen in a typical crowd, is that you end up with a judgment of average quality, not the best judgment.

The problem with lowest-common-denominator judgment when it’s applied to quality is that it uses the primitive part of our psychology. In other words, expect exactly the opposite of wisdom.

So when it comes to quality, a typical crowd is going to be either a judge of average quality or an unwise judge. And nothing else.

Where does that leave the ‘Wisdom of Crowds’ movement? (in the garbage bin of history in my candid opinion.)

### Toward a Democratic Society

A hierarchy that doesn’t listen to the crowd (or that forces and manipulates the crowd to listen to it) is a dictatorship (e.g. North Korea, Iran, the 3rd Reich, etc.)

However, mixed ‘hierarchical + crowd’ system, which ideally allows the crowd to adjusts the judgment (of the system), is a democracy.

Therefore, Web 2.0’s [un]wisdom-of-crowds model needs to be fixed by adding the concept of a non-arbirary hierarchy that is by the crowd (or people) and for the crowd (or people.)

Below is one example, using ‘digg’ as the Web 2.0 application, that shows a prototypical transformation from Web 2.0 to Web 2.5 (or from “hunter gatherer” to “democratic society.”)

Electing Leaders in a Democracy: Building the System

In an application like digg (or the “digg killer” to be exact) writers, content producers, social figures, business figures, and others, who are higher in the food chain than the consumer, and who are collectively referred to herein as ‘taste makers’, should be allowed to start their own channel (or page) where they list links they think are cool. If enough people ‘bookmark’ a given page then that means that the taste-maker in question is worthy of being positioned into the system’s hierarchy at a higher level than that of the consumer. The taste-makers can then rally their followers (those who use them as taste-makers) to digg the links the taste maker has chosen to put on his/her page.

This is similar to parliamentary democracy where members of the parliament have to get enough votes on a given issue from their district in order to pass it into law.

The key here is that the ‘trusted’ taste-makers get to decide which links to promote for votes from their followers.

At the same time, people in the crowd should be able to vote the taste-makers in or out of the system’s hierarchical structure by bookmarking or un-bookmarking their page.

Anyone who has followers can become a taste-maker, but they would have to replace an existing taste-maker as the system has a finite hierarchy with finite number of taste-maker positions (e.g. in the thousands.) And once someone is elected as a taste-maker they would stay in the role for a certain period before they can be voted in or out of the position by their followers (assuming another contender has nominated himself/herself for the position.)

This is a very simple ‘hierarchical + crowd’ system that implements a very simple form of leader-follower democratic process.

The perils of letting the crowd decide without giving them a democratic structure and process is to let lowest-common-denominator and averaged judgments become the norm.

Leaders and Crowds need to work together within a democratic structure and process to assure the best judgment possible.

BTW, this is not much different than the process whereby the crowd selects its taste-makers (e.g. Radio DJs, Wise men, etc.) except this provides a structure to formalize the process, which would be too costly and time-consuming in the real world. So may be this would also apply to how society elects its taste makers (outside of social bookmarking sites.)

The reason this system would kill digg is because it will have an aggregate quality of judgment so much better than digg.

Related

1. Web 2.0: Back to The Hunter Gatherer Society
2. The Future of Governance (outdated by next link)
3. Toward a Natural World Order

Tags:

1. There is a certain flux in the blogosphere, it is searching itself, adapt itself to the new models. Of the millions of blogs out there, a fraction of them are interesting. There is a lot of spam and junk out there. Hierarchical models are a must.

2. As usual, I’m a bit more hopeful than you, Marc, though I do entirely take your point.

The blogosphere is a long way from perfect, but if you look at the top ten blogs, you’ll find a million times more intelligence, rebellion, diversity and wit than you’ll find on the average newsstand, airport book shop, or indeed the top ten regular internet sites.

I think news-voting sites are just a little newer and haven’t yet learned the lessons they need to become more intelligent, whether that’s creating a hierarchy of voting influence, being more sceptical, making it harder to pass judgement, or whatever mechanism become the norm.

3. The picture I present is exactly how it is but only focusing on the negative side… That’s because the positive side does not need any fixing.

I’ll probably be in London next Spring so we’ll get to chat about it then and see if putting these ideas out will have had any positive effect on the Web 2.0 community. Basically, I think there should be amove towards Web 2.5.

I’ve said enough on the Web 2.0 and Wisdom of Crowds stuff. :)

Marc

4. I was pondering a question approaching the last leader + follower thinking just yesterday when I was perusing the “top diggers” section of digg, and was struck by the thought “would / should / can individuals in these crowd-based systems gain recognition (and following that, influence) by nature of their large involvement?” This seemed to be really more a question for digg than anywhere else, as you can quickly determine that someone who prodigiously contributes to a community somewhere may be more reputable than some suspicious new player… but I then considered that there is no indicator of “power pushers” that I MYSELF felt a victim of, I had no compulsion to check every story and see the % front page of the poster, for example.

different clearly what the elections you are talking about because there the hypothetical taste maker would be put in a position of prominence not because of their tastes but simply because of their enormous appetite (more numbers, higher rank–everything else second)?

Anyways I know I just failed to articulate myself intelligibly, but that’s not the point! I can’t help but shake my impression that the next step up (or down, as the case may be) from this crowd-methodology will be towards smaller, slightly smarter crowds within the mob. A sort of specialization. Be it digg cliques or blogs cross-linking, the same community involvement just doesn’t seem to stay the same as the system scales up. Breaking into smaller pieces to increase surface area, as it were, seems logical.

5. For something as fundamental as governance, while it’s tempting to try different and new ways, the end result is that we have tried for tens of thousands of years every conceivable way to govern society and we have ended up going with what was invented by the Greeks (i.e. democracy: hierarchy + crowd) as the ultimate form of governance.

Now, anything that can build on top of what is proven should be explored but ideas like going back to collectivism (aka ‘wisdom of crowds’) are not going to work: we tried them. They didn’t work. We moved on.

Something as fundamental as the process of governance is not immune to innovation but innovation in the core process happens over tens of thousands of years not in a handful of years.

The e-Society concept I explored in a more recent post can sit on top of a democratic structure without supplanting it. I believe it is much safer to give the e-Society concept a try than to embrace the ‘wisdom of crowds’ which has never been proven to work, not in del.icio.us, digg or CCCP.

Marc

6. This comment is a follow up to a discussion I’ve had with Sam Rose on this subject.

On Truth, Belief and the Future of Governance

What is Truth?

I believe that the conscious act of planning, thinking and experimenting is the only truth there is, and that the particular thoughts, plans, processes and results that we generate in the process are transient artifacts.

What is Belief?

I believe in beliefs, i.e., I believe in making basic assumptions.

Future of Governance

My basic assumption is that the process of governing human societies in cyberspace will ultimately go back to the classical model we have today in the Western world. It may take 10, 20 or 50 years of experimenting with but I believe we will come full circle to what we have today.

I believe that the core governance process that is our democratic process (which is in essence the same basic idea as that invented by the Greeks, with several important innovations built on top of it) is immune to innovation in the short range. This belief applies to our core governance process now or at any time, i.e. it will always be immune to innovation in the short range. Change in such a process that is fundamental to our existence and progress tend to happen every so many thousand years, but it cannot be forced.
I don’t believe that we will succeed in changing the core process that is the current process we have today. I believe that we can innovate on top of it.

That does not mean we shouldn’t experiment with ideas.

So let’s think, plan and experiment and let’s do that differently (on purpose) because the particular thoughts, plans and experiments are nothing. The act of thinking, planning and experimenting is everything.

Marc

7. […] The Unwisdom of Crowds does web 2.0 mark a return to a hunter-gatherer culture? (tags: essays web2.0) […]

8. It’s a good argument as far as it goes, but you don’t explain why
an averaged judgement is worse. I think you also didn’t define
quality either. I think the operative question is ‘why is the
average judgement worse?’ and ‘Worse for what?’

9. I can’t define quality on your behalf or on behalf of other readers

Quality is a subjective thing so it cannot be defined in objective terms.

Your second question “why is the average judgment worse?” is interesting but I did not say it’s worse, although it may be implied in some way (since I see my own personal judgment as likely to be above average, i.e. most of the time.)

The average judgment is worse only if your judgement is above average.

However, if someone’s judgment is average or below average then it’s better to go with the average judgment.

I believe this is a good thing to clarify explicitly, although it did state in the Pros section that averaged judgment has the advantage of averting bad judgments (i.e. below-average judgments)

:)

Marc

10. maybe at the beginning we get back to hunter-gatherer mode, but perhaps it is necessary to learn new technology, saying differently, exploring the new plain Web20 fully before we proceed. In the meantime the ideas of taste-makers could develope, and then after we shall see the effects.
I don’t agree with the idea of basic insticts of crowds, because sooner or later, everybody finds more usefull sites. Besides those who create sites are really increasing quality.

11. Hang on a minute. AFAIK, in Greek city states people didn’t elect leaders. All men could vote themselves, no?

But I agree that a nice mix of direct and representative democracy, i.e. “liquid” or “delegative” democracy could yield better results. But I lean way more towards the direct side of things that you appear to above.

I’d recommend at read of Democracy 2.1 and Delegative Democracy pdf available here: http://open.coop/background+docs

For liquid democracy see http://wiki.uniteddiversity.com/liquiddemocracy and http://campaigns.wikia.com/wiki/Liquid_Democracy

Also, have you seen http://smartocracy.net/ ? I think you may like it.

• Hi Josef,

The concept of voting as a governance model is what I was referring to by saying ancient Greeks had invented democracy.

I had written the article some 3 years ago and my thinking has evolved, even though the thinking expressed in the article still holds in some sense.

Right now I’m focused on the difference between the outcome of running he Snowdrift Game model and the Prisoner’s Dilemma model in hierarchies vs. meshes.

Recent studies show that unpaid cooperation (e.g. Snowdrift Game) backfires for the cooperators in hierarchies but works really well in case of meshes. The same studies also show that the Prisoner’s Dilemma, which is more of a classical survival game, works best in hierarchies. I will update the P2P Energy Economy once I have wrapped my mind around these conclusions.

Marc