Bumblebee Labs

After this, he gave a much longer, more detailed talk about social software and small group theory, "A Group is its Own Worst Enemy": http://www.shirky.com/writings/group_enemy.html

The talk is an accurate and accessible adaptation of 50 years of sociological research by Bion and others in "small group research" and "group dynamics."

If this sort of thing interests you, you might also like social interaction designer Xianhang Zhang's compelling anecdotes: http://blog.bumblebeelabs.com/social-software-sundays-2-the-evaporative-cooling-effect/

And my own essay on the Austin design community, with hypotheses about how communities work at the end: http://distance.cc/issues/01/01c-Vitorio-Miliano.html

There are about 150 footnotes in the essay, but I pulled out all the references into a separate PDF, and linked to all of the papers and talks that are available online or out-of-copyright, which would be a good starter kit for anyone looking to do social software correctly: http://distance.cc/issues/01/vitorio-colophon.pdf

( I follow up that essay with another one, talking about professionalism in design and the future of design practices, and Distance is on clearance now for $5 if you want to read it: http://distance.cc/ )

After I wrote my essay, a couple scholars produced a book called "Successful Online Communities: Evidence-Based Social Design." I haven't read it cover-to-cover, but the parts I did read were very good, lots of well-researched advice for new social…

First off, for anyone looking at web reputation systems, read the book on the subject: Randy Farmer and Bryce Glass, Building Web Reputation Systems:

Book: http://shop.oreilly.com/product/9780596159801.do

Wiki: http://buildingreputation.com/doku.php

Blog: http://buildingreputation.com/

I can pretty much guarantee there are elements of this you're not considering which are addressed there (though there are also elements which Farmer and Glass don't hit either). But it's an excellent foundation.

Second: If you're going to have a quality classification system, you need to determine what you are ranking for. As the Cheshire Cat said, if you don't know where you're going, it doesn't much matter how you get there. Rating for popularity, sales revenue maximization, quality or truth, optimal experience, ideological purity, etc., are all different.

Beyond that I've compiled some thoughts of my own from 20+ years of using (and occasionally building) reputation systems myself:

"Content rating, moderation, and ranking systems: some non-brief thoughts"
http://redd.it/28jfk4

⚫ Long version: Moderation, Quality Assessment, & Reporting are Hard

⚫ Simple vote counts or sums are largely meaningless.

⚫ Indicating levels of agreement / disagreement can be useful.

⚫ Likert scale moderation can be useful.

⚫ There's a single-metric rating that combines many of these fairly well -- yes, Evan Miller's lower-bound Wilcox score.

⚫ Rating for "popularity" vs. "truth" is very, very…

The overall readership size isn't of great concern (the more the merrier), but it breaks down when people start interacting at scale. It breaks down when posts start getting thousands of upvotes or tens of thousands of comments. At that point, why bother?

Depending on where you step in Reddit, it's either steeped in a common shock culture (relate everything back to racism, sexism, nazis, 4chan dirty memes, ...) or just friendly nonsense. You never know what will show up where.

All this has happened before, all this will happen again. http://shirky.com/writings/group_enemy.html — http://blog.bumblebeelabs.com/social-software-sundays-2-the-evaporative-cooling-effect/

You have to fit your thinking to fit the social masses because you can't manipulate the masses enough to fit your clever business model (unless you can demand monopolistic lock-in to force behavior).

A slightly better article on the same topic from a few years ago: http://blog.bumblebeelabs.com/social-software-sundays-2-the-evaporative-cooling-effect/

It's funny we couch it in such intellectual language. There's an everyday word for this: Potemkin village.

http://en.wikipedia.org/wiki/Potemkin_village

Or, a thought experiment. What's more valuable? A social network with 10,000 people who are all at most two degrees of separation from everyone else in real life? Or a social network filled with 10,000 people from around the world, selected at random?

Could you even call the second thing a "social network?"

Oh, please. That comment is in reply to one at the very, very bottom. (As I'm viewing it.)

Cherry-picking a single--frankly, not really that rude or terrible--comment isn't proving anything.

In either case, the "good old days" never existed. See: http://blog.bumblebeelabs.com/social-software-sundays-2-the-evaporative-cooling-effect/

I get it, though. It feels good to be self-righteous. So let's go nuts, let's have this little drama in this thread and give PG and the mods another headache they don't want or need. As the owner of a social network for over five years, I feel for them.

You may be interested in "the $5 Guerrilla User Test", which was posted on HN last week:

http://news.ycombinator.com/item?id=1120820
and
http://blog.bumblebeelabs.com/the-5-guerrilla-user-test/

the basic idea is to go into a bar and find someone who's a little bit drunk who'll spend 15 minutes using your software in exchange for a drink. "Watch your application crash & burn as people do all sorts of ridiculous ass shit they would never do in a lab but constantly do in real life", and learn from it.