Thanks to all of you for participating in this Dialog. Let me begin with the introductions. We're lucky enough to have gathered together here Jeffrey Veen, Executive Interface Director for Wired Digital; Gurl.com community director Heather McDonald; Firefly interface designer Eric Liftin from Mesh Architectures; and Andrew Shapiro, author of The Control Revolution.

As far as the content goes, our topic is both practical and philosophical: How do we design more enriching communities? What kind of communities do we want online anyway? Let's start with the front page of the New York Times. Yet another study has appeared that argues that heavy users of the web are a disconnected, anti-social bunch.

You can read the whole thing here [http://www.stanford.edu/group/siqss/], but let's jump right to the fighting words:

"The Internet could be the ultimate isolating technology that further reduces our participation in communities even more than did automobiles and television before it."

I'm assuming that we all agree that this is by no means a definite outcome. But what should we be doing as designers and critics to ensure that this doesn't happen?

-- SJ

Date: 2/22 2000 23:27
From: Jeffrey Veen

I've always feared the word "community" when applied to whatever Web project I happen to be working on. There really isn't anything wrong with online communities -- and I'm sure we'll discuss that in greater detail soon -- but the term has become a magic bullet for business plans and the product managers that implement them. Paired with "monetizing eyeballs," "viral marketing," and "opt-out signups," it becomes yet another shaky beam in an already crumbling foundation.

Erasing time and geography is wonderful. I feel well connected to the sea kayakers, risotto cookers, and human-computer interface researchers with which I interact on a daily basis. But community is work. Hard work. Ask any majordomo list owner how much time she spends a day admonishing rambunctious users, weeding out bad address, and taking care of the other mundane chores that go into maintaining a strong online community. Add to that the monumental task of keeping the discussion on topic and you can start to see why so many groups have failed in our new medium.

So it's no wonder I cringe when I hear someone going on and on about how their new Web site will most certainly have chat rooms, message boards, instant messaging, and whatever else is attracting venture capital these days. "Think eBay meets Slashdot!" Oye...

I'm pretty interested in the sites that have managed to incorporate community into their basic fabric. Epinions.com is a good example of how strong architecture can get people talking (and arguing!) about the products and services they love and hate. The notion of a "web of trust," in which your rating of other reviewers affects your view of the site, is an interesting example of how we can encode into software the way humans interact in the real world. I'm sure there are others...

-- JV

Date: 2/23 Feb 2000 23:06
From: Eric Liftin

This Dialog is likely to spur debates on two related topics: first, is the Internet bad? ("the ultimate isolating technology"); and second, if it's not inherently bad, how can we foster successful online communities?

Our moderator cites the recent Stanford study claiming, provisionally, that the Net displaces social experience. While I always appreciate a good bar-graph laden study, this one seems to offer the same facile critiques that people who have never even seen a Web site often recite. For every complaint that the Net is anti-social, there is a complaint that the Net is anti-meditative. Stanford Professor Norman Nie says "E-mail is a way to stay in touch, but you can't share a coffee or a beer with somebody on E-mail or give them a hug." Last week an acquaintance couldn't imagine that reading the screen could ever equal curling up with The New Yorker.

Think about it: the ascendance of the telephone (spontaneous, social) drove people to bemoan the death of the letter; the explosion of E-mail has got people lamenting the dear old phone call. Similarly, many critics expressed alarm that the entertainment dominance of movies (social) would displace book reading (isolating). Now these people fear the extinction of the social cinema.

The Net is unlikely to replace any of these activities, although it will most certainly steal time from all of them. At its best, the Net helps people save time by rendering information accessible while extending the lives of its inhabitants into a new realm.

As designers, we face the same paradoxical issues that city planners face. How does one meticulously plan, design, and build spaces that foster spontaneous, varied social activity? Web designers have witnessed, already in the Web's infancy, the same phenomenon familiar to urban planners and their clients: a well-conceived, carefully planned community is often less successful (less socially gratifying) than an unplanned one.

I would start by hypothesizing that a successful community (I would like to hear people's definitions thereof) is built on the coexistence and balancing of opposing forces. When these forces are in equilibrium the community thrives, when one force dominates, collapse ensues.

I would like to invite people to contribute to a list of these dynamics in an attempt to identify the forces. Here are two to start:

1) Accountability. This is necessary to ensure that members act in good faith. There must be enough accountability to encourage responsible, constructive behavior. On the other hand, too much accountability stifles difference and leads to repression. Ebay has been very successful at creating a system that encourages responsible behavior by building profiles of members' actions. The system rewards with cute icons those who conduct good transactions without treating harshly or locking out someone who screwed up. Ebay also remains open to new, untested members, which brings us to...

2) Accessibility. The community must be accessible enough to grow, develop, and attract suitable members. But if it is too accessible -- and I include ease of discovery in accessibility -- it cannot maintain its distinct identity or prevent the casual or irresponsible intrusion of unaccountable visitors.

-- EL

Date: Fri, 25 Feb 2000 10:48
From: Andrew Shapiro

I think Steven has gotten us off to a great start here. The Net is ours to design. It is not inherently isolating any more than it is inherently democratizing, paradigm-smashing, or [insert cliche-of-the-moment here]. It's about potential and what is realized. So how do we design the Net to maximize the social good?

With regard to community -- I think we have to acknowledge that there are features of today's Net (and today's networked culture) that could lead to greater personal isolation and social fragmentation, that could test our ability to get along as neighbors, even as a nation. The main culprit, to my mind, is the potential for excessive personalization: Too much MyAOL, MyWay, MyThis, MyThat cuts us off from shared information, common experience, and eventually empathy for one another. (Notice, by the way, that I'm not talking about "online" vs. "offline" communities, an increasingly meaningless distinction.)

Granted, there's lots of conjecture in the preceding paragraph, which I'd be happy to unpack upon request. But let's assume for the moment that we agree about this. That way we can move on to the more interesting exercise of figuring out how we design around such potential harms.

I like Eric's suggestion that we start by identifying those things that make community -- whatever that is -- successful or worthwhile. Eric says that there are forces that need to be in balance in order for community to work, starting with Accountability and Accessibility. I agree with the balance idea, but see those two values working together somewhat differently, more like this:

1. Accountability <--> Do-Whatever

Without some accountability for our actions, community breaks down. (Trust is lost, etc.) At the same time, too much accountability deprives us of the right to be different, free, occasionally irresponsible. So balance #1 is between having some recourse for "bad" behavior and allowing people, to some extent, to do whatever they want.

2. Accessibility/Fluidity <--> Stability

To avoid staleness, communities must be open to newcomers and must allow folks to leave (thus the term Fluidity, which suggests ease of entrance and exit, not just the former). However, a community that doesn't balance fluidity and stability won't have any identity. Moreover, too much ease of exit may sink any real communal spirit, because folks can just up-and-leave whenever they want. Contrast that with the best communities, which often seem to work because actors know they're stuck in it together and so they just have to deal. As Jeff said, community -- whatever that is -- takes work. And that means some rules, some structure, some glue.

Now, how do we code that?

-- AS

Date: Fri, 25 Feb 2000 10:17
From: Heather McDonald

The strange thing for me about this research issue is that maybe sometimes I feel this way as a producer of Internet content, as a person working at an Internet start-up at the end of the day when I need to stop myself from IM and emailing my coworkers, friends and lovers and need to transition into actually speaking with people.

However, the gURL.com user base and active community members are in high school and obviously living a different reality than adults using the Internet. I recognize that the Internet can be isolating and I think I sympathize as much as a single person living in NYC can with the parents of teens watching their Internet savvy kids surpass their own ability to collect information, find anything, and participate in a virtual community. (It is also very isolating for people who don't have access or don't feel that they have the skills to understand this media.)

However, when I read these studies I can only think about what I do and how I hope it is making high school girls feel less isolated. When you are high school (unless you are homeschooled) you do have at least six hours of forced socialization every day so I think the Internet can provide a nice zone for alternative forms of communication even with the people you know from school.

-- HM

Date: Sun, 27 Feb 2000 14:09
From: Eric Liftin

Heather's post reminds me of the old days of Firefly. For those that don't remember, Firefly launched in mid-'95 as a music-recommending site. With proto-buddy lists, instant messaging, intramural email, rudimentary chat functions, and, most importantly, crude personal home pages, something we called a community soon formed.

Most of the members were teenagers. Attention to the technology-showcasing music-rating areas of the site began to wane as teens surfed personal pages and buddy lists. Kids adapt easily, and these well-documented effects indicate directions of interest in the medium. The software tools allowed communication and information sharing to flourish while the music function was used just enough to help define subcultural spaces. Firefly continued to draw hundreds of concurrent visitors months after the company had been bought by Microsoft and completely dissolved.

Its popularity was due substantially to the important dynamic on the site between public and private. The personal pages displayed a decent quantity of information: friends, favorite music and movies, age, region and whatever people wanted to type and upload. Privately there was the messaging, which was safely confined to the site, away from personal email addresses. The public display enabled people to surf and identify each other. The private channels enabled the formation and deepening of relationships.

-- EL

Date: Sun, 27 Feb 2000 17:14
From: Steven Johnson

Eric writes:

The public display enabled people to surf and identify each other. The private channels enabled the formation and deepening of relationships.

I think this gets us right to the heart of one of Andrew's concerns (and something implicit in what Jeff was saying too.) Is there a risk in the "surfing and identifying" phase that communities will form that are excessively focused? In other words, do you end up with a series of demographic closets, with all the Housemartins fans crammed into one virtual space, completely oblivious to the Billy Bragg fans down the hall -- and on another planet altogether from the Marilyn Manson devotees? You want people to be able to find people who share their tastes, but is there a point in our real experiences with online communities where the groups gets too insular? And what kind of design keeps that from happening?

-- SJ

Date: Sun, 27 Feb 2000 20:11
From: Heather McDonald

I think one thing that keeps groups from getting too insular is the participation from people who totally disagree with everyone else. I have found in most active online communities people with very strong anti-Housemartins feelings are always out to find the Housemartins lovers' sites and rile everyone up.

I think as designers of online communities we need to be concerned with creating environments where users are really the designers. So many sites out there are fabricating communities which seem empty, lonely, and obviously contrived. It is easier to have honest interchange in a world that is created by the needs of its users rather than the needs of a sponsor or the ideas of an editorial staff. The challenge is to listen to the user and get them as involved as possible.

-- HM

Date: Wed, 1 Mar 2000 16:28
From: Jeffrey Veen

Heather McDonald wrote:

I think as designers of online communities we need to be concerned with creating environments where users are really the designers.

Ah ha! Now we're getting at it. Part of the problem with "designing communities" is that we fail to distinguish between rules and tools. We've been talking a bit about how much work it is to keep communities healthy and active, but what we haven't touched on is HOW to do that.

Designing rules into a community is probably a necessity. "Keep your posts on topic." "Don't be abusive to other community members." "If you don't follow the rules, we'll kick you out." Fine. But the reality is that rules only go so far. What Heather is referring to are tools that empower community members to create self-sustaining conversations and interactions. Maybe some examples would help.

Spend some time on eBay, and you'll start to see how simple tools are used to keep people on track. In the strictly commercial world of auctions, eBay has added a self-policing feature that plays very well towards a healthy community. The concept is simple: Community members can rate the experiences they have with sellers. The more honest the sellers are about the quality of their auctioned goods, the better their rating will be. It gets right into the interface design heuristics of eBay: Your name is forever bound to an icon proclaiming how good or bad you are. Follow the rules or we'll make you wear a big scarlet letter A on your chest...

Likewise, both Amazon and Slashdot give users the ability to rate posts. Community members (and the "anonymous cowards" who don't bother registering) are motivated to make thoughtful and productive responses to ongoing threads. And again it affects the user interface: Well-received posts bubble to the top; crap sinks.

So good tools, I think, can make people follow rules. But can it make them cross-pollinate? I wonder...

-- JV

Date: Wed, 1 Mar 2000 22:50
From: Andrew Shapiro

Interesting question, Jeff. I think the answer is no. Though I said earlier that the Net is ours to design, tools alone cannot make people do anything (cross-pollinate, talk to our neighbors, avoid solipsism). That's particularly true if we're talking about site-specific interfaces which can be jettisoned with just one click. It would be different if the Net were as top-down as AOL, but (fortunately) that kind of macro-design seems impossible.

So maybe we're left with a bit of a design dilemma. Our culture has embraced a communications technology -- distributed, interactive, bottom-up -- that resists social engineering. Individual site designers might add all sorts of features to get people to stay engaged, or whatever, but the benefits of this would seem to be slim.

At some point, people themselves have to decide (free will!) whether to take affirmative steps to avoid narrowing their horizons, to seek out difference, to embrace community. The question is whether the tools to make this happen are readily available. It's not so much When Bad Tools Happen to Good People, but rather A Good Tool is Hard to Find.

I do think good tools are hard to find -- tools, for example, that facilitate sustained communal activity online. Anyone agree? Disagree?

-- AS

Date: Wed, 01 Mar 2000 23:52
From: Eric Liftin

The LOOP discussion on this topic has raised an important point: Any kind of visitor interaction is not necessarily "community," and in fact "community" is often less desirable than other forms of online interaction. The term "online community" is too vague to describe the varied, genuinely novel dynamics that develop online.

Online software "tools" are capable of provoking certain behavior or facilitating actions. It is often fascinating to observe the results of putting a chunk of free software out on the Web. Our site called Random Access Memory (http://randomaccessmemory.org) is a very simple tool for archiving human memories and browsing through them. There is no community, but this "tool" has provoked entry of over 2000 memories.

Andrew rightly observes that site tools are generally too parochial to enable social interaction that is complex and rich with difference. The next logical step is to move beyond community to a wider field that is capable of enfolding multiple sites, multiple actions, multiple levels of interaction, both public and private. This step is toward Web Urbanism.

Just as in cities, the layers of infrastructure provide sites and the potential for all kinds of actions, the Web will transform itself into a multilayered field. Software tools will be added that underlie multiple sites instead of residing on one. Andrew's concern is that the very elements that characterize Web sites tend to isolate them from each other and thus to isolate their visitors. His dilemma is how to encourage friction in a ruthlessly efficient medium.

Web Urbanism takes cues from successful cities. People choose to live in polluted, overcrowded cities because they're drawn to the resources -- the layers of infrastructure, both technical and social. If we build sophisticated, open systems that allow for internal difference and degrees of autonomy -- that balance dynamics of public and private, access and restriction, accountability and freedom -- these electronic spaces will be irresistible.

-- EL