Interview by Marti Louw :: May 2008
Louw: You're an academic researcher within a business school yet your research centers on online communities. Can you tell us a little bit about your work and interests?
Butler: I've been looking at online communities in different ways and with different frameworks for about 15 years. Some early work I did was around listservs and email-based communities and groups. Currently I'm looking at Wiki technologies for supporting online groups. The question that particularly interests me is: "How do you build and evaluate these online systems?" My focus is not so much what a particular person will do, but rather how individuals behave in aggregate and the dynamics behind that — so looking at how people come and go, how content evolves, how things grow over time and what happens when you get a whole bunch of individuals together who may not act like the average individual. Is that a good thing, a bad thing? How does that play out? How do you design for that? A lot of my work ends up being design consultation for online communities that brings together sociology, psychology and organizational research.
Louw: In terms of online communities, what are hot topics or "special issue" kinds of journals subjects you are encountering?
Butler: There is a lot of research attention being given right now to communities in Second Life and virtual worlds. There are a host of really interesting research questions around what to do with these environments — trying to understand how they are different; how you make decisions about when to use them and characterizing how they work at the individual and group level.
A second area with a decent amount of scholarly activity currently is in evaluating and figuring out what to do with emerging online technologies — so social networking sites like Facebook, MySpace, and knowledge co-creation tools like Wikis and even blogs to some degree. There are lots of examples of how to use them outside of an organization. But what happens if you place them within an organization or an existing community? How does that work and how do you think through that design problem? It's not just creating a Wiki. It's trying to figure out how to strengthen your community, broadly speaking, or your organization. There's a lot of research that'll be about making those decisions.
Some other areas of recent interest seem to be in language, discourse analysis and conflict in these online spaces. Some researchers are looking into diversity and what happens when lots of people come together in online settings that wouldn't otherwise be together. And there are some interesting ideas about how online culture works in those spaces.
Louw: We have so many terms for describing online communities — learning communities, social or knowledge networks, knowledge building or sharing communities, communities of practice, purpose, interest, electronic communities, virtual communities — just to list a few. Are there categories or useful dimensions to characterize the different flavors of communities? Is a typology of online communities emerging?
Butler: When characterizing online communities, one important dimension that comes to mind is looking at whether a site is more social or information-oriented. There is a lot of research going on that's trying to understand what happens in a online community and why people participate. For example, some researchers (Ren, Kraut, & Kiesler, 2007) are looking at role bond vs. identity in online communities. This really has to do with why people are there. Are they there for the other people or are they there for the group as a whole? That seems to be a distinction that's useful for lots of reasons. Understanding whether people are participating in a community because of the people or because of the information directly shapes how community designers should present the community to potential participants, what kinds of features should be included in the members interface, and how to evaluate the viability of the community.
And I've seen some people try to argue that learning communities are a unique type of online community. There are some general principles and ideas that seem to apply across all types of communities, for example having to do with member dynamics.
So my short answer is 'yes' — there are certain typologies you can use, but the challenge is figuring out whether you are dealing with a problem where you can learn from other communities or are dealing with something that's unique to a particular situation. Rather than getting fixated on typology, though, I think that it's important to determine what characteristics matter and use those to intelligently determine which cases you can learn from and which you can't.
Louw: During our two-day CAISE Inquiry Group, we talked about six professional sites presented (ITEST, MSPNet, ExhibitFiles, InformalScience.org, SMILE, Coalition for After-school Science) as varying along a continuum from being project-oriented to field-oriented. They also varied in terms of user contribution from being closed to open-access. Can you talk a little more about your thoughts regarding this continuum?
Butler: Well, we talked about how some of these sites seemed very project-oriented in their focus. In these cases, the project exists because of NSF funding, or a set of partnerships, or as a private endeavor. And so the technology that's being used to support any organizational entity is independent of the site. In contrast, other sites had a more open, ad hoc organizational strategy. In those, if the website disappeared, the particular online community it supported probably wouldn't be able to assemble again. It's part of a larger community in which no particular organizational associations exist other than the site.
The sites we reviewed seemed to fall along this continuum of being organized around projects to the field. We had clearly project-oriented sites like ITEST and MSPnet that have certain characteristics in common — the sites seem to be funder- or goal-driven, and focus heavily on supporting project teams. Once a project is completed, the motivation and requirement to participate drops off. Project-based sites are bounded in the sense that you don't have membership that changes radically over time because it's so temporary. These sites are very different when compared to the field-building, professional community-oriented sites like ExhibitFiles and InformalScience, where projects are used as an organizing unit in a very different way. Here, there may be an infrastructure for supporting the projects, but people don't necessarily come to the website because they are part of a project. Users come to browse the work of others, post exhibition reviews or evaluations, comment on projects and belong to a community. The SMILE Pathway seems to be aiming to serve both. From a systems perspective, the back-end may have a project-oriented infrastructure, but the front-end is really aiming to create a community that in itself exists for other reasons. The Coalition for Science After School web initiative also seems to want to bring together a diverse group of professionals from different arenas that wouldn't normally interact except for the fact that they're interested in and drawing on the same resources.
Louw: Can you talk about what you think constitutes a "community."
Butler: I think we have to use caution when we apply the word "community" to an online group or set of affiliations. When I wrote my dissertation in the late '90s, I really didn't want to use the word "community" because it's such a loaded term. You say community and people imagine a small town where everybody knows everybody's name. One can get into these big long arguments about whether something is a community, a group, a voluntary or professional association or a loose set of affiliations. It turns out that the dynamics and theories associated with those types of organizations actually describe online social structures better than studies of small groups or traditional organizations. This is in part because of differences in how fast people can join and/or leave. In most studies of small groups, it is assumed that people stay involved for the duration of a session/meeting/or project. But in studies of traditional communities, it is assumed people can come and go – but that it is a high effort activity. In studies of voluntary associations, however, entry and exit are low cost (and hence quick, high-frequency activities).
In my dissertation work, I actually tried to come up with another term and ended up talking about online social structures because I wanted to use the term "collective"... but the business school professors thought I was turning into a leftist. So, on the one hand those terms mean something very precise, but the reality is that they're all being used synonymously and people aren't thinking very carefully about what the difference is between a network and a community.
Louw: Knowledge-sharing is frequently described as being an important aspect of online communities that is focused on professional development and peer learning. Can you talk about this?
Butler: Knowledge-sharing is another term I would be cautious in using. A knowledge-sharing network or community provides an extremely limited framework for thinking about what goes on in these sites. The literature talks about knowledge-sharing or knowledge-contribution, but one of the things that academics forget is that public speaking is right up with death and taxes in terms of what people are anxious about. The thing we blindly call "knowledge-sharing and contribution" is, to most of the rest of the world, a form of public speaking. The reification of oneself and one's ideas online is extremely off-putting for many people.
Looking at learning communities in terms of knowledge-contribution tends to make our analyses very mechanical and resource-oriented because, in reality, it's a very complicated social act to say something — to publicly ask a question whether you do or don't know the community. So by focusing narrowly on knowledge-sharing, you miss a lot of the psychosocial dimensions of behavior that have huge impacts in these spaces.
If you want to talk about learning communities, contribution in some ways assumes that learning happens only on the consumption end and that the contributor is not getting anything out of it. But from what I understand in the education literature, we know better — that teaching somebody something is actually a really good way to solidify your own learning. So by looking at knowledge as resource contribution you turn it into an atomic, isolated action where the only things that really matter are how much time it takes and what your motivation is.
Louw: How do you see online communities being used most effectively for professional development? What seem to be the most valuable aspects of participating in an online community?
Butler: To the degree that professional development is a learning activity in the traditional sense, it comes down to figuring out how to do something. There are relatively straightforward benefits to learning in a social environment, for example the opportunity for peer feedback. The professional development aspect of online communities, which I find more interesting, is the opportunity to rehearse oneself as a professional. So it involves becoming professional beyond the formal training aspect, and is concerned with being able to speak, both inside and outside of work in a way that reflects the expected norms for that profession. I think that the potential for learning the discourse patterns and social practices of a field is an important professional development activity that online communities can support.
In my mind, the more interesting thing to think about in online settings is how you can help people that, for whatever reason, may not have access or time to continue those social aspects of professional development. People may go to a class in order to keep their certification or a training seminar to learn new skills, but that doesn't team them how to participate, how to talk and how to be an expert socially. That's why thinking of professional online communities not just as websites is important — because you're making the social aspect of things more open and something that people can learn from. And active participants gain visibility and build reputation. Even if people don’t actively participate or contribute, they can track the flow of ideas, the conversations and the nodes of influence in a professional community.
Louw: How do you think about or define professional development?
Butler: To be an expert, there are certain things you have to know — and there is skill, and then practice. You have to know how to do some stuff and more importantly be able to actually do it. You have to be familiar with language in terms of knowing how to talk about what you do. Then there is networking and relationships – you have to know who the people in your field are and be able to create relationships with them. In academic speak, this is social capital. You have to be able to learn new things about the subject of your profession. So just having knowledge, skills and a network doesn't do you any good if you can't learn about emerging ideas and research. And then last of all, you have to have reputation – that may be a global reputation or it may be local. But if you're in an organization and nobody knows that you're an expert on X, then in a sense you're not. Your ability to have an impact and do the things you want to do is very limited if your professional expertise in not recognized. So those six things are crucial – knowledge, skills, professional discourse, network, learning and reputation/visibility. To evaluate a professional development website in terms of the NSF Framework for Evaluating Impacts of Informal Science Projects we've been discussing, I could probably force all but perhaps the reputation and the social capital aspects of professional development into the framework categories of knowledge, engagement, skills, attitude, and behavior.
Louw: During our discussions, it was frequently mentioned that to succeed, these professional development websites must fit into or draw on existing practices. Do you have any favorite examples to counter or support this that describe how a communication technology has changed or affected professional practice?
Butler: The example that I usually give is how the business memo as a communicating genre was adapted for e-mail. If you think about how we construct an e-mail message, it starts with three fields – From, To, CC Subject. This follows the format of a business memo rather than a business letter, the latter of which having much of that information placed at the bottom of the page instead. It turns out the business memo format comes from filing cabinets and how filing systems work. They wanted to have that information on the top so they could easily find documents placed in a filing cabinet. Joann Yates' book traces the evolution of the business memo into an e-mail and more generally tracks how successful communication technologies adapt pieces of existing practice by only introducing small changes.
I don't know how successful it has been, but I saw this intent to augment existing practices and activities with ExhibitFiles, which is clearly trying to change what you do upon completion of an exhibit project and who you tell. What is promising is that they're asking exhibit designers to share pictures and stories about their exhibits, not write long summary reports. So they're essentially asking them to go do extra things that they've never thought of doing before. And they're using terms consistent with how they described their exhibits when they made brochures or collateral materials. What's different is that they haven't spent a whole lot of time communicating with other designers about that exhibit. That's just not where the focus is because that's not what gets funded and not what their boss says is important.
So they're using the tools, the language and existing structures of professional practice and saying, "Let's use this to share and start a discourse about exhibit design." This adds a reflective step and makes designers more critical in a good way — it urges them to think through, evaluate and make comparisons that they otherwise might not have done. So I could see the potential for this site to be a loop that builds on an existing practice, but then also transforms it.
Louw: Do you think we're undergoing a shift or a change in practice with some technologies now that we're not aware of? Something similar to email with the more participatory or social media kind of things that are coming out that are going to stick and change everyday communication?
Butler: Let's use designers and developers of informal learning experiences as an example. I suspect for people at a small number of institutions that have traditionally had lots of resources, that could travel and have lots of professional networking opportunities available to them, that these new technologies are not really going to change their life very much at all. But a large percentage of other professionals will now have choices that they didn't have before. The technologies will result in an expansion of who's visible, who's changing and who's leading the field. That's the kind of thing you can't predict in advance and it will probably take years to play out. But I think that the online community technologies we're talking about have the potential to change the dynamics of a profession — both in terms of the speed at which change happens and the potential to catalyze new professional areas. I don't think it's a new dynamic, but who can be involved and how much it costs will change over time.
Louw: Will these new technologies be a democratizing force in professional arenas — reducing barriers to entry and breaking down insider clubs that tend to exist in fields or will social networks just form in new ways?
Butler: Well, those clubs exist for reasons. You can't interact with everybody in the whole world. It's too much. So there will always be boundaries and there will always be structures and networks. And technology comes in and in some cases forms new structures that replace the old ones. Of course, in some cases the old structures adopt new technologies and adapt.
There's actually a study of use in organizations that essentially asks this question — if you bring these new technologies into organizations (this was done in the early '90s so it was looking at e-mail), do they produce more centralization and control or do they produce more decentralization and what you might call democratization. And what they found was that in organizations with already highly centralized control of resources and technology, centralization just increased. And in organizations that were more decentralized, they because even more decentralized (Pinsonneault & Kraemer, 1997; Pinsonneault & Kraemer, 1993).
So is technology going to magically change an organizational structure that is highly centralized and insular into an open, decentralized one? Probably not. Democratization is not an inherent outcome of technology. Wikis are a good example. You can take a Wiki and make it incredibly hierarchical, self-regulated and centrally controlled or you can have it open to anybody. And there's a wide range in between.
Louw: In 2008 the Informal Science Education division of NSF issued a new framework for evaluating projects for both public and professional audiences. In the Inquiry Group, we spent a long time discussing and trying to apply this framework to some real sites. From a research standpoint, did you have any overall thoughts about this framework?
Butler: I think there was enough ambiguity in the framework to allow for broad interpretation of impacts. That's good, because otherwise you can get into endless academic debates about what the differences are between awareness, knowledge, skills, and behavior. But the distinctions probably don't matter for NSF's purpose as long as you can fit one of your intended impacts into in one of their boxes. I think attitude is one that was hardest for people to get their head around — why you would want to change someone's attitude? But when you equate that with norms and values, that impact category makes more sense to me.
Louw: We spent two days talking about NSF-funded professional online community websites for informal science education and how to assess their impact and value. What are your thoughts on how determine the impact and success of these web initiatives?
Butler: As Randi Korn said in her talk, the exercise of stating what it is that you actually want to have as the impact, even if you can't measure it, is an incredibly valuable exercise. But then you have to try to understand what's going on in the community that's achieving that impact. You have to think about your measures depending on whether you are trying to achieve community operation objectives or field impact objectives (see diagram).
I think you can actually use the evaluation framework to identify awareness of a community, knowledge of how to participate in the community, engagement and interest in participating in the community, attitude toward participating in the community, behavior of actually participating in the community, and more. And you can think more specifically about what you mean by participation in an online setting. Do you mean reading something? Do you mean contributing something? Do you mean commenting on something? Do you mean having an extended discussion? All of those would fit back into the impact categories as they relate not to a professional field, but to a community member. However, if your impact objectives lie only in community operation goals, then you run the risk of becoming a bureaucracy or an entity that exists only to propagate itself rather than an agent of change aimed at impacting the larger professional field.
And so one possibility would just be to say that if you're going to talk about online communities for impact, you can use the professional field — but you really have another area that you have to pay attention to, which is the community member. If you don't pay attention to the impacts on both of those, your site isn't going to be sustainable.
Louw: How do you distinguish between a professional and a community impact?
Butler: It really comes down to understanding the reasons underlying participation on your site. For many users in most cases, their participation in an online community site is not going to lead to a professional or field impact. That is, unless a site has such a standing that just saying, "I contributed something to InformalScience.org," for example, constitutes a professional activity like publishing in a journal. But in most cases participation is going to be something that is an internal activity, and you need to understand its relationship to professional development. So to give a concrete example, being aware of preexisting work in an area is a professional impact; being aware of ExhibitFiles is a community impact.
You could say that you need to have impacts on the larger professional community for the impact categories. But you also need to think about the impacts on the behavior of individuals involved in the community. One way of dividing up the space might be to consider field-level impacts, community-level impacts, and project-level impacts.
Louw: If the unit analysis targeted is greater than an individual impact, what might be the measures?
Butler: For example, ITEST and MSPNet seem to be focused on project-level impacts because they're primarily providing an infrastructure for project teams. They're interested in making the project team is more successful, which is different than each individual investigator involved in the project being more successful. Interestingly, I've been involved in several projects where all the investigators would say that they were individually pretty successful, but the project was a disaster.
For ExhibitFiles or SMILE, they may be looking at larger scale impacts where they want to know about things like whether the field is becoming more innovative, or self-critical. The tricky question is how to find evidence of such impacts.
Assessment is a nasty problem once you get above the project level. How do you assess something in the field without it becoming mostly about individuals? What it comes down to is that the data is exactly the same, but interpreting it makes the difference. Let's say I do a survey and find out that 35% of people download at least three evaluation plans or view three project pages or reviews before exiting a site. Is that an indicator of individual behavior that affects their ability to be a better informal science researcher or practitioner, or is an indicator of a behavior being propagated to look at multiple sources. It really means both. So how you take that into consideration and which unit of analyses you apply depends on the impact level you target.
I think someone said something in the Inquiry Group about how you're almost always going to be talking about some of the higher-level units of analysis and units of impact because NSF, a private foundation or any other funding agent is usually interested in bigger impacts.
Louw: There is an "Other" impact category provided in the framework. What kind of impact for an online community could you imagine for that category?
Butler: Efficiency, perhaps. Being a business school professor, I guess I would probably come up with that one. Again, part of the complication is that behavior, attitude, knowledge, and engagement cover a wide range of impacts. So at the individual level, the cost is time spent. So if I reduce the amount of time it takes to get something done, such as writing a proposal, that cost efficiency is not really reflected in the other listed impact categories.
Louw: Are there any impacts we might be missing by looking through the lens of this framework to evaluate professional development or online communities?
Butler: Well, if you treat behavior and knowledge narrowly, then yes. For example, language and communication have great impact on others. If you're only looking at what professionals are doing, not how they do it, then you might be missing some impacts. And there are probably some things about how you do it that are missing if behavior becomes these actions. For example, if you count actions, you could easily say, "Well, we want to see if people are using audio technology more effectively because they've participated in an online community discussion about audio tools." And so you survey people and ask them to tell you the number of times they've downloaded a Podcast in the past six months. You could do that, but it's probably more interesting to know how they used a Podcast and whether they tried it out and decided that it was of value. Getting into that won't be quantitative; it won't be counted. They'll be messy, qualitative cases. But that's probably where you're actually going to learn more about professional development and where you'd actually be able to see real impact with a bigger consequence. If you define behavior and knowledge too narrowly, you're going to miss interesting behaviors because we're not used to thinking about talking as a behavior.
Louw: Can you see online communities supporting collaboration and innovation?
Butler: Yes, online communities clearly support collaborations. At the very beginning of the Inquiry Group, each of us was asked to describe how we used the web for professional and personal development. Many of us said that we use the web to coordinate what people are doing. In general, online technologies are allowing us to carry out much more distant collaborations. The ITest and MSPNet sites are allowing professionals to collaborate on a larger scale with more diverse groups than before. And I think that there's direct value in that.
Trying to spark innovation within a community, or more broadly a field, is trickier. If a project or individual does something innovative and that gets recognized even in some small way, that exposure and feedback increases the likelihood that an individual will try it again — especially if they're in an environment that does usually recognize the innovative quality in their solutions. In this way, online communities can be incubators for local innovation, especially for people working in isolated environments that lack that positive feedback. Idea validation is incredibly valuable in sustaining somebody who's trying to do weird things — which describes innovation. Increasing external recognition also increases the chances that innovative individuals and projects become leaders in their own organizations and change the status quo. And change requires somebody that's willing to do things that are in many ways irrational.
Moreover, online communities allow micro-cultures around concepts to develop. The sense that there's a bigger group out there from whom you can get help and support is really important in providing external validation. And that goes a long way towards increasing innovation at a local level.
Louw: Can you talk about how you would apply the ISE evaluation framework categories to professional online communities?
Butler: The framework seems to apply well to public exhibitions and programs and their impact on audiences. It make sense to me in terms of professional audiences when you have people that go through an explicit professional development activity such as a training seminar. These framework categories become more difficult for me to understand when you apply them to an open group of people coming to a website for many different sets of reasons. We don't know how they're using the resources; their intentions and so forth, so doing a survey is complicated because if you have some users that come for social interaction and others who come for information and you give them the same exact survey, you might miss the point of their activity. It's not an unsolvable problem, it’s just a complicated one because what you do varies based on what you're interested in, and that's hard to figure out from an assessment standpoint.
One thing I will comment on, which I thought was really interesting to see in the discussions, is that evaluation is based on a scientific model of assessing causality. Yet, it seemed that most of the evaluators had social science training.
Complicating this, you have an evaluation framework that's not really a science evaluation framework — it's aimed at verifying causality, because informal science education projects wouldn't get funding if funders didn't believe that the causality was there in the first place. The evaluation framework seems be far more about how things ought to be working in our organization or with our audiences. In other words, they're confirming or disconfirming a narrative which may or may not have a strong causality component to it. So it seems that evaluators are trying to assess design purposes, improvement or funding. These objectives differ from scientific research which is always after causality. So the evaluation and research methods might be similar, but they seek different objectives.
Louw: Most, if not all of the of the sites presented during the CAISE Inquiry Group are top-down, funder-initiated efforts with clear agendas — be it to improve research and/or practice, increase the cumulative research and evaluation, support project innovation, or aid in networking collaborations in a field of endeavor. Do you see anything problematic arising in these top-down initiatives?
Butler: There is one view that says a top-down approach to communities won't work. Thompson (2005) has an organizational science paper describing the development of a community of practice where they make a distinction between a controlling structure and a ceding structure. A controlling structure is an entity or process put in place to ensure that something happens. A ceding structure is an entity or process designed to enable something to happen. Both are top-down approaches that have merit in different situations. Controlling structures provide ways to make sure nothing unexpected happens. Ceding structures (in fact one can argue that in a large sense that's what NSF is) are top-down, centrally-controlled things that allow unexpected things to happen. You can also have bottom-up controlling structures — for example, nosy neighbors acting as watch dogs. The bottom-up entities can be just as restrictive and just as damaging to a community as top-down controlling structures. It's really an issue of balance. And whether top-down or bottom-up, there will always be controlling structures to make things happen and ceding structures that let things take place. They come into a balance and vary from community to community.
It seems most of the sites we looked at had this top-down ceding approach. What I haven't seen yet is an organically emergent kind of community that gets scaled up and is perhaps more authentic in its origins than these agenda-driven sites. I do wonder whether these ceded ones are sustainable or not. Top-down, especially controlling top-down, creates the illusion that something has been started and therefore it will either succeed or fail. And bottom-up, grass root initiatives may last for a little while and then fade away. But is that a failure? No, we don't necessarily call it that. We don't call attention to it; nobody counts it.
Louw: Why is it so challenging to sustain an online professional community and what are some of the issues, especially when the funding runs out?
Butler: What people forget is that the flow of resources, financial, expertise, time, whatever you like to call it, is rarely self-sustaining. When you cut off the funding, you don't have enough resources for it to continue.
There's a distinction that the poet Robert Bly makes between a network and a community that I find useful to thing about. A network is a collection of like-minded people and the community is a collection of diverse people that choose to be there and may not be as like-minded. His point was that a network is less likely to be self-sustaining because if someone or something unusual comes in, either everybody's going to be able to solve the problem, or address the issue, or have the resources needed — or nobody is.
And so you have the situation on one hand where voluntary choice and human tendency pushes us toward networks that are less self-sustaining in a resource sense, but more personally relevant in terms of complementary needs and benefits — and that’s this thing called a community.
This is made more difficult in online communities because the set of design possibilities is greater. Van Alstyne and Brynjolfsson have a paper (2005) that supports the idea of: if you reduce the costs of switching, people go wherever they want. Their argument is that people are going to tend to go toward homogenous networks. But that doesn't happen because those types of networks are smaller and incredibly costly to sustain, so people don't want to do it. A network (i.e. a homogenous collection of people) works fine as long as the only thing that needs to happen is the activity that all of the individuals involved find to be high-value (or easy). However, as soon as something needs to be done that is critical to group survival, but is perceived by the typical member to difficult or low-value (such as organizing the content or disciplining a disruptive participant), the network runs into a problem. Since no one will be willing to take care of the issue, the group will suffer and perhaps even cease functioning.
In contrast, a community which has members with diverse goals, needs, and preferences, will be better able to deal with these kinds of community needs – because it is more likely to have members who have the skills, resources, and disposition to address the problem.
So while homogenous networks of people may be easier to form (since they play to basic human tendencies), they are more likely to be transient. On the other hand, if you can manage to form a community with more diverse members, it will be more likely to have the capability to do what needs to be done to survive over a longer term.
Louw: So if I asked you to be presumptuous and tell the Informal Science Education division of NSF and Center for the Advancement of Informal Science Education (CAISE) how best to support and manage a portfolio of online professional communities, what would you say?
Butler: I suspect that this is in some ways what the Inquiry Group was designed to do. I think it's incredibly important to figure out what set of objectives or impacts we can reasonably expect online community sites to have. For example, what are the minimum criteria for determining if and when sites become sustainable? I've heard one commercial consultant say that when he talks to someone who wants build a online community, the first thing he asks is, "How many people do you think are out there in the world or in your possible set?" If they come back with anything less than 5,000, then he stops talking to them. Part of what he's basing this on is that if you have less than 5,000 people, the 5% or 10% of those that actually contribute and do interesting things is too small for the site to be interesting and sustainable. Now that may be different in a setting when everybody's a educator, or they have a set of shared values and goals, or it may be different in a community setting where there's a physical space and an identity. Getting a better understanding about what basic things need to be in place for an online community to work is crucial. And having baseline critieria for success will help proposers and funders understand whether you have a large enough potential community, enough money, enough staff or whatever else.
"People aren't thinking very carefully about what the difference is between a network and a community."
"The thing we blindly call "knowledge-sharing and contribution" is, to most of the rest of the world, a form of public speaking."
"A network is a collection of like-minded people and the community is a collection of diverse people that choose to be there and may not be as like-minded."