A Conversation with David Weinberger About the Future
By Rich Bailey
One of the highlights of Chattanooga’s recent Startup Week was a lecture by David Weinberger, Harvard researcher and author of Too Big To Know, about how the Internet has changed, not just what we know, but how we create knowledge. He spoke on the Public Library’s Fourth Floor about “the future as a platform.”
Rather than interview him, I enlisted two of Chattanooga’s own future makers — Mike Bradshaw, executive director of CoLab, and Nate Hill, assistant director of the Public Library — to have a conversation with him about the Internet, startups and Chattanooga.
This is the entire transcript, edited slightly for clarity.
Rich Bailey: Can your research be used as model for understanding what goes on in communities, which has so much to do with information but is not information technology?
Mike Bradshaw: Can I add to that? That is the question. It operates on two levels here with respect to both the community and the startups things we are doing.
[It’s a] measurement problem. People want to apply traditional forms of measurement to the effects of our work here, the emergence of this start up community, which is really kind of a transformative phenomenon that defies our ability to measure it effectively. And even the act of doing so might interfere with it, put it into traditional modes of organization, structure. We’ll start managing outcomes that are declared from above rather than springing from within.
To connect this — and then I’ll be quiet and love to listen to what you have to say about it — all of our startups have to use basically a Newtonian model to project their future plans. They start with an initial state and apply dynamical rules to it, which is their business model and then [they measure in] one year, two years, three years. Everybody knows it’s bullshit, but you have to do it.
David Weinberger: You have to get to 10x in five years.
Bradshaw: It’s the hockey stick.
Weinberger: I’ve done that.
Bradshaw: Yeah, me too. So it seems like what you’re hinting at is some sort of a means of expressing possible futures without having to go through this really Newtonian process. It doesn’t seem to be applicable. Nobody believes it, yet we’re required to do it.
Weinberger: That’s really helpful. I will be using that, so thank you the business model is a perfect example of Newton gone wrong. I want to keep repeating: Newton works and Newton does predict, but yeah absolutely. [Several times in his lecture, he acknowledged the validity of Newtonian physics and a generally Newtonian approach to business but critiqued their limitations.] I think the Internet innovation, entrepreneurialization space is way ahead of me in this. I’m not coming up with a new business model for them. This is what they’ve been doing.
One of the most interesting people is Evan Williams who, with a partner, invented one of the initial blogging platforms for free and then was one of the founders of Twitter and now is doing Medium. In all of these cases, he created an application that had a very particular use but was really open ended in what it would be used for and what its bus model would be. He was absolutely right to do this.
Twitter started out with a very different idea of what its use was. It has been extremely open to enabling the uses that it is growing into because this is how its users are inventing it.
Do you know medium.com? As I understand it — and I don’t understand it really well at this point, I use it sometimes — Medium is a place where you can go and you can write and post something. It was designed to be an esthetically comfortable writing place. But as far as I know, they were explicit about not knowing what it would be used for. They said, “Here’s a writing platform. Write stuff, and let’s see what happens.” And it’s unmonetized. How are they going to make money? Where’s their business plan? In typical Internet form, they don’t yet have one. Twitter sort of has one but launched without a particular one.
It seems to me that that approach is exactly the sort of thing I’m talking about. Namely, we’ll start off with a thing that has some uses, but we know we don’t know how it will be used. We’ll maximize our success if we enable this thing to grow into people’s uses rather than hoping that we hit the expected and anticipated use correctly and we’re providing THE functionality that users need. This is how software and everything was designed forever, but the notion of putting something up and letting users expand it and decide what it’s going to be is new. It looks like lack of planning, like they’re not mature business people, but they’re way smarter business people. They know that they cannot predict what this thing will become. If you don’t know what it’s going to become, it’s very hard to lock it into a business model and how you make money exactly. And you hope you figure it out.
Reddit’s going through it now. The founders of Reddit have been very explicit. Alexis Ohanian has a book out less than a year ago that is exactly about this, that is, about trusting the community and letting this thing gain value. Because the community has a sense of ownership of it, and you figure out how to make money out of it. Don’t know what its going to become, cannot predict, but that’s how to let it gain value far beyond any value that you could have predicted.
Nate Hill: This is exactly the sort of model we have used for how we think about the Fourth Floor. We think of it as a physical platform for the community. [To Mike] Any time you’ve talked to me over the last couple of years and said “What’s your plan, what are you going to do next,” I usually dodge your question a little bit, right? Because, ultimately, I don’t have a plan that I rain down upon the fourth floor. It’s a community-driven thing. Think of it as a place with books and open APIs, as a place the people can come and create that next iteration of what the library is going to be for them.
I think it’s really important to note that this kind of concept is not entirely bound to the Internet. The Internet has changed the way we think about all these things, but we can apply it to urban planning, to business models, to libraries as well.
Weinberger: Exactly. It’s easier in the Internet because everything is just bits and you can find everything and it’s very cheap to fail and its easier to build on. But it’s an old idea. Jane Jacobs [wrote about it with urban planning]. People will want “here’s where the barber shop will go” or “make the sidewalks this wide” and they don’t anticipate bicycles or food carts.
This is far from a new idea. It is easier to do in the online world. And because from my point of view, the online world is becoming the paradigm by which we understand the rest of the world, it’s naturally…
I’ll give you an example from the library we were talking about earlier. [To Hill] This is maybe your example so may be you should say it.
Maker spaces are all the rage in libraries, because they want to have a physical place where people can come and engage. And the typical one has a 3D printer.
They’re an interesting experiment. It’s great for libraries to try them out, but they tend frequently to overanticipate what user needs are: “People love to have a 3D printer.” Well, yeah they might, and many of them do, but in many cases they’re not organic. They’re providing anticipated services to users. Those services may be great, but it’s a different model than the one here, which is to provide a place where people can try out, experiment, build new things and see what they’re doing and address the interests that emerge. Which seems to me to be a much richer environment.
Hill: I think the thing that’s special to this library and Chattanooga in general is that we work really hard on the culture of the place. With providing new services on the 4th Floor, in some cases we have totally dropped down a piece of equipment and been like “All right, let’s see what can we do with this.”
Weinberger: That doesn’t contradict what I said.
Hill: But then we work really hard to build culture by working with other organizations and other people around the community to turn it into something more than a piece of equipment that sits somewhere. And that’s not necessarily a library thing. I think Chattanooga is really good at…
[To Bradshaw] You guys do this. You built the Gig Tank culture. It’s not just about the particular things that come out of it. People know that this is the Gig Tank place. That attracts a certain kind of talent and it turns into a different kind of feeling for the city, right?
Bradshaw: Yeah, and that’s exactly the point that I was asking about in one part of the earlier observation. Was that unplanned emergence of something that defies standard quantification but yet it’s reality is indisputable? How do you talk about it? And when you think about the dominant metaphor changing, as we deal with the Internet and elaborated networks becoming the dominant metaphor, how do we think about these things so they don’t have to have this highly quantified [approach?]
Weinberger: Events like Startup Week — as I understand it — are about thickening the culture, which means increasing the connections, which is what a platform is really about. The fact that you make a decision to invest in some piece of equipment because people will find it helpful or provocative… of course you do. But just as you say, you’re doing this in an environment where its use and meaning can be elaborated by the culture and the users. That’s what culture in general has always been about. It’s been about shared connections around share experiences, shared values, shared things.
Hill: “Thickening the culture” is a wonderful turn of phrase.
Weinberger: It’s from social networking research. That’s hugely important and it’s also really difficult to measure, especially if you set out a goal. What are you going to do, say everyone should have seven networked buddies?
Bradshaw: That’s right. And yet that’s sort of the way it comes down when you look at — first — people seeding the waters, adding some sort of energy into a self-organizing system. The available free energy comes in the form of capital and people’s interest and creating nodes on the network. After a while, they are going to go, “How many mentors did you attract into your [project], and what’s the investment amount that you secured for these startups.” And they’re setting these things, which do become your goals if you want them to keep funding you.
Weinberger: “If you can’t measure it, you can’t manage it” is the idea.
Bradshaw: And I believe that’s completely wrong. You’re right in what you say.
Weinberger: I think often what people mean by that is “If you can’t measure it, you can’t kill it.” So you do this library experiment and somebody says, “Well that’s fine, but how do we know if it’s working?” And the right answer is “Come into the damn library. If you can’t, then okay let’s talk but…
Hill: Were there any VC angel investor type people in the audience, Mike? I’m kind of curious how this rings to the guy who has all the money and is thinking about making an investment. What does it sound like to this dude who chooses to invest in this startup that’s like two really talented people and maybe their idea isn’t the greatest idea yet.
Bradshaw: Most of the time, even though they’re going to ask you for the projections, they’re really looking at you. This person and that thing you’ve got inside of you that you can respond to the environment appropriately. The actuality of it is when you’re talking through things with the investor, you go “Okay, these projections really just display the quality of my understanding of the current state, and here are my assumptions, and you can take a look at them.” And the investor is sitting there looking at the founders thinking, “Are you the person that’s going to be able to manage all this when that doesn’t happen?”
There’s something else. Have you ever watched a Mandelbrot Set draw itself? [This is fractal geometry, in which a huge and complex form creates itself by the repetition of small, simple shapes.] It fills in, fills in, fills in — that’s incrementalism to me — and then bang its out there. That seems to me to be a great model for the effects of incremental [changes on a startup]. At a certain time, under the proper conditions — under some conditions, and generally unpredictable ones — a new level of organization springs in.
Weinberger: It’s a phase change.
Bradshaw: It is a phase change. It’s a qualitative shift rather than an incremental shift. And that’s sort of what I saw happen here in Chattanooga, first as an observer and then a participant. I saw a phase shift in the culture. And cultures like the Internet don’t have gravity and mass. They’ve got a certain kind of inertia that’s ghostly but also they change.
Weinberger: Without some inertia, there wouldn’t be incrementalism. It would be totally new every time.
Bradshaw: That’s right. There’s a certain momentum.
Weinberger: You want that, right? There’s a certain coherence.
Bradshaw: You’ve got to be able to relate that to somebody’s understanding of what they’re getting into and where this thing’s going and everybody wants to know it. So how do you feel back about “Just sit back and let it happen”? It’s the best answer I’ve got.
Weinberger: You’re trying to get a phase shift in culture, and it starts with a seed, a set of people who understand it and are pushing it. And you have to constantly explain to people, even well meaning people — forget the antagonists — you’re constantly explaining to them, [saying] “What’s important about what we’re doing on the Fourth Floor isn’t the particular instantiation that you see” and “We love the open data portal, that’s great, but you should understand that what’s really important is…” You put it many ways. It’s the culture, it’s the openness, it’s the fact that we didn’t have these ideas when we started and they evolved.
So you constantly have that discussion to try to make the culture clear. And one by one people get it. What’s going on in the library… we can’t use the old measurements for that. It’s not the number of books checked out. It’s not even necessarily the number of seats of people who come in, because it has a wider impact.
Over time you win that argument by giving people a vision of the culture that they embrace, which means giving up on the old existing measurement ways of judging values.
Hill: Putting that at CoLab, when I think about how you iterate on what the verticals are every year [for Gig Tank], the way that you have taken additive manufacturing and the way that you [Mike] remind everyone constantly how it relates to the Gig and everything, you’re constantly nudging people — “it’s the right track” — so that the culture can keep on building on itself, right?
Hill: The fact that you do seasonal things, I think is really awesome and helpful to your momentum. I’m jealous of that from time to time. Because we have this “We’re here every day, we’re the library” thing. Your seasonal thing really helps you.
Bradshaw: Cyclicity is really important in a certain way in setting these things up. It’s a natural phenomenon as well.
Weinberger: That’s super interesting.
Pulse: Mike, what marked the phase shift in Chattanooga’s startup culture?
Bradshaw: You can trace back historical stuff: this happened, this happened, this happened. But I don’t know if you could have — I’m certain you could not have — been at the beginning and said “Now we’ll do this step and we’ll do this step.”
[To Hill] Actually I’m surprised to hear I ever asked you what your plan was, because Tia [CoLab communications director Tia Capps, who was observing] would be amazed to hear that because she knows I never do that.
Hill: I get that from people all the time.
Bradshaw: I do that too. I put a plan out there and basically I’m setting attractors out and then nudging towards them.
Bailey: Was there a marker for the phase shift?
Bradshaw: No, and I think by definition there isn’t [a marker]. What happens is you kind of look at it and go, “There it is.”
Weinberger: I haven’t researched in this area, but it seems to me, mainly from personal experience, that there often is a moment looking backwards that stands for the phase shift. And, clearly, that’s totally artificial but also helpful. “Do you remember when we had the very first Startup Week? That was when everything changed.” And of course you know that’s not really…
Hill: Maker Day. [A 2013 expo of 3D printers on the Fourth Floor]
Bradshaw: Maker day was a big one in a further phase shift. And I describe [Maker Faire] this coming weekend as being the third number in a sequence. But, of course, even though I said it would be great if we did this, this and this — no idea that was going to happen. Then things move along and we’ll see.
In the case of the startup community, I think the precipitating factor was Sheldon Grizzle and the CoLab. The fact that it was able to come into the context of Chattanooga four or five years ago and nucleate a lot of activities.
Bailey: Because at some point it shifted from being a project [of CreateHere], even one with a lot of substance…
Bradshaw: …into a phenomenon. And at the same time you had the incredible power of the Lamp Post Group that injected into it and stabilized that structure. Then you had Benwood and the foundations that funded all of this, adding what I look at as the ambient energy: primary funding and the credibility that “Hey, this should be taken seriously, this is a valid experiment, we need to let it go.” And they hadn’t gotten to the point yet where we need to measure it. They just let that happen.
I would say the emergence of Lamp Post Group and CoLab was at that point in time where things emerged into a new structure.
Bailey: How did that work? Because they’re officially separate.
Bradshaw: They’re officially separate, but I think you would look at them as a synergistic relationship more than anything else. Lamp Post was willing to step forward with funds to do this thing. At the end of the day, they decided to collaborate rather than compete. I’m looking at it historically on the outside looking in. I started volunteering at the organization shortly thereafter, but that’s what it looked like to me once I was inside looking back — that these guys decided to work with each other rather than stake out their territories, and that was immensely helpful.
Bailey: [To Weinberger] Which is very typical of Chattanooga.
Bradshaw: [To Weinberger] We have “the Chattanooga way” here. That’s what they call it. And in a way it’s almost an open source model.
Bailey: [To Weinberger] It’s almost frustrating, because people here know it well and say “Oh yeah, okay, the Chattanooga way.” its like a piece of jargon [that makes it easier to communicate because you don’t have to explain a big concept,] but sometimes stops you from thinking. To me it really needs to be examined. I’ve seen a lot of it in action, but I want to understand it better.
Nate Hill: I think it’s going to be really interesting to watch because the city is growing. Part of the Chattanooga way is interoperability and my ability to be like “Mike, what’s up. I’m working on this thing. You do this thing. You know this other guy who does something like this thing.”
We’re very good at connecting to each other in this way. How are we going to keep that real as it scales up?
If you use Twitter as the example, one of the things people like to say — I haven’t read this book about the history of Twitter that somebody just put out that’s apparently fantastic — with all these social networks, they’re awesome when you first start and there’s like a few hundred people, and they’re your people, and you got your geeks on there and it’s your crew. And I feel like we’re kind of in that phase as a community now. Once Chattanooga triples in size, what is the Chattanooga way going to look like?
I really don’t know because I agree, I think it is special. I’ve worked in some much bigger places. It’s wild to see the interoperability. The lines are very blurred, right? So that we can all work together and get things done for this city.
And I don’t know what this has to do with interviewing you any more. I’m just going on.
Bailey: This is exactly what I wanted to happen.
Bradshaw: I think that’s completely parallel to David’s thesis.
Hill: That cellular automaton stuff, I love that. I can’t remember the name of the game…
Weinberger: The Game of Life. [An artificial life simulation Weinberger showed in his talk.] More here.
Bradshaw: I think the basic thing there is that from these simple rules the thing possesses, these elaborated non-periodic patterns come out.
Weinberger: That’s basic emergence, chaos theory, how termites make their homes.
Bradshaw: Yeah, it’s how all this stuff gets made outside of our built environment. I think it’s the most applicable model. We just don’t have the metaphors to work with yet, I don’t think. That’s part of that explaining process. It takes so many words to talk about what you’re doing and what happened here versus “Here’s the spreadsheet, it shows the results.”
Weinberger: One of the ways you know the phase shift is occurring is that you begin fewer sentences with “What’s really important here is…” People get the value system and the premise and they’re on board. There’s a set of people that are ready to hear. As they make the transition to the new idea, it’s inevitable that you’re spending a lot of time dealing with near misunderstandings. You just need to nudge the person who is somebody likely to get it. They’re really close. They’re not an antagonist, but they keep wanting to misunderstand it in some crucial way and you just have to keep saying, “No, that’s right but what’s really important here is…”
Bradshaw: And you know you’re getting somewhere — you’ve gone through that phase shift — when you don’t have to have that preamble anymore.
Weinberger: Among your cohort, you don’t.
Bradshaw: Part of what I had to do at CoLab is effectively engage these entities who operate by an entirely different dynamic than the one that we’re talking about here and still create an API [application programming interface, a tool that allows programmers to work with a piece of software] that could handle the non-Newtonian physics of somebody who actually operated by that principal by necessity. Because [organizations like funders and governments] had to stay organized, they had to be hierarchical in a certain way. They couldn’t exist without it. I think a university is like that in a lot of ways.
Weinberger: Government is one of the hardest entities to change, one of the last to be affected by this. You can see this not only in the fact that most have not, but where there is change, it’s positioned as peripheral. That it’s, “Don’t worry. We’re still doing the same old things” — we actually sort of wish they weren’t — “but here’s a web site you can try.” And it’s the easiest and most incidental stuff that can be managed. Getting to the core, which in many cases is where we really want the reform, is going to take a long time. Governments are very proud of the peripheral experiments they are doing but very reluctant — for good and for bad reasons — to let this become core to how government works, so that we can have a network relationship to our governments, which is what many of us want.
Hierarchies and networks are different dimensions. It’s very difficult for hierarchical organizations to participate in networks. I mean this really concretely. Hierarchical corporations have found it a tremendous challenge figure out how they can participate in the open Internet and have made more mistakes than — what’s the opposite?
Bradshaw: [Trying to build networks working with hierarchical organizations,] what we found — I’m running into this now, and no ding on the organizations — is that the incentive systems of the hierarchy, which is what really controls the motion of their members, doesn’t reward them. And so that network takes an inordinate amount of energy to sustain.
Weinberger: That’s a really important point. Just look at the PR messes when companies try to engage on the Internet. Rarely do they get it right. It’s unclear exactly what “right” means, but they put a PR person on [social media] who only says positive things. They put up a site that’s incredibly boring and nobody pays attention to it. We go to where people are actually talking like people to us. People we can trust generally are not the people who are in the business because those people’s incentives are not the same as ours. You get many wonderful, hilarious examples of companies trying to be hip on the net.
Internally, the incentive thing is very crucial. It’s money, it’s power. That’s true in lots of places, including universities.
Hill: Help me fit the emergence of all the sharing economy applications into this larger platform idea.
I’m thinking of Uber and Airbnb. We have Campus Bellhops here. I’m just thinking out loud. I don’t know where I’m going with this. It seems like [with] something like Airbnb it’s really hard for them to have an open API or to encourage some kind of emergent behavior in something that’s essentially already breaking laws and codes. Or rethinking those, pushing the boundaries of those.
How is there room for a sharing economy application to iterate and become agile like with Reddit or Twitter, these things that start off as one thing, but you had no idea was going to be used for. Is there a difference between those kinds of applications and leveraging the crowd’s resources?
Weinberger: Airbnb… I can’t remember his name, the founder, the kid. I heard him speak this summer, and he’s fantastic. He has it so contextualized as one piece of a larger shift, a paradigm shift in the economy towards sharing.
You don’t really want them to put up a platform that lets me find out where people I don’t know are going to be. That’s just not good, right? And in the same way, the library doesn’t want to let who’s reading what become publicly available, unless the person opts in, of course.
So there are real limitations that serve the needs and interests of the user. And Airbnb has a particular set of them. But one can imagine if they were to think through the information they have, the metadata about trips, you have to think very carefully about making sure the stuff can’t be re-identified. But first of all there’s a set of research data that would be absolutely fascinating: to see how people are traveling. Maybe that can’t be anonymized enough, I don’t know, but if it can that’s a good research [tool] to learn lot of things, including maybe some things that would be useful for epidemiology.
And second of all, with another set of meta data, one can imagine users, other people building applications that could be useful to people who are using Airbnb. Maybe Uber wants to know if people are staying and they can make an offer. Maybe neighbors would like to welcome in people who are coming from particular parts of the world.
Hill: These are fantastic ideas.
Weinberger: You don’t know what people are going to come up with. Maybe nothing. Right now, the hotels for decades have been on purpose very isolating and alienating. You’re kept apart from every other hotel guest. Many of us are quite comfortable with that. I know I am. But I can imagine some people who would like to have more of a social network around a hotel and maybe around an Airbnb.
Bradshaw: Talk about transference of a model. I haven’t verified this. I think I read this morning — it might have been last night — that they just crowdfunded in Egypt the new Suez Canal. They couldn’t put it together to come up with the government funding for it. They said in just a couple of weeks they raised the funds. I thought “Wow, isn’t that something.” This thing created to help little startups move along has just like completely… It’s not April first. It might not be a joke. [It has, indeed, been reported.]
Weinberger: If that’s true you couldn’t come up with a better example of that movement. That’s amazing.