I originally wrote this email back at the end of April, 2013 as an introduction to someone who has since become a friend.  However, it is one of the more detailed written descriptions of my thinking on a number of issues and maps out the history of how I stumbled onto those insights / opinions.


 I’ll start with the basics of background, history, interest etc…

I grew up as an Army brat and lived in Europe, Asia and all over the US as a kid.  By the time I was 18, I had come to the conclusion that the world was largely dysfunctional, but that this dysfunction was not the result of “bad people” or anything like that but rather stemmed primarily from institutional structures that gave individuals perverse incentives to behave in ways that didn’t serve their communities well.  Disruptive behaviors were a consequence, not the cause.


For the last 15 years or so, I’ve studied a wide variety of disciplines with an eye toward how we might be able to restructure institutions of all types in ways that would improve outcomes.  For much of my 20’s, I focused on institutions of governance and of communication.  However, I had found not much more than dissatisfaction with the “best solutions we can muster” — i.e. national representative democracies with capitalist economies, leavened by some socialist style institutions of education, law enforcement etc… occasionally going to war with one another because the international arbitration/mediation bodies had no real enforcement powers etc….. (you are familiar with all of this, so I’ll stop here).

Prior to law school, I set up a “sweatshop-free” clothing company and became interested in setting up a standards organization for fair-trade certification of textiles.  The business was not a success — probably because a) I hadn’t yet learned how to delegate well and b) I’ve always had a significant amount of disdain for “fashion.”  Once I realized that others were starting to make headway on the “sweatshop free” stuff, I lost all interest in continuing to spend my time trying to convince people to buy a shirt.

The interesting thing that did come out of that experience was my grappling with this question: if I were to set up a “Fair Trade” certification standard, what set of policies / behaviors would that “certify?”  Would fair trade mean “workers make enough money to pay for food and transportation? Medical? For themselves? For a family of 3? For a family of 5? What about in contexts where the state took care of medical expenses? What age would we consider “too young” to work? 17? 15? 13?

Customs surrounding those practices vary from place to place.  In order to clearly communicate with the person making a purchasing decision in the clothing boutique, we would need to have a clear standard for many, if not all, of these areas.  The issue is that each of the decisions that we make for what “makes it into the standard” and what does not, will map to the values of one purchaser well, but to the values of another purchaser poorly.  A simple example would be the “minimum working age” issue — while Jane might be comfortable with someone from African country “X” to begin working at 13 because she understands that school in that country ends at 12 and if they are unable to work, there is really no other option for them at that point, Mary might have a more rigid belief about work only being done by people aged 16 and up — maybe she thinks that if we insist upon such a structure, it will place pressure on those countries to continue educating their children longer.  Both beliefs are acceptable.  The selection of a single standard makes it so that at least one of them is not able to get the relevant information “for their own decision making” from that certification.

I applied to law school with an application essay that basically stated that I didn’t intend to become a lawyer but instead intended to start and grow organizations (whether for-profit or non-profit) that solved problems and sustained themselves on that basis (the term social-entrepreneurship was starting to be used by this point, but I hadn’t yet encountered it).  My belief was that if I wanted to succeed in re-structuring institutions, I had better understand the existing framework as well as I could and should probably accumulate some bona-fides on the way (silly as it seems to me, the fact remains that other people put stock in prestigious educational brand-names — so I figured, I might as well pick one up and try to explore as much interesting stuff as I could while I was there).

While at Berkeley, I dabbled with business school courses and projects, mostly surrounding social entrepreneurship and the issues faced by the folks trying to figure out how to do “triple-bottom-line” type accounting comparisons across firms — I found that question interesting in the context of the “we don’t agree about the relative value of different outcomes” conundrum that I had encountered when exploring Fair Trade Certification. I also started focusing on information systems and took a number of courses at the “i-school.”
During my time at Berkeley, there many minor “aha” moments, but four major ones stood out.


1) Culture trumps Law

2) Privacy, Taboos and a vibrant “marketplace of ideas.”

3) Violence and Ostracism as Regulatory Tools over time (the history of — and factors affecting — the functioning of Reputation Economies)

4) An information Architecture that could enable a reputation economy to function at global scale


While writing the final exam for my Constitutional Law Final (with my favorite professor, the brilliant Goodwin Liu) it struck me that no matter what you wrote down on paper — whether in your corporate mission statement, a state statute, your national constitution or the UN Charter of Human Rights — you would only get, in-practice, on the ground, the outcomes that individuals demanded.  If a cop enforces the law more rigidly with certain ethnicities than others, your equal protection norm is, in practice, unevenly applied — and will continue to be so unless someone insists upon something different (and does so with enough force that it “coerces” the actors involved to change.)  Another example that springs to mind has to do with corruption in different countries — in the form of government officials asking for bribes.  In places like Indonesia (where I lived for 6 months a couple of years back) there are laws on the books against the giving and receiving of bribes.  However, everyone in that society knows that those government officials are not paid very well and need to make up the difference by “exacting tribute.”  As a result, though it is a begrudged bribe, it is one that people are willing to put up with.  The cultural norm is one of “bribe if you have to, but try to get a good deal.”  In the US, and many other western industrialized nations, the cultural expectation is that giving a bribe and receiving a bribe are taboo.  If caught, not only will there be official punishments etc…, but more importantly, there will be shame.  Bribery is one of the areas in our culture where the threat of shame plays a major role in deterring a specific behavior.

The other part of this realization was that you cannot legislate “wisdom.”  Again, no matter what you write down on paper, if your people want to go drive themselves off of a cliff, they will.  This explains, in part, why the “drag and drop” of text from a western constitution into some new country’s constitution doesn’t have immediate impact.  It takes time for people to decide what they want the system to really be and old cultural norms and expectations concerning “how to get things done” will trump, at least in the interim.  That said, laws and other written codes can serve as tools to help build a consensus over time about what is appropriate and how to behave.  However, it is just a tool for helping bring about a cultural consensus.  If that agreement does not already exist, it will take time for it to emerge (if it does so at all).  The community (or portions of it) might determine that such a norm isn’t all that useful for them and will continue to disregard it.

While in a class with a lecturer that worked as an attorney for EFF, I opted to write a paper about privacy issues with regard to location information.  This was before phones widely had GPS on them, but you could see it coming and we already had RFID tags for toll-passes and the ability to triangulate the location of a cell phone via cell towers.  In diving into that specific topic, I came to the realization that in the US, we don’t have a coherent theoretical framework for why privacy is important.  We have widespread agreement that it is important.  We just don’t know why.  I set about using this paper as an excuse to map out such a theoretical underpinning.  The basic gist of it is this:

Communities exist within some set of circumstances.  Over time, those circumstances change (whether through changes to the environment, new neighboring threats, the success of their previous efforts resulting in population expansion, the creation of new technologies etc…).  As circumstances change, the set of social norms/policies etc… that were well suited to their previous context might become less well suited to their new context.  As a result, pressure will emerge to shift away from old norms that are no longer serving them well and to new norms that better fit the new context.  In this context of shifting contexts and changing norms, John Stuart Mill’s concept of the marketplace of ideas is at play.  Mill argued in “On Liberty” on behalf of freedom of speech, in part so that ideas could “battle it out” in the marketplace, with the idea that the cream would rise to the top.  What I came to realize was that in that “marketplace of ideas” you have some ideas that might be well suited for the “new context” that happen to be taboo currently (presumably because they were poorly suited to the “previous context”).  Privacy allows behaviors to be kept secret and thus enables people to explore taboo subject matter with some protection against suffering the social consequences of violating that taboo — in short, if nobody finds out about, no consequences will result.  If the experimentation and exploration that goes on within these private spheres begins to result in seemingly useful ideas, over time those ideas will begin to creep out of the private sphere — or more accurately, will initially be published more broadly within private spheres until they eventually start to be explored openly in the public sphere.

An example: If our location is being published at all times and recorded, stored and made available for anyone else to view, we are less likely to attend that union meeting, that communist party meeting etc…  Privacy with regard to location information thus gives us the “freedom” to explore taboo subject matter with less worry of getting fired, blacklisted etc…  Obviously, this same logic applies to other ideas as well, whether they have to do with the use of recreational drugs, sex before marriage, homosexuality, exploring a political philosophy, organizing a coup etc…  Technology, including technology that enabled more intrusive recording of audio and video, distribution of that material, search and synthesis of that material all has made “private property” a less robust vehicle for the creation of “privacy.”

In addition, without privacy, the ability to exploit ones own innovations becomes hampered — trade secret enables you to come up with a “formula for success” and to use it to your advantage.  Without privacy, that advantage would disappear almost immediately and thus the incentive for investing effort in the innovation in the first place is eroded.

So to summarize, privacy creates value in part by allowing individuals to share more freely, because they are able to share in a limited context and it enables people to experiment more broadly, because they are (literally or figuratively) “behind closed doors.”  The broader innovation and experimentation that results from a well-functioning private sphere results in a more vibrant “marketplace of ideas” and the community ends up being able to better navigate and adapt to the changing circumstances that they face.


About midway through law school, I was studying “Chinese Legal Institutions” and learned about the governance structure in Chinese villages roughly 2000 years ago.  Local magistrates passed tests of culture etc… to win appointment as “the emperor’s man in town,” a position that I envisioned as being roughly a mayor/judge type of role.  Among their other duties was that of mediating disputes among citizens.  If Lau gave Chen a bag of rice today in exchange for a chicken tomorrow, but Chen ended up not delivering the chicken, Lau could request to have the magistrate mediate the dispute.  The magistrate would try to get the two sides to come to a resolution.  However, if Chen insisted on being stubborn and they were unable to come to an agreement, eventually the magistrate would excuse them without having imposed a judgement upon them about the resolution of the dispute.

My reaction was: “How the hell would that work? Wouldn’t your society just fall apart with people screwing one another over constantly if there was no arbiter forcing them to keep their commitments?” And then, almost as quickly, the realization hit, “No. This was a small village. Everybody knows everybody.  Confuscian etiquette structures were particularly clear about the appropriate way to behave in specific social interactions.  If Chen gets stubborn and acts contrary to those standards of etiquette, there is a real risk of loss of face — of status.  Being unwilling to “be reasonable” when in front of one of the highest status members of the community would be to risk losing status broadly and if that happens you might not have people willing to do business with you any longer or find them unwilling to allow their daughter to date your son.

In short — the village had a reputation system that worked.  Reputation, and the threat of ostracism, played the dominant role in keeping society intact and functioning harmoniously.  The threat of violence, the other primary coercive tool that communities have had at their disposal for regulating behavior played a role as well, but it was a smaller role.

I quickly realized that this basic structure — widely agreed upon informal social norms, enforced via a functioning reputation system — had been the dominant regulatory regime not only in china, but in communities spanning the globe.  However, though it obviously did still play an important role, by the late 1990’s and early 2000’s it was obvious that violence had come to dominate the philosophical landscape as the primary tool that could be used to get people to “behave well.”  To summarize these theories briefly —
1) We don’t want roving gangs terrorizing our citizens
2) Consequently, we want the government to have a monopoly of violence
3) but in order to ensure that the government does not abuse this power (exercised through the police, courts, prisons etc…)
4) we try to check the government’s ability to use this power in ways that do not serve the citizens
5) democracy (and the accompanying institutions of freedom of speech, press, assembly etc… that foster it) are useful tools for keeping the regime in check, however

6) there is the threat of the majority in a democracy oppressing smaller groups (tyranny of the majority).

The selection of cultural norms in a reprasentative democracy largely consists of voting for representatives who will then hammer out the details of laws (norms) that will be enforced, either directly or indirectly, by the threat of violence.  We pay the parking ticket, because we know that refusal to do so, will eventually result in potential imprisonment.

This violence based enforcement system has significant defects ranging from the selection of social norms (including problems in the electoral and legislative areas) to enforcement of those norms.  The fact that we have to select a single approach means that for some, the law will run counter to their beliefs.

It seems obvious that reputation based systems faded over time as populations expanded and trade expanded.  Interacting with strangers more and more led to situations where one person could wrong another and then hide in the anonymity of future interactions with individuals that had no knowledge of that history.  However, beyond the problems that arose as a result of “larger communities” expanded trade also introduced problems associated with disagreements concerning what is appropriate and what is inappropriate behavior.  As pluralistic societies became more common, problems associated with disagreements about etiquette, norms etc… arose.

It was apparent at that point (2006) that the internet was allowing us to share information, including reputation information.  Communities like eBay were functioning relatively harmoniously and were doing so because of reputation information — and not fear of lawsuits — nobody is going to sue over a $20 pair of shoes.  However, it also seemed apparent that this reputation based regulation was only functioning well within communities where there was widespread agreement concerning which behaviors were appropriate.

And so the thought occurred — how could you enable that type of informal social norm structure, enforced by a reputation system but on a global scale, in a pluralistic society?  It took 8 months or so of exploring various parts of the space and diving into existing communities that included remix culture (creative commons and connexions from Rice University both got a fairly deep dive) but then, while listening to Obama’s second book (via a download from Audible.com) the solution sort of presented itself all at once.


In The Audacity of Hope, Obama talks at one point about the process that a legislator must go through in negotiating environmental legislation.  He talks about prioritizing your wants because you are going to have to make trade-offs and you would prefer to trade away low-priority items in exchange for accomplishing high-priority objectives.

I hit pause and thought to myself, could you accomplish environmental regulation via a functioning reputation economy?  What would that look like? How would you accomplish it.  If people can see that buying this loaf of bread would pollute the local river by this amount, but buying a different loaf wouldn’t pollute at all, they might “punish” bad actors directly rather than indirectly.  The thought occurred “but what if people don’t care.” Then the river will get polluted.  You will only get the outcomes that your people are willing to stand up for.  However, if I not only know that the river will get polluted, but that down the line other people might be able to discover that I contributed to that pollution (and knowingly so), I might decide to go with the non-polluting bread, not only because I care about the river, but also because I care about other people thinking that I care about the river.  On the other hand, if people don’t actually care about the river, it will get polluted.  And eventually, if they feel any sort of repercussion from that pollution, it will cause them pain and they will then start to care about the river.

Any community is faced with a wide variety of threats and opportunities and the ability to experience pain or benefit serves as a primary signalling function that helps them decide which problems to dedicate time, effort and resources toward solving. There is no reliable way to figure out these priorities before hand.  Hayek wrote about the signalling role that price plays in a market economy and about how that indicator enables the members of a community to self-coordinate without having any knowledge beyond price.  If copper prices go up, a manufacturer that relies on copper in their production process might switch to a substitute material.  This is true whether the price of copper went up due to a labor strike, a storm that disrupted delivery or a big new customer that is driving prices up as a result of their demand for the material.  If you build the information architecture appropriately, you can create a reputation economy that similarly allows the actors within a community to self-coordinate regarding policies and social norms without forcing them to come to a consensus on them.

Currently, I’m hosting a weekly meetup on Tuesdays at 7pm called Build the Collaborative Internet.  This weekly meetup is intended to serve as the “church” for those people that are passionate about bringing about this next generation version of the web.
I’ll start by sharing a little about my vision for how that system works and what it enables.  Prior to starting the meetup, I’d referred to my version of this system as FLTRD (pronounced “filtered”).
Here is some text from a company snapshot that I had written last year:====
The FLTRD Social Search Platform (FLTRD) will be a collaboratively constructed – yet custom filtered – database that gives each person on the planet easy access to information that is Relevant, Reliable and Readable for them.When I walk into a grocery store, I will be able to pull out my smart phone and scan the barcode on a bag of potato chips using my FLTRD application. Based on my user preferences, FLTRD will return a handful of items:1) a graph comparing these chips and the three most popular alternatives with info on price, calories per bag and the average “taste” rating.

2) a map displaying the the cheapest place to buy each of these 4 options in my neighborhood and

3) a warning that these chips might contain peanut particles.

If a friend decides to “follow” my subscriptions, but doesn’t care about peanut particles she can modify that portion of the subscription with a quick swipe and a tap. Next time she does a similar search, she benefits from all of my other subscriptions but doesn’t see the peanut warning. Alternatively, if she wants to alter the “taste” metric to count only the votes of people who are members of Facebook; or are female; or are between the ages of 23 and 35; or are members of “the Bay Area Food Buffs”; or any other way that she can dream up– with just a few taps, these results – and her future searches – will be FLTRD accordingly.

Because users are able to piggyback on one another’s efforts, small incremental contributions by individuals result in collaboratively built – yet customizable – reports. This enables even a first time user to benefit from comprehensive sets of expertly curated subscriptions while being empowered to select an alternate version or adjust any system component to better suit their own needs.

With minimal effort, users of every experience level will be presented with information that is relevant, from sources they trust, filtered any way they like and presented in the format of their choice, while allowing them to modify any aspect of the system to fit their own preferences. This is FLTRD’s key differentiator – its modular structure enables users to collaborate on a massive scale while simultaneously freeing them to customize those collaboratively created works in any way they see fit.

The FLTRD Social Search Platform will allow users to comment on any item – whether a real world object, an internet resource or a specific piece of content created by a FLTRD user. Once content (of any type) is created, users will be able to:

SHARE: publish with flexibility and control;

SUBSCRIBE: determine which sources (and other system components) they will rely upon;

REMIX: Add to, subtract from or simply modify any of these subscriptions and system components (in accordance with any underlying permissions and licenses) and

RE-SHARE: publish in any way that complies with “upstream content’s” permissions and licenses.

To put it another way: Take the pieces that you like. Ignore the stuff you don’t. Remix it any way you want. Repeat.


At present, I am working with the members of the collaborative internet meetup to build the various layers of this system. Some of the conclusions that we’ve come to have to do with guiding principles.  Others have to do with techniques for implementation.

At one recent evening we mapped out the principles that seem to be at the core of what we are working on.  Thanks to Jack Senechal for instigating and documenting this:

Core principles of the distributed cloud ecosystem
– Massively distributed data & functionality
– Unix design principles – atomic units of functionality
– Data interoperability
– Forking
– Versioning
– Unfettered duplication
– Content addressability
– Replaying agent actions / pathways
– Spheres of data – be as open or closed as you want to be
– Community self-regulation & curation
– Eventual consistency

Things we need to build
– Forking / collaboration
– Visualization / navigation
– Trust networks & other filters
– Underlying data graph (the matrix…)
– Annotation

Finally, at the moment, it is looking like we have a strategy for moving forward and getting various actors on board.  I will be attending a conference this week that is focused on companies in the Sharing Economy space.  These companies’ business models already depend on having functional reputation systems in place.  I am looking to start coordinating between them and an organization like Mozilla to have users be able to store their own data in their own “personal clouds,” likely at first via Mozilla’s “persona” identity system.  The play here is to get the AirBNB’s, ZipCar’s and other “sharing economy companies” to recognize that if they enable their users to “sign” and “export” their information to the cloud, even if they don’t get anything back, the exporting company benefits.Example:
If an AirBNB user exports all of the comments that they have made and all of the comments that others have made about them to their own “personal cloud server” and are later able to share that feedback (perhaps a review they gave of someone that left the dishes in the sink rather than washing them) with the members of another community like ZipCar (who might care about the tidiness and how considerate community members are) the new community benefits by having a broader pool of reputation information that they might be able to filter on the basis of.However, more importantly, the organization that opted to allow their users to publish reputation information beyond the boundaries of that community (here, AirBNB) also benefits because the fact that that reputation information might get used by other users and applications outside of their own “micro-community” gives the feedback of users within that community so much more power.

If the feedback that you receive here has the potential to follow you through all future interactions and to be sliced, diced and synthesized so that people get a useful and actionable snapshot of who you are and how you might be expected to behave — that creates a real incentive for individual users to “behave well” during their interactions within this micro-community (again, AirBNB).

This is just step one — getting the reputation information exported from the developer’s servers and onto servers controlled by users.  But that step is an important one in enabling a broader reputation eco-system to emerge.

I’ll leave you with two last — and I believe — important points:

The people who will be using these systems won’t be overly conscious of the fact that they are using a regulatory system.  They will simply be pursuing some end, whether that is finding cheap gas, avoiding clothing that is made with child labor or showing gratitude for great service by leaving positive feedback.  However, despite the fact that “regulation” might not be their goal, it will be the result.  These actors, in pursuing their own ends, will enable the “invisible hand” as Adam Smith referred to it, to function oh-so-much more effectively.

Finally, The critical component is to enable people to follow pathways that are relevant to them.  This will likely include not only relying upon information that has been created by others, but reinterpreting that information in some way.  This remix capability is a critical component, for without it you end up with giving users information overload, rather than custom synthesized information.  It will result in competing “versions of the truth” within the system.  But so long as people are able to route to sources that they find trustworthy, the system will work for them.

There is much more detail surrounding all of this that I could dive into, but I think you probably have enough to chew on here for now.

I look forward to exploring more of these and related topics soon.


-Matthew Schutte