Non-Profit Ideas

by Philip Greenspun in October 2007

Site Home : Non-Profit


As a companion to my pages on early retirement, this document presents some ideas for useful and fun things to do with one's money.

Simple: Donate money to Wikipedia

If you're too lazy to start a new non-profit organization, donate money to Wikimedia Foundation, the folks who publish Wikipedia. In 2007, Wikipedia was the ninth most popular Web site in the world and probably accounts for a significant portion of the world's self-education. What's the budget for this cornucopia of knowledge, provided free to the world in hundreds of different languages? About $1 million for 2007. If you gave $1 million to a big university, you would fund their operations for a few hours at most. By contrast a $1000 donation to Wikipedia will significantly advance their operations and, incidentally, light up your name on their benefactors page.

Internet Content Prizes

Suppose that Joe Generous writes the greatest English-language novel ever and posts it on his Web site for anyone in the world to read at any time of day or night. Joe is losing out on some expected income from his writing, but perhaps he has a lucrative day job and values the increased audience size more than the money that commercial publication would bring (recall that the poet Wallace Stevens went to work every day as an insurance executive). Could Joe be honored with a Pulitzer Prize for his work? No. A Pulitzer Prize can only be awarded to works published in hardcopy with a commercial profit-seeking motive. Could Joe be honored with a Booker Prize for his work? No. A Booker Prize can only be awarded to authors who have tried to make money from every person who seeks access to their work.

Most writers get paid minimally by publishers. It is rare for an academic author to earn more than $10,000 from a book, despite all of the hard profit-seeking effort by many people in the publication and distribution chain. Why shouldn't it be possible for an author who publishes his or her work freely to earn more than $10,000, plus honor and distinction, by winning a prize?

Here are the categories for a set of annual prizes:

How much should each prize be? To ensure that this is the most prestigious and sought-after award in the English-speaking Internet world, the prizes need to be larger than any other cash prize (right now there is no competition in this area) and larger than what the author could expect from a commercial publisher (not larger than the maximum, just larger than the average). It seems safe to assume an average of $20,000 per prize, with short works getting smaller prizes and long works, such as novels, earning prizes perhaps up to $50,000.

How are winners determined? In the late 1990s, a group of friends and I set up a similar prize system for people under 18 years of age who had developed interesting non-commercial Web sites or services. We set up a server to let anyone nominate a site and to let the public vote on nominees (one day of programming). We sat down for one evening and sorted through the nominees in descending order of average public rating and picked our favorites. We had one organizer who invited all of the awardees to a public lecture in Boston, arranging accommodation and airline travel, then kept them with us for a few weeks while we taught them a course on building Internet applications before sending them home with a Unix shell account on a machine with a relational database management system. This was to enable them to build more ambitious services in the future. The cost and effort involved was fairly modest.

The cost of running ? Assume 20 prizes will be awarded annually (not every prize will be awarded every year). We assumed an average of $20,000 in cash per prize or $400,000 for prize money. Assume $400,000 to run an awards event with associated publicity and approximately $1 million annually for continued PR and administration. The cost of substantially increasing the incentives for Internet authors thus comes to $2 million per year. To earn $2 million in investment income, after inflation, annually, it would be necessary to endow this prize with $50 million.

Teaching High School Math and Science in the Context of Engineering

Folks love to argue about whether or not evolution should be taught in public high schools. Left out of both sides of the debate is the fact that the average American high school student does not learn or remember any of the math or science that is taught.

Why don't students learn math or science? For the same reason that most of us adults don't learn math or science: we don't need it to get our projects done.

Suppose that we told a student entering high school "You are going to design a bicycle, build it, and ride it away on the day that you graduate." Algebra, geometry, calculus, physics, and chemistry can all be studied in the context of accomplishing this engineering project. Chemistry, for example, is used to figure out what materials would be appropriate for making different parts of the bike. Biology would be tough to fit in, unless we want to consider what happens when the rider falls off the bike...

This could be done relatively cheaply as an open content curriculum development effort: pay people to write textbooks, homework assignments, and lesson plans that can be freely used by interested high schools. A more expensive option would be to open a physical school that would use the materials.

[The author of this article has been a volunteer tutor at the most expensively funded public high school in the Commonwealth of Massachusetts (nearly $25,000 per year of taxpayer funds per student). Students in 11th Grade "Physics" were unable to "locate the point in the x-y plane where x=0 and y=0", use a graphing calculator, or grasp the idea that an equation corresponded to a set of points in the plane that satisfied the equation. In a small attempt to inspire a handful of kids to hit the books, the author created an annual aviation prize program.]

Internet Classical Music Free Library

Go down to your local symphony hall or opera house, a building that very likely has recently benefitted from a $50-100 million renovation. The audience will be so old that you wonder why they didn't add a Medevac helipad on the roof. That seniors suffering from cardiac arrest aren't hauled out at every intermission is a hard-to-solve mystery.

Classical music organizations bemoan the graying of their audiences and try various gimmicks to attract new listeners. They will pay headline conductors in the neighborhood of $2 million per year to show up for a handful of performances. They will indulge in expensive building campaigns. They will experiment with selling tickets for less than $100/seat. They've tried everything except actually delivering music to young people.

Where are the young people? On the Internet. Do they buy CDs? No. Why not? First, their apartments are too small for the clumsy "music in a physical package" idea that Edison developed in the 19th Century. Second, more sophisticated industries have collected 100 percent of their entertainment dollars. After paying their cable TV, Internet, Netflix, and XM Radio bills, the average young consumer doesn't have any money left over.

Classical music is good enough to sell itself, but it hasn't been packaged very well for distribution over the Internet. A classical work might be broken up into 32 individual tracks. Who wants to download, organize, and name these tracks? How can you guarantee that they will play in the correct order after transferring to an MP3 CD to play in your car?

For a tiny fraction of the cost of rebuilding a concert hall in a part of the city where nobody under 50 has been seen for 20 years, the entire classical repertoire could be placed online under a Creative Commons license.

How big is the core classical library? With 10,000 albums it would be almost certain that a requested work was available. How much would it cost to license 10,000 recordings for free Internet distribution? Most of the classical repertoire has been recording by smaller labels in smaller countries. Most of these recordings are on analog tape sitting in vaults. They were distributed on LP record and are unlikely ever to be demanded in sufficient quantities to make distribution in a physical medium profitable (the artists of the 1960s and 1970s might have been talented, but they are not known by today's consumers). Probably an average of $500 per recording would suffice, resulting in a total cost of $5 million to establish the library, but the licensing costs might fall as the site became more popular. Because the site would have actual music, unlike any other classical music site in the world, it would be the most popular place on the Internet for classical music lovers. An organization such as the Cleveland Orchestra might decide that they wanted to promote themselves by having their recording of a particular piece be the canonical recording. It seems likely that eventually performance groups eager for ticket sales might pay to have their recordings in the library.

What about operating costs? Bandwidth and servers are getting cheaper every day, and peer-to-peer distribution software can be deployed to good effect here, but it still seems worth setting aside at least $1 million per year for operating costs. To earn $1 million in investment income, after inflation, annually, it would be necessary to put aside $25 million.

Total cost to start and run in perpetuity a free Internet library of classical music would be $30 million.

English-language School in Peru's Urubamba Valley

Peru is one of the world's most beautiful countries (see my photos) and the Urubamba Valley is one of the most pleasant places to live within Peru. There are quite a few poor kids in the region, however, and the educational opportunities are not comparable to what is available in Lima.

We start a tuition-free English-language school in one of the pleasant towns of the Urubamba Valley, surrounded by Inca ruins. We admit the smartest 20 children who apply and provide them with a free education starting at age 7 or so. All classes will be in the English language. If a graduate of our school gets into a prestigious Peruvian high school or a university, we pay for that.

The cost? A good private school teacher in Peru earns about $5,000 per year; let's have two. A comfortable two-room school house should cost no more than $12,000 per year to rent and heat. Budget another $5,000 per year for computers, Internet, and other supplies. Budget another $10,000 per year to rent a nice house in which visiting tutors from the U.S. (us and our friends) can stay. Out total budget is approximately $40,000 per year or, per pupil, more than double the average private school tuition in rural Peru. If 10 of the students end up in Lima at a fancy boarding school, we're on the hook for roughly $100,000 per year in fees, depending on scholarships. A similar amount should suffice to support our graduates through university.

Who benefits? Mostly us. This gives us (and our friends) an excuse to fly down to Cusco, hop a taxi or a train for 1.5 hours to the Valley, and hang out with smart motivated kids for a month at a time.

Total cost? $40,000 per year for 6 years, rising to $140,000 per year for years 7-15.

Computer Science University in Africa

At the request of the United States Agency for International Development, in 2007 I visited computer science departments at universities in Tanzania, South Africa, and Uganda. There is a tremendous enthusiasm for learning among the students and tremendous demand for computer science education. Unfortunately, the universities there have uncritically imported the traditional American and European university ways of operating.

The organizations that function best in Africa are those that have adopted the best First World quality control techniques, including the dreaded ISO 9000. What quality control techniques would an African university have if it copied the best American university? None. Consider an incompetent and lazy teacher at Harvard University, our most prestigious institution. He or she wants to spend maximum time on career-enhancing research, minimum time on career-killing teaching, and is not good at teaching in any case. The Harvard professor teaches his students nothing and, to suppress complaints and disguise the fact, gives all of the students As and Bs. As long as each professor grades his or her own students, there can never be any control of quality. I asked the African professors "How come more people fail the Microsoft Certified Systems Engineer (MCSE) exam than your most advanced and supposedly difficult computer science classes?" The answer is that that MCSE exam is administered by an impartial authority.

See my notes from a talk to a CS department for a general assault on the utility of giving lectures to engineering students.

The good news is that teaching a handful of Africans to the highest standards would have enormous implications for their economies. There is tremendous demand for capable software engineers (currently virtually all African organizations need to depend on software developed in First World countries or, increasingly, in India and China). Computer science is cheap and easy to teach since no laboratories need be built. It really isn't that hard. All that someone would need to do is tell the faculty "You are henceforth forbidden to give lectures and instead must spend your time walking around the lab helping students with projects."

Model: Olin College of Engineering (start up cost around $400 million, including deluxe real estate in the Boston area)

Latin American-style Towns for the U.S.

When computer nerds get rich, their charitable thoughts turn to helping Africans (see Bill Gates). They make a spreadsheet of the quantifiable aspects of the human condition, sort by misery, and the Africans come out on top. Peace Corps workers who return from a couple of years in small African villages tell a different story. They come home to their parents' materially magnificent suburban homes and immediately suffer from loneliness and depression. Maybe we should feel sorry for Americans who live in suburbs and need to get in the car to shop or work and need to make an appointment before there is any possibility of seeing a friend.

What are the problems with suburban living, the dominant mode of American life? A shallow problem is that the car is required to accomplish any task outside of the home. Suburbanites waste their lives, and a lot of energy, driving to the strip mall to shop, driving to their place of work, driving to see friends or to an entertainment venue. Suburbanites cause horrific traffic jams that turn their nightmarish 45-minute commute into a hellish 2-hour commute. Giving how spread out houses are in the suburbs, it is impossible for a business that depends on pedestrians or bicyclists to succeed. A strip might might support a coffee shop, but it won't be a place where people drop in as a casual part of their day. The more serious problems with the suburbs start with social isolation. You won't have a chance encounter with a friend when you drive point to point. A suburbanite in theory could make an appointment to see a friend, but this is tough to arrange when everyone works 8-9 hours per day plus commutes for another 1-2 hours. Zoning laws ensure that nobody can run a business, even one that is clean and quiet, from his or her home. Thus the typical suburban youth will never see an adult at work. As far as suburban teenagers are concerned, cash is something obtained mysteriously by adults and brought home after an exhausting commute.

Latin Americans often come up near the very top of the world's happiest people, despite a material prosperity that is very pale compared to that we enjoy in the United States. Nearly every small town in Latin America is built around a central plaza where the citizens gather at various hours to meet friends, play chess, eat meals in restaurants, etc. Small streets radiate from the plaza and hold all of the shops that are essential to daily life, including supermarkets and hardware stores. Housing is built up to a three story height, dense enough to support businesses, but not so dense that people are isolated in concrete towers with elevators. Smaller workshops are mixed in with housing, introducing young people to the texture of business.

The U.S. offers some enjoyable walkable neighborhoods, mostly developed before the rise of the automobile. Examples include many neighborhoods within New York City, San Francisco, Chicago, and Boston. These neighborhoods, however, are small and can hold only a tiny minority of Americans. Consequently, houses within walkable neighborhoods typically cost over $1 million. As the U.S. population heads toward 500 million, these livable neighborhoods will become even more out of reach of the average citizen.

The market economy will not deliver Latin American-style living. We have to assume that building tract houses along the Interstate, served by strip malls a few exits down the highway, is the most profitable way to develop real estate. The handful of "New Urbanism" communities are not substitutes for the Latin American town. At Disney's Celebration (near Orlando, Florida), for example, residents must drive more than 20 minutes to get to a supermarket, a hardware store, or a bookstore. It would be illegal to start a small business in most areas of Celebration.

In the exurbs of a rapidly growing metropolitan area, such as San Francisco or Los Angeles, we build a Latin American town, complete with central plaza ringed by three-story high buildings, the ground floor of which holds shops. We offer free rent to supermarkets, hardware stores, and other essential services. We encourage residents to start small non-industrial businesses in their homes, partly to provide jobs within the community and partly so that young people can see what adult work looks like. Once completed, the buildings are sold off for market prices and the money is recycled into building the next one.

How much would it cost? To achieve critical mass, we need to build housing for 10,000 people plus a bit of retail space. Let's assume 700 square feet of living space per resident plus another 300 square feet of retail or public space. That's 10 million square feet that we need to build, at an average cost of $125 per square foot, or $1.25 billion. We could probably get bank financing for 75 percent of the cost, which means that approximately $300 million in cash is required at the outset. Having started with the assumption that this is an economically suboptimal way to build, we predict that we will lose 10 percent on the final sale of the property, which means that we'll lose $125 million per community constructed. Thus the cost of making each person in this community happier than he or she would have been living in a suburban hell is $12,500 per person. A side effect of building these towns is to take some of the pressure off prices in the handful of desirable neighborhoods throughout the U.S. and also to reduce traffic congestion. Folks living in our new towns will walk or bike to accomplish their errands and, in some cases, get to work. Our higher density will be conducive to running public transportation from our central plaza to the downtown areas of the nearest big cities.


philg@mit.edu

Reader's Comments

I have a suggested alternative to building Latin-American-style (in your view) towns from scratch in the U.S.

It seems to me that the kinds of towns you are talking about are built organicaly, but a very large number of small developers, all making individual choices over a long period of time. In some sense, you simply *can't* reproduce that by fiat: the development is inherently cumulative, incremental, and deeply involved with the culture that dwells there. You risk creating a cargo-cult of a village if you try to plan it out as you've suggested.

Rather: look to the most impoverished areas and use your charitable largesse to aim for break-even investments in the development of small businesses owned and operated by the people there. People say things like "The Web destroyed mainstreet" but I think that's only the story *so far*. Historic retailers are (mostly were) irredeemably old fashioned in how they managed their supply chains, retail space utilitization, and more -- geeky entrepreneurs should be able to find a lot of new efficiencies, sufficient to support at least small local enterprises.

In that way, you can build a constituency of community leaders with whom to, next, turn to questions such as regional development policies.

-t

-- Thomas Lord, November 20, 2007

I like your ideas.

Wikipedia is amazing. The power of volunteer groups in general is surprising to me.

The "Latin American Village" would be hard, but it might work due to current demographic trends. The current baby bust and divorce is causing a lot of loneliness in the suburbs.

You would need a special zoning exception from the city, and any home owners association might quickly kill most of the in-home businesses. People do no like the idea of businesses springing up next door and bringing a daily traffic of strangers past their door.

Latin Americans lived the way they did out of necessity and because extended families were still economic in the hinterlands.

The ideal American life in the 1970's was like Bonanza, with Pop, Adam, Hoss, and Little Joe plus some women and children. The show had a sad tone to it without wives and children.

People in the 1970's were trying to recreate that family isolation in the suburbs, but I think that may shift again to the practice of the 1950's when there were no fences allowed between neighbors. That idea became fashionable among the intelligentsia during the baby bust before WWII.



-- David Lee, July 27, 2008

I like your idea of Internet content prizes, except that this would ultimately lead to less choice on our bookshelves, virtual or physical, as only one prize per category means only one author in a category of thousands of authors who actually gets paid. If all authors turned to this over night as a way to earn money, the volume of authors and with it the volume of published works would diminish over time.

I do think its high time the copyrighters recongnised that with Internet distribution, the price of the information (read that as author) should be the bulk of the price consumers should pay. I don't understand the cost model behind iTunes et al charging the same price for an online virtual CD as Best Buy does for the same instore physical CD.

Authors should get paid. That's my $0.02 worth I'll be collecting some day :)

-- Tim Fisher-Jeffes, July 30, 2008

re LA Villages - you have to solve the problem that in the USA, every street must be wide enough for 2 fire trucks to pass each other - if you don't you won't be able to insure any of the property.

Possibly engineering a different solution might add $10-20 million (as a guess, design places for pumper trucks to plug in to the underground pipe system nearby but not on that street) , but the real hurdle will be the building codes and the ability to get the insurance companies to write insurance on the affected properties.

-- Patrick Giagnocavo, February 18, 2013

Add a comment | Add a link