• Lyft, Sidecar, and Uber’s taxi service changed San Francisco for me. I found cabs on the street – which, c’mon, never used to happen in SF – but often didn’t bother, because ordering a Lyft/Sidecar/Uber Taxi was easier and sometimes cheaper.

• “What, we’re all okay with gypsy cabs now that they come from an app?” one irrate yellow-cab driver asked. And the answer, basically, is yes: cars from an app come with drivers with background checks and known bank accounts, and these startups know more about drivers’ and riders’ current locations than the yellow-cab company. The world in which I feel safer in a “Lyft gypsy cab” than in a regulated yellow cab doesn’t feel that far off.

• I’ve started to believe the leverage in the “sharing economy” will be in opening regulated industries. SF cabs were atrocious because there are too few medallions. (Turns out the medallion holders, keen to restrict medallion supply, were well-incented lobbyists, as any good monopolist should be.) The revolutionary part of Lyft and Sidecar is that those companies decided, forget the medallion battles! and let’s just increase the number of drivers on the road.

• I’d suspect, but don’t know, that each Lyft, Sidecare, or Uber Taxi driver earns less than a yellow-cab driver did for the same work because of the influx of drivers. Though in the short run, new demand might keep those rents afloat even as the monopoly cracks.

• Within two years, I bet the east part of the Mission will supplant SOMA as SF’s startup hub: it’s sunnier, gritty (or so; there’s also Central Kitchen and Blue Bottle), and young teams can get cool, warehouse-y spaces for less than an elevator’d building off Third Street. (Who likes those buildings anyway?)

• Dropbox is the whispered-about company of the moment – “can you believe they hired [this person]?”, “have you heard what my friend is working on there?” etc etc. – and it’s becoming the latest pool of very very good people in the tech industry. (Facebook used to be this company.)

• Building codes that discourage density are San Francisco’s greatest threat. There’s been much written about SF rent increases, which are a nuisance if you start or work at a well-funded startup but more dangerous if you’re just starting out: talented people are not always rich, and we’d all lose if they chose not to live in the Bay Area. (Though to be clear: New York’s residential areas aren’t dense enough, either.)

• Did it feel like a bubble? Mostly not. Some things felt off, and others unsustainable, but it also feels like we’re coming up on a golden age of software, as we realize we can and must understand computers now, and software moves from media toward other, huge sections of the global economy. It’s all exciting.

I joined USV in 2010, not expecting I’d work at a venture capital firm so soon after school. The last two years were amazing. USV’s partners – Albert, Andy, Brad, Fred, and John – are fantastic, the team with whom I worked – Gary, Dorsey, Nick, Brian, Zander, Veronica, and Gillian – unbeatable, and the entrepreneurs the firm backs brilliant. I thought very seriously — torturously, even (hi, friends!) — about staying at USV, and I’m extraordinarily grateful to Albert, Brad, and Fred for taking a chance on me.

Why did I want to do something different? In part, because I wanted something that felt more tangible. But mostly because the story of the internet continues to be the story of our time. I’m pretty sure that if you truly want to follow — or, better still, bend — that story’s arc, you should know how to write code.

I also wanted to learn to write software out of stubbornness. It felt that you’re not really supposed to do that after school, and such low expectations amazed me. When did we decide our time’s most important form of creation is off-limits? How many people haven’t learned to write software because they didn’t attend schools that offered those classes, or the classes were too intimidating, and then they were “too late”? How much better would the world be if those people had been able to build their ideas?

Programming wasn’t entirely new to me. I’d taken classes, and I’d worked with a few APIs while at USV. I didn’t talk about these projects often (the people with whom I met were solving fantastically harder problems), but it sometimes came up. Most didn’t take my coding prospects too seriously, which stung at first and then became annoying.

Admittedly, my need to learn to code wasn’t as obvious to most everyone else. Empirically, you don’t need to program to invest in software companies. Empirically, you also don’t need to understand software to invest in software companies. Venture capitalists have other tools to evaluate a new company beyond asking technical questions, and often, those other tools are used first. But evaluating companies isn’t what I wanted to do.

So writing software is what I’ve been doing. For the past few weeks, I’ve been working on a web app. Unsurprising to those who know me, it surrounds books. I’d like to think of it as a digital bookshelf, a conversation starter amongst friends … but it’s not there yet. It’s got a database though – which none of my side projects had – and it’s got caching, OAuth integration, a few queues for jobs that can be made asynchronous, and other things I’d never explored before. It’s also got some users, friends who put up with the site’s lumpiness and frequent ill behavior.

I’ve more thoughts about writing software, and learning to do so “later in life”, but one thing’s certain: though I’ve written about the tools of software development becoming more accessible, and the process of software-creation becoming simpler, it’s still really, really hard to build a good product. I don’t have one yet. I do have a changelog, a day-by-day record of what I’ve been through. Just like the site, it’s scattered, idiosyncratic, and lopsided. And while I don’t think I’ll spend much more time on it, it’s given me sea legs.

On stable ground, it’s easier to stop worrying about the future or the past and simply enjoy what I’m doing now: learning every day, testing and discarding my opinions, and making something that oh-man,-real-people-use!

• M-PESA is the world’s most widely-adopted mobile payments system. More than 50% of the mobile payments in the world happen over M-PESA. M-PESA operates in Kenya, Tanzania, South Africa, Afghanistan, and India, though almost all transactions still take place in Kenya, the first launch market. M-PESA was developed by Safaricom and Vodafone (Safaricom’s parent company). The IP is still held by Vodafone.
• M-PESA’s premise is simple: users conduct instant peer-to-peer money transfers, using phone numbers as identifiers. Cash enters and leaves the system through M-PESA agents or traditional ATMs.
• Safaricom introduced M-PESA in Kenya in 2007, initially to lower operating costs for microfinance institutions. During the pilot, loan recipients began using M-PESA more broadly, and the product was expanded. This history is documented in Money, Real Quick.
• Today, 70% of Kenyan adults have used M-PESA, which is greater penetration among Kenyan adults than Facebook has among US adults. The equivalent of 30% of Kenyan GDP flows through it each year.
• Safaricom, the incumbent monopoly, is by far the largest telco in Kenya. In rural areas, rival telcos struggle to provide any cell reception; Safaricom has coverage, data, and M-PESA agents.
• The network of human agents that broker M-PESA-to-cash exchanges is over 40,000 agents strong – meaning M-PESA has more physical points of presence than there are bank branches in Kenya across all banks.
• Other telecoms have replicated M-PESA’s technology, but they’ve yet to replicate its agent network. Load-balancing cash requirements seems to be the hardest piece. (Cash tends to flow from urban to rural areas.)
• M-PESA agents are everywhere. It feels like every fourth shop in Nairobi had an M-PESA sign. “M-PESA is like your shadow”, as the saying goes.
• It appears that the local entrepreneurs who double as M-PESA agents do so not for margin from the service but to drive foot traffic into their stores.
• Similarly, Safaricom makes razor-thin margin on M-PESA itself but makes up for it with improved customer retention.
• At Nairobi’s shopping centers, the ATMs that dispense or transfer money from M-PESA had the longest lines.
• M-PESA adoption has been strongest where its alternatives are worst. Bank transfers and checks are practically unavailable in Kenya. Before M-PESA, internal remittances were sent on a bus with a human courier and utility bills were paid in cash to a teller. Now, internal remittances and larger recurring charges are two of M-PESA’s dominant uses. M-PESA offers a better version of both checks and bank transfers.
• M-PESA is used less often for local transactions – a restaurant bill, a product at the market, transportation, small services like shoe shining – because cash works sufficiently well most of the time. Its flat fee per transaction model makes smaller transactions disproportionately expensive, which disproportionately affects its cash-constrained users. In current form, M-PESA is not a better version of cash.
• M-PESA doesn’t yet offer a product that grants credit, though one is purportedly imminent. Banks and microfinance institutions offer loans, but credit cards are rare, and credit card acceptance rarer still. The internet offers no salve; many ecommerce sites based in the US or Europe reject outright credit cards issued in Kenya for fraud risk.
• The Kenyan Government announced last week that it planned to tax cash withdrawals from M-PESA at 10%. It’s probably ill-conceived: it’s a consumption tax that hits the most price-sensitive consumers, and may become an example of “the tax that shifts consumption so far it destroys more revenue than it creates.”
• M-PESA’s IP is held by Vodafone in London, though it was largely developed by IBM in Germany. People who develop software atop M-PESA say that technology’s stilted, the service is erratic, and they’ve little hope for its improvement. Still, while development can be slow and brittle, success means access to the world’s most widely-adopted mobile money system.
• M-PESA slipped past financial regulators and banks, who seemed to think little of the service, and by the time it took off, its market position was sufficiently strong that it could avoid the onerous regulation requirements under which banks fall. This lowers the system’s cost structure, relative to that of a bank, and broadens its potential customer base. Banks have petitioned unsuccessfully for regulation; it might help the Kenyan Government holds a minority interest.
• Telecoms like Safaricom should be in the pole position: they’ve got brand recognition and millions of networked customers, whom they can contact and with whom they’ve a billing relationship. Yet they differ from the slow-moving, multinational incumbents found in the US or Europe: they still operate in a market that’s yet to suffer from a barrage of smartphones.
• If you’re curious for more, this Kindle Single is one of the best things I’ve read about M-PESA.

The Hare with Amber Eyes by Edmund de Waal
A well-written history of a family of wealthy Jewish financiers who crossed paths with Proust and Rilke and then lost everything when Germany annexed Austria. At times, I couldn’t tell if the book was fiction or nonfiction. I never fell into this book.

Love and Shame and Love by Peter Orner
This book felt lazy, as if Orner-the-short-story-writer wanted a novel and so chained short stories together, rather than write a proper novel. I do have friends who loved the book.

Open City by Teju Cole
This novel doesn’t have much of a plot, doesn’t differentiate speech from narrative, and has a marginally-present main character — and yet I loved it as much as the deafening reviews suggested I would. “Ordinary solipsism” and all the rest.

Triumph of the City by Edward Glaeser
One of my favorite nonfiction books of late. It’s fast and overwhelms – in a good way – with evidence for city life. I came away more conscious of housing stock levels and land-use regulations. Definitely recommended, particularly the chapter on Detroit.

1. The paper’s introduction seems to suggest, without being direct, that the existence of reversible transactions (and the resulting need for dispute mediation) explain micropayments’ poor adoption. My impression had been that micropayments’ lack of success had more to do with marginal costs other than fraud being prohibitively high for both spenders and financial institutions. So even if we eliminated all fraud risk, I don’t believe micropayments would make sense.

1a. Transaction batching is often cited as the way we’ve gotten to what looks like micropayments. (Payments on Apple’s iTunes store and Flattr both look like micropayments, but neither are.) If anything this seems to support the it’s-transaction-risks-other-than-fraud hypothesis; batching transaction would seem to make chasing down any instance of fraud more difficult but reduce total costs because it’s (mc*1 transaction) rather than (mc*n transactions). Though transaction batching makes dispute resolution more difficult, we’re still willing to do it.

2. Bitcoin’s gold-mining analogy will likely become my at-hand anecdote for technology that borrows from but does not replicate “the real world.” Much has been written about the reasons social software should mimic the real world; I was once one of the advocates. Yet a few recent conversations have convinced me the best technology doesn’t mimic the real world but rather extends the real world.

3. The privacy models laid out in section 10 referenced banking, but they’re general. Here’s the models from the paper in which “Traditional Privacy Model” is a traditional bank and “New Privacy Model” is Bitcoin:

Here’s Facebook and Twitter repurposed in Bitcoin’s language. Interestingly, Facebook’s model is precisely that of a bank: it sees all transactions, knows the attached identities, and mediates between parties. Twitter is a bit different; it doesn’t shield anyone or anything from the public.

What social software divorces all content from any sort of identity (pseudonymous, anonymous, or identified)? I’m looking for something that has no concept of public identities, even those defined temporarily like 4Chan’s.

This is all very high level, so more practically: I would love to see a version of Twitter that includes replies, RTs, and favs but drops usernames and profile photos — if anyone works on this, please let me know.

It’s also very technical, but the interesting part here is in the feelings: I’m very curious about the experience of reading tweets on an identity-less Twitter stream. What would the product feel like? Would different tweets spread in an identity-less network, and if so, what does that mean for Twitter? How large must the difference be between identified Twitter and identity-less Twitter before you’d reconsider your beliefs about the social web? How large before you’d change your beliefs about Facebook?

1. I agree with the three general reasons Tyler lays out: “You are productive fairly quickly, you make good contacts with other smart people, and you can demonstrate that you are smart, for future employment prospects.”

1a. The productivity point may be controversial, but it shouldn’t be. If nothing else — and there is much else — it’s rendered true by salary in the medium- to long-run. Most corporations don’t (and arguably can’t) pay a $60-120k starting salary for 22 year olds. Maybe it’s because they have trouble identifying which 22 year-olds will be sufficiently productive to justify that salary, but maybe it’s because 22 year-olds just can’t be that productive in larger firms.

1b. These jobs pay well in part because the higher-ups in each field actually work with the recent graduates and get utility out of working with smart people. It’s often worth an extra $x,000 from the partners’ pool to get the recent grad whose market rate is $75,000 rather than the $(75-x),000 recent grad.

2. These fields are considered impressive and prestigious. Talented students who guide themselves toward increasingly prestigious problems were and probably still are likely to spend time in consulting, finance, or law. These fields are the output of “the Ivy League hill-climbing algorithm,” as one friend put it. (I do believe software engineering is gaining ground here.)

3. Consulting and finance firms have figured out how to allow smart young people to live in New York/Boston/San Francisco but work in Dayton/Piscataway/Bentonville where they are most needed. This is non-trivial.

4. Adjusting for quality of life, I don’t think consulting, finance, law (or even operations at Google) pay differently. There’s time at work and workplace culture and people’s utility functions — and then if that fails, there’s expense accounts. Consulting firms often have the best ones and pay less than banks and law firms at similar experience levels. But take-home pay matters less if you charge expenses 4.5/7 days of the week to your employer.

5. Coming out of school, declaring you “don’t know what you want to do,” and devoting time to “figuring it out” is hard for many high-achieving students. Those that pull it off still face awkward questions from friends and family members. Consulting and finance (and to a lesser extent law) seem like “places where one can go and be productive and figure out what one should really do.” In these fields, career swaps are the norm.

6. That these jobs are not the most exciting things one could be doing is often the worst that can be said. They are very rarely “bad options.”

Another way to phrase this question is “why can’t [some specific field] attract more smart young people?”

In Other Rooms, Other Wonders by Daniel Mueenuddin
Recommended by someone whose book-taste I trust, especially when it comes to stories like these. (How many “here’s a story of the hardships of daily life in South Asia” have been published in the past decade in the US?) It’s actually quite good though — Mueenuddin can write well, and his characters act and speak like people, not caricatures. Start with his New Yorker short story if you’re skeptical.

Poor Economics by Abhijit Banerjee and Esther Duflo
Skimming or reading this book gives an overview of the last decade-ish of development economics literature, in which Banerjee, Duflo, and their Poverty Action Lab have had a hand. The book explains, in clear English, the “paradoxes” of life in poorer countries without assuming irrationality or pity.

Tide Players by Jianying Zha
Stories of major figures in China’s business, intellectual, and cultural sectors. I didn’t find I empathized with or liked any of the people profiled — which admittedly isn’t the point — but did find the stories narrow, interesting windows into modern-day China.

Dancing in the Glory of Monsters by Jason Stearns
One of the best books I’ve read on the conflict in the Great Lakes. Stearns avoids moralism, even temporarily. The first three-quarters of the book, mostly historical, were better than the last quarter, which was mostly forward looking.

1. The internet offers products competitive to Hollywood: there’s loads of non-Hollywood content on the web that people enjoy. More entertaining stuff finds its way to the internet every day, and as content distributors (e.g. Amazon, Netflix, YouTube) start producing content, seeking to improve their economics, they’ll be more “premium, Hollywood-like” content.

2. Labor supply is changing too; some of the people who would have gone to work in Hollywood earlier are instead choosing to work on the internet.

3. Until streaming becomes a substantial revenue source for studios, they’ll continue fighting everything that seems like to interfere with box-office receipts or DVD sales – including the internet. Parallels to Hollywood’s initial opposition to VCR sales abound. Yet Hollywood can’t rival the distribution offered by the internet, especially with its current cost structure, and piracy is most common in places where content isn’t legally available. Studios cannot build their own networks of theaters and DVD sellers that rival the internet’s reach and would do well to embrace internet distribution.

4. Wider distribution means more popularity, more of a role in the “cultural consciousness,” and more merchandise sales. Moving forward, it’ll not just be the Twilights that have merchandising machines; more films and shows are turning toward merchandise as a revenue stream (e.g. Mad Men’s collaboration with Banana Republic.) A film industry that embraces distribution as a pathway to revenue is a film industry that’s likely to embrace the internet.

5. Box office revenues have been falling, but they still matter because they seem to be the best-available indication of a movie’s future cash streams. (For how much longer will the bragging rights of box-office receipts matter? When can we start using Facebook newsfeed data to predict box office takes?) Yet without revenues to securitize and sell off upfront or better control of production and marketing costs, studios will make bets that appear to be safe. In practical terms, this means more sequels and cookie-cutter plots.

6. Competition from the internet should improve the movie-going product. Theaters will have to offer a more differentiated experience than simply a big screen, surround sound, and comfortable chairs (you can have all that and more in your home.) I expect we’ll see “super-premium” theaters in big urban areas and theaters selling alcohol in many more areas.

7. Despite the downward price pressure from Moore’s Law and componetized software, movie-making technology is becoming more expensive. Call me techno-optimistic, but this seems unsustainable.

8. One interpretation of Kickstarter’s rise in the indie movie world is that it provides a proxy for market demand. A Kickstarter project’s success can be used to assess movie-product/market fit. It is also true that Kickstarter — and “the internet” more broadly — increases the chance an indie movie will be discovered; as grocery stores move to upper floors, movie marketing will matter less.

9. As other forms of media have moved online, they’ve gotten more participatory. Participatory projects succeed when: 1) the bigger project can be parceled into smaller, lower-cost pieces; and 2) there’s a large pool of potential workers available. Figuring out how to break apart movie production while controlling or lowering video-production costs will open up new forms of “movie media.”

10. The economics of Hollywood movie production seem sufficiently unattractive to support current production costs, but today’s studios aren’t being wholly irrational either. (Put differently: it would still be difficult to produce J. Edgar, even if you started from a Silicon Valley garage; put differently still: movie production isn’t yet garage ready.) Yet Hollywood, the high-cost and high-quality producer, has begun hitting diminishing returns, and there’s an opening for “a new type of movie” that starts out looking inferior but, over time, is able to win through relentless increases in efficiency.

1. Sandberg joined the company in March 2008; in December 2008, Facebook was being financed at a $15bn valuation.

2. Let’s take $15bn as Facebook’s valuation when Sandberg’s comp package was issued. (The number might be too high; maybe there was a premium placed on preferred shares, and the company certainly grew in 2008, but it’s in the ballpark.) From March 2008-January 2011, Facebook increased it value 6.67 times. ( = $100bn/$15bn.)

3. The Times calculated Sandberg’s comp package at $1.6bn if Facebook were valued at $100bn. What’s worth $1.6bn at a $100bn valuation was worth $240mm at a $15bn valuation. (Assuming no stock grants, options resets, etc.)

Punchline: Facebook’s board granted Sandberg a $240mm comp package because they thought she could build a business that would generate billions in profit and would increase the company’s value by way, way more than she was getting paid. She did it, and that’s why she’s the number-two at one of the most valuable companies in the world.

In many ways – all practical ways? – our class isn’t a design class. There’s some stellar design classes at SVA, but there seemed to be less awareness of entrepreneurship, or entrepreneurial-ish things, among students.

And so our fairly-explicit goal is to maximize the number of students who will start companies. We’re using stories from founders, which we call love letters, to humanize “the mythic entrepreneur,” and we’ve tried to use the projects to give experience and quick wins. The final project is literal ramen (or cereal) profitability.

Gary and I are very bullish on designers’ ability to create transformative products and services.

While our views have been informed by working at USV, we’re not assuming the “end goal” is or should be a venture-backed company. We are assuming the internet will be relevant to whatever one does.

We’ve also designed for openness: the syllabus and assignments are available online. We’ll be tumblr’ing about the classes and posting audio recordings of many guest speakers. The students will be sharing their work online as well. Please do follow along if you’re interested.