Matthew Russell recently interviewed me about Mifos for the PayPal X Developer Network. This came about because Adam Monsen, Van Mittal-Henkle, and I gave a talk called “Ending Poverty with Software” at OSCON 2010.

I don’t believe that only software can end poverty. But if we want to reach the 2-3 billion people in poverty, we definitely need software.

Maple and WiShield running the WebServer demoPerry Hung just ported the AsyncLabs WiShield 2.0 (ZG2100) drivers to the Leaf Labs Maple board!

Here’s picture of the two boards running the WiShield WebServer demo. You can see the wireless connection light is on.

Maple is a sweet new Arduino-like physical computing platform. It’s got analog and digital in and out like the 16MHz, 8-bit Arduino, but is based on a faster, 72Mhz, 32-bit ARM Cortex M3 processor from ST Microelectronics. It’s about $50. The WiShield is about $50 too. The ZG2100 is only about $30, so it is probably possible to build a lower-cost combined board.

HTTP is the new device driver! A new age of iPad and iPhone compatible, Arduino-like devices are upon us! Yay!

P.S. If you want to check out the code yourself, see
Perry’s GitHub repository
. Check out the wishield-dev branch like this:

git clone -b wishield-dev git://github.com/iperry/maple-library.git

Build from the command line like this:

$ make flash
$ make install

LANL is working on low cost, IEC fusion power reactors. This is Inertial Electrostatic Confinement-based system, otherwise known as the Farnsworth-Hirsch Fusor, invented by Philo Farnsworth, who also invented television, back in the day.

Robert Bussard, Eric Lerner’s Focus Fusion, and others have also been working on inexpensive means of fusion. These are important, because unlike the hugely expensive ITER, these types of reactors are small and affordable by cities or companies. Reactors at ITER’s scale require funding from multiple nation-states.

The inexpensive fusion technologies really could usher in a whole new world – with cheap electric power, we can cure many of the world’s problems that are related to scarcity*: water shortages, food shortages, poverty caused by lack of clean industrial development. It means fast and inexpensive space travel, too – possibly enabling us to survive world-size disasters like asteroid impacts.

I’m hopeful that more resources being put into this kind of power generation will yield results faster than ITER, which has been slow and expensive.


* Yes, if we can also stop fighting so much. I’m hopeful that when cheap fusion power is prevalent we will at least stop fighting wars over resources.

Caleb Elston:

A good product is not a bucket of features, a mere bulleted list of
things users can do. A good product is an exercise in exclusion. A
good product is defined as much by what it doesn’t do as by what it
does. This is especially true at the start. At the onset, people don’t
care about you, they don’t know what you do, and they certainly won’t
put up with a confused product. This apathy must be acknowledged and
combated in a product that does one thing extremely well; otherwise it
won’t stick.

(Via Hacker News)

One thing that can be said about Julian Assange is that he is effective at radical transparency. Large hierarchical organizations (militaries, corporations) now fear for their secrecy. John Robb has a good article on Assange today:

Julian Assange

…what’s really interesting about Wikileak’s Julian Assange is that he is one of the most important innovators in modern conflict alive today. A true global guerrilla. In other words, he’s a person that has effectively adapted modern methods of warfare/conflict to engage and defeat the biggest nation-states in the world.

Robb uses the language of warfare a lot – which can be offputting. But if you translate “warfare” into “conflict” you are entering the world of Sun Tzu’s Art of War, where conflict is a natural part of human life, and is not good or evil. What matters is how to achieve a victory that can “take whole,” subdue conflict while doing what is right for all people involved:

Therefore, one hundred victories in one hundred battles is not the most skillful.
Subduing the other’s military without battle is the most skillful.

(The Art Of War, Denma Translation, Chapter 3: Strategy of Attack)

This is very interesting when you consider major human problems as the “enemy”: poverty, warfare, disease, starvation, oppression. Robb says in another article on Open Source Warfare:

Open source warfare is an organizational method by which a large collection of small, violent, superempowered groups can work together to take on much larger foes (usually hierarchies). It is also a method of organization that can be applied to non-violent struggles. It enables:

  • High rates of innovation.
  • Increased survivability among the participant groups.
  • More frequent attacks and an ability to swarm targets.

Here I think Robb is using the word violent to mean “acting with or marked by or resulting from great force or energy or emotional intensity,” not necessarily causing physical harm to anyone.

This has application to the “war on poverty” – following Robb’s ideas, and Assange’s lead, if we actually want to eliminate poverty from the world, we’d start a movement of open-source wealth creation that had the characteristics outlined above, an open source war on poverty. It would be able to innovate quickly, be resilient, centerless, and would quickly swarm around key opportunities, key points where progress can be made quickly and on wide scale.

Your vote on American Express’ website will help us win a $200,000 grant – we’ll use that money in our quest to end world poverty. Vote here (after clicking the link, you need to register the first time you vote).

For more information about why you should vote, one of our interns Fred Graves made this informative and entertaining video!

Great post on Business Model Alchemist on using Customer Development for social business. This is what we’re doing where I work at the Grameen Foundation’s Mifos Initiative, as I described in a post a few days ago.

Business Model Alchemist does an imaginary case study for the very real PeePoo company, makers of a biodegradable bag you can use as a portable toilet. PeePoo is trying to mitigate the problem of flying toilets that are found in slums worldwide – the practice of peeing or pooping in a plastic bag, and throwing it into the air away from your house. It’s a significant problem, because more than 2 billion people live without access to basic sanitation.

The BMA article covers a couple of important points:

  • You need to find a business model that makes money, but also one that will help significant numbers of people.
  • The business model you start with – that you think up in your offices – will probably be wrong. You need to go through several or many iterations that incorporate learning from the field.

Peepoople, for example, has a proven product/technology that works and was tested – the Peepoo bag. However, the company acknowledges that this is only a starting point. The management team knows that it has to think through several possible business models in order to find one that is sustainable AND has a substantial impact…

The Customer Development process assumes that many of the initial assumptions about your business model are probably wrong, which you will find out in the second step of the process, Customer Validation. It is only when you start testing a business model or aspects of it with customers that you will find if your hypothesis were right or wrong. Hence, the Customer Development Process builds in an iteration loop to fix the shortcomings of your business model. Eric Ries, who built on Steve’s work, coined this business model iteration loop the Pivot.

Richard Hamming gave a talk on what it takes to be individually great. He says,

Richard Hamming

In summary, I claim that some of the reasons why so many people who have greatness within their grasp don’t succeed are: they don’t work on important problems, they don’t become emotionally involved, they don’t try and change what is difficult to some other situation which is easily done but is still important, and they keep giving themselves alibis why they don’t. They keep saying that it is a matter of luck. I’ve told you how easy it is; furthermore I’ve told you how to reform. Therefore, go forth and become great scientists!

It’s a long talk, but well worth reading. It does point the way for individual greatness. This is in reach of many people, although there currently aren’t many people who want to follow that path. But we do need a reliable way to produce greatness – great solutions to the many problems humanity is facing. But how?

The best idea I have is – that it’s through teams. Not just any team, but teams that can consistently produce great results or output the way the few great people do each generation. Hamming cites some great scientists, since that’s his field: Fourier, Ampère, Watt. Shannon, Fermi, Teller, Feynman. There are analogous teams. I will call these “multipersonal entities.”

This is a term originated by Jim and Michele McCarthy to describe teams that can aggregate the greatness of their members without being hindered by the non-great parts. Teams like this can also spread the personal growth that is necessary to be great among the team members, lowering the barriers to greatness. I was telling a friend about this, and he said, “It just sounds impossible.” But it’s not.

In a conversation today, Michele said, “I want team psychology to be as well understood and commonly used as individual psychology. I want it to be well known how to create great teams.” And I think she also meant, I want it to be standard practice that you use great teams to solve great problems, and standard practice to create them when you need them.

This sounds impossible, eh? Hamming says,

One of the characteristics of successful scientists is having courage. Once you get your courage up and believe that you can do important problems, then you can. If you think you can’t, almost surely you are not going to.

Therefore, go forth and become a great team!

(Thanks to Cal Newport for getting me to re-read this before I talked with Michele yesterday.)

Indian women in a microfinance group

So it was 101 days ago that I attended Eric RiesStartup Lessons Learned conference. Going to that conference set off a chain of events, which is why I haven’t posted since then.

The SLLConf morning session was all on Agile, and since my team is 100% Scrum and Extreme Programming, I didn’t go. I did go to afternoon sessions, though, and in particular, the ones on the Customer Development case studies were brilliant – Dropbox (which I use and love); PBWorks; and Food On The Table. Each presenter started out describing their business, each making slow progress. But slow progress is death to a startup. So they found ways to accelerate their feedback loop, their OODA loop. They all got it down to weeks or even days.

Days! This means, find a customer problem, come up with a solution, test and build software, ship it to customer, see how it works, and do it again, in days! I also know Eric’s last company, IMVU, did the same. This is the essence of Lean Thinking.

And the whole time at SLLConf, I was thinking, I am doomed! I am doomed! Because my business is too damn slow. At Mifos, we sell (er, sold) enterprise software, installed on-premise, to medium sized microfinance banks (50k-500k end-clients each). The implementations take up to 6 months. I can only do 2-3 implementations at a time. My software releases take 3 months each, and customers don’t even install them right away. How the hell can I pivot anything? The feedback loop is too slow.

Especially when I compared myself to the SLL case study companies – these were companies moving at web speed, shortening their learning cycles, and learning and pivoting like mad to find a business model.

So before the conference the Mifos leadership team had all read Steve Blank’s book, Four Steps to the Epiphany. And we had been discussing customer development and trying to implement it. But with little success.

The “I am doomed!” thoughts made for a miserable few weeks. However, as a team we realized that if we couldn’t learn faster, and pivot faster, we would really be doomed instead of just thinking that. Our funders and customers wouldn’t stand for it. And we wouldn’t be able to end poverty.

So we pivoted, and repositioned Mifos around hosted (“cloud”) services. Yes, pivoting is painful. I think it’s important to make it less painful, I’ll say more about that in another blog post.

Our app is web based, so now we host it on Amazon EC2. We are going after smaller customers (microfinance institutions, MFIs) because they’re faster to decide, faster to implement, and Grameen Foundation and other investors can actually invest in them. We are no longer selling enterprise deployments, unless they want to be implemented in the Cloud. We host the software, so we can push updates to it whenever we want, as long as we don’t break customers. We have the data, so we can tell whether what we’re doing is helping or hindering MFI success.

We are also building a version of IMVU’s “cluster immune system” of stages of automated tests mated with a cloud provisioning system to support rapid pushes of software to our test, staging, and production servers. We already had continuous integration so we can build on that knowledge.

We just did our first 5-day on-site deployment (down from 6 months!) in India with an MFI there, Light Microfinance. And we got another small Mexican MFI, Podemos Progressar, up in 5 days with no on-site visit at all. It will take another several months or so to see if these MFIs are actually doing well using the software. Meanwhile we have a bunch more small MFIs in the pipeline and aiming to do simultaneous developments which would vastly increase our capacity. We are going to rebuild our team structure around Customer Development and Lean Startups. We also have some cool ideas about what software and changes we can make so we can go even faster.

And along with that we’ve been collaborating closely with a couple of other Grameen Foundation departments, Social Performance Measurement which does the Progress Out of Poverty Index methodology for measuring whether a microfinance institution is helping people raise themselves out of poverty, and Capital Markets Advisory Committee, which invests in socially responsible, fast growing MFIs.

We have more hypotheses we’re testing, and I’ll say more about those in coming weeks and months as they become public. In the meantime, check out our new Mifos.com website that explains much of this!

It’s been an exhausting, exhilarating, and sometimes painful 101 days. And we still have a lot to learn!