The Longest Now

OLPC Afghanistan currently works with school in Kabul, Jalalabad, Herat, and Kandahar.   This is one of our most politically complex and interesting deployments.  The initial schools involved tend to be on the wealthy side, but are still often in areas with poor power and connectivity.

Jalalabad also houses Afghanistan’s only FabLab – which set up the first “FabFi” mesh network to serve the surrounding community.  After the deployment of OLPC laptops to a local school there, families began to have access to the Internet, and to Wikipedia, for the first time.  Here are three generations of one family, outside on their roof, browsing Wikipedia together:

An Afghan family browses Wikipedia together outside

(As it happens, one of the university students who helped localize the software into Dari and Pashto is also a Wikipedian.)

Over a year after that deployment finished, I am working with FabLab folk to figure out what a similar lab and community wifi setup might look like in Herat, where we also have an OLPC school and may add another.  They’re refreshingly fun and competent people to work with, and full of great stories about young Afghans taking interesting ideas and running with them, turning them into amazing art projects or montages or startups.   Any city trying out cool new technical innovations should have a fablab to amplify the joys of being on the cutting edge.

Today we have 4,000 families connected to eachother and to the Internet in Afghanistan through OLPC; we hope to have thousands more by the end of the year.  And now I’m wondering if we can get fablabs started in the US cities where there are significant OLPC projects.

The movement roles of individuals, informal groups, and our many wiki projects need to be discussed by a different group of participants, reflecting the diversity of community and editorial efforts that make our projects work.  This discussion will receive more attention from the current MR working group once its recommendations are published this summer, but can be pursued independently from the current formal-entity discussions.

This set of issues is very broad, perhaps the broadest set of issues raised during strategic planning.  Topics on organizational structure, dynamics, and communication all have analogies in more traditional movements and organizations.  However the constellation of independent wikiprojects, ad-hoc groups, and active individuals is closer to the structure of a town than that of a non-profit; and we have had less in the way of concrete advice on how to organize and plan such work.

By the same token, these issues are central to the original success of the Projects, and to pressing questions such as how to increase participation, openness of projects to new types of contribution, and communication across projects. What groups have the role of helping wikiprojects communicate about their work, or organize and maintain their efforts?  Responding to floods of new users?  Responding to spam, vandalism, and abuse of project policies?  Maintaining  accuracy and quality?  Who are responsible for protecting contributors who are harassed or placed at legal or personal risk?  Who manages messaging on the main pages and banners of the projects?  And who prioritizes updates and improvements requested by each project?

Anyone interested in starting this next phase of movement roles analysis is encouraged to do so on Meta – and to join the current working group even if the ‘formal entity’ topics are not of interest.

Brand has a lovely interview in The European this week (auf Deutsch) on his ideology and thoughts on language preservation and nuclear power.  Worth a read, even if only in translation.

Here is one of Michael Boshernitzan’s favorite math problems, which came up again recently.  I’d like to find an optimality proof for it, let me know if you have come across one:

General Karmov orders a 100-man company to line up for inspection. He is in a foul mood, and unfortunately the commanding officer doesn’t know his quirks. So when Karmov demands they form a single orderly line, the CO forgets to order them by height. “How sloppy! Must I do everything myself?” Karmov asks noone in particular, stepping back to consider the whole line. “No, don’t move.” He pulls out his pistol and walks down the line, shooting soldiers as he goes, until what remains is ordered by height. He stops from time to time to reload. Being a mathematician, he shoots no more soldiers than necessary to order the line…. What is the minimum number of soldiers remaining when he is done?

Projects like EveryBlock have a noble goal – to have information about every block in a city for cities around the world, to let you follow information relevant to where you live and work.  But they tend to stall at the level of a few thousand new entries about a city each day — far less than even the collective newsrooms in a city process.  And they don’t have many ways for individuals to contribute information about where they live, or to distribute the task of seeking out new govenment data and posting / tagging it where appropriate.

How do we make things like this real?  How do we identify the hundred or so large ongoing tasks for a city – from posting its laws and regulations and codes, to sharing any information about its public works, to sharing updates from residents about the state of its infrastructur, to crimes and concerns, to social events and new business openings, to apartments for rent and neighborhood committee meetings?

Commons hits eight figures of media. The WMF blog post about it is lovely.

Commons growth continues to be geometric and visually stunning. And the extra horsepower running it (and making regular dumps!) marks a great improvement from last year. Now we need to help the community there keep up with its popularity!

Today the Tehran Times, an English-language paper based in Tehran, and other Iranian news sources, engaged in a bit of Internet journalism, copying some satire (‘Saudi king offers to buy Facebook for $150B to end revolt’) — down to a misspelling of Zuck’s name — into a summary of news on the King’s announced plans for social reform (providing cheap land for housing). This got its fifteen minutes of fame on forums and Twitter, enough to draw a brief official denial.

It’s not news that minor news agencies can be too busy to check facts or worry about copyright, but you’d think they would be more sensitive to satire. All I have to say is: Freshrant made the joke first.

After early exponential community growth, editing on Wikipedia has slowed recently. The number of readers, on the other hand, grows steadily. Over the past 3 years, the number of active monthly editors in all languages has declined by 12% (and twice that in English). But the effective change in active editors per reader may be 4x as large.

This change in how many readers become editors points to both a problem and a short-term solution: On the one hand, we have many more people coming to the projects who don’t know they can edit, find no reason to do so, or are discouraged before becoming active. On the other, we reach many more people than in the past, so effective changes in messaging, tools, or policy have a larger impact.

Mako and I were discussing this last night, leading to some back-of-the-envelope calculations (using some of the many great stats resources the Foundation maintains) and a heady R + ggplot session, which turned into a beautiful post on copyrighteous:

Unlike all those other [encyclopedia projects] Wikipedia is the encyclopedia that anyone can edit. Wikipedia is powerful because it allow its users to transcend their role as consumers of the information they use to understand the world. Wikipedia allows users to define the reference works that define their understanding of the their environment and each other. But 99.98% of the time, readers do not transcend that role. I think that’s a problem.

Read the full post.

We still need better demographic data, and an understanding of our own sample biases, as this recent floatingsheep article indicates.

Hooke liked to note discoveries he had made before he had found time to exlpain and prove his discoveries.  He used the simple mechanich of anagramming an entire phrase:

The true Mathematical and Mechanichal form of all manner of ARches for Building, with the true butment necessary to each of them. A Problem which no Architectonick Writer hath ever yet attempted, much less performed. abcccddeeeeefggiiiiiiiillmmmmnnnnnooprrsssttttttuuuuuuuux

This code, not decrypted during Hooke’s life, was revealed on his death to anagram to: Ut pendet continuum flexile, sic stabit contiguum rigidum inversum — “As hangs a flexible cable, so inverted, stand the touching pieces of an arch”.  The modest original context follows; (more…)

We’ve been working on a few different visualizations of the OLPC community around the world.  The most enjoyable and colorful is olpcMAP, a collaborative mashup designed by Nick Doiron that blossomed after last month’s map sprint.  (Nick is an avid map hacker and long-time OLPC volunteer who has also written the popular Map activity for offline Map-creation and -marking using XOs.)

Before this map was launched, the sorts of global visualizations we had were limited to large established groups (mapping chapters and major deployments), average statistics by region, or thousands of scattered individuals without a coherent feel.  olpcMAP combines this with personal and class projects from hackers and teachers around the world, adds search and an API for reuse, and feels above all approachable.

At the moment you can import JSON data and can choose between Google Maps and OSM layers.  The search matches both on locations on the map and on keywords used in marker descriptions. It is designed around the Google App Engine, and the growing olpcMAP API lets you request images, iframes, or KML to use this as  backend for further remixing (say, embedding a screenshot or overlay of part of the map elsewhere on the web).

You can browse the olpcMAP code and try setting up your own instance.  The framework is quite general, and it is straightforward to brand it for other communities.

I would love to see this sort of map of Wikimedians around the world, for instance — I suspect that we would see a very different picture of ourselves as a community than our current self-image.  The distribution of 10th Anniversary events this month was a first step in this direction, and was a surprise to many people.

And it would be amazing to see comparative maps of different global communities — Firefox users, Ubuntu hackers, Red Cross volunteers — using this model.   If you’ve tried to set up your own olpcMAP instance (if this becomes a general community-mapping framework, perhaps we should pick a more universal name), or have features you would love to see implemented, please let us know.

An interesting parenting discussion is underway on Quora. There’s primarily a focus in these conversations on performance, knowledge, and skill; with a few asides on inspiration and creation.

The anecdotes in question are too general to say anything about genius and will, though I have similar questions there — William James Sidis is on my mind. (ᔥBoing^Boing)

…and when they make loose with a few orders of magnitude.

Rouder and Morey critique some recent work by Bem on “Feeling for the Future”:

“[O]ur assessment is that Bem’s experiments, collectively, provide some evidence of psi phenomena, but not enough to sway the beliefs an appropriately skeptical reader…

…There is [a] surprising degree of evidence for the hypothesis that people can feel the future with emotionally-valenced nonerotic stimuli, with a Bayes factor of about 40. Though this value is certainly noteworthy, it is several orders of magnitude lower than what is required to overcome appropriate skepticism of such implausible claims.”

The framing of the questions and hypotheses here is most amusing, and worth a read. Rouder’s face sums up this whole debate.

Hat-tip to Cassandra Vieten at HuPo.

olpcMAP is ready for launch, and we’re hosting a map sprint at OLPC headquarters this week. Come help us design the next iteration of the map, and add your favorite projects to it. And encourage everyone you know who works on an OLPC project to add themselves as well — this map is designed to be a reusable repository for map data, so anything stored here will be easy to query and use in other map contexts in the future.

If you can only join for only part of the mapping sprint, try to come on Tuesday Dec 28, when you will get to see a special screening of Audobon Dougherty’s study of the impact of Internet access in rural Peru.

We like to think of humans as defined by being tool-users and language-users. But while we respect people who create new tools and languages, we don’t prioritize such work, nor have we developed fields that study how to become more efficient at developing, measuring, and improving theoretical tool and language designs and implementations.

There is the idea, in each case, that undirected evolution over time will sort out the best new tools or words or languages, organically producing [successful, widespread] inventions and [popular, widespread] terms that address all significant opportunities for us to become more effective [as tool- or language-users].

I’m not sure where this idea comes from. Three people whose thinking I admire have independently offered a version of this idea as a rationalization for why we the current level of interest in tool- and language-crafting is ‘optimal‘ or ‘sensible’. I think there are quick ways to quantify the extent to which this is not the case.

(As an example of this idea of default optimality: my clever linguist friend last night explained that there is a popular assumption in linguistics that “all living languages are equally good at transmitting all kinds of ideas,” modulo new vocabulary.)

As an example of quantifying what is missing: mathematics & physics in the last century have very actively started creating new collections of axioms, and trying to use them as a language to define what is known about math & the world. If one frames this as language-formation, it was consciously designing a better, more elegant, more expressive language — and designing one that is capable of explaining in simply terms new complex things that we observe or have discovered.

Stephen Wolfram makes the case that there are an enumerable number of different systems of logic (on the order of 50,000), and that we chose one fo these long ago which we’ve built up into moder mathematics and logic, and use to define which srots of theorems or proofs seem ‘elegant’ and ‘simple’ and can be derived quickly from its axioms. He suggests that choosing other systems of logic (and repeating the process of building out an infrastcuture of theorems and propositions) will provide fertile ground for further advances in understanding the universe. What I like most about this argument is the attempt to identify opportunities for understanding which we cannot yet approach conceptually, for lack of language to take us there.

One could do the same by moving backwards in the history of mathematics, trying to describe problems of broad modern interest without concepts and terms developed in the last 200 years. But in that case one could still imagine a single broad highway of ‘increasing sophistication’ along which we progress, adding more nuanced language as we go — when in contrast I feel that here, as in most walks of life, we have made a choice at some point to limit the building blocks of subtle communication, and are filling in the space of ideas that follows naturally from those early assumptions, but are no longer able to see what other building blocks would make possible. In particular, we have no way of estimating gaps in our understanding, or how to reach them.

So the question is: how do we reframe our development of languages outside of math so that we can start improving them consciously, measuring their effectiveness and acknowledging successes that we have discovered in the past through random-walk exploration? How do we merge the valuable properties of different spoken languages; create new auditory or visual languages; develop better sublanguages for effective communication in negotiation, love, large-scale collaboration? How can we use modern tools (wordnik, ngrams) to take control of the language-creation process, identifying trends and demands, and helping visualize new discoveries across all languages?

(more…)

Simple, passionate, efficient: a solid global implementation of one of the cheapest irrevocable life-improvements (fixing cleft palates) modern medicine can offer. Smile Train is a remarkable charity.

They are fascinating on a few levels, from the clarity of their work and tactical training local doctors, their selection of spokespeople and Oscar-winning short documentary, to the controversy of their fundraising tactics, and the lingering drama of an early fork. If you haven’t heard of them before it’s worth taking ten minutes to find out what they do.

NMAtv gives the secret DailyShow cabal a saucy McRibbing. (Hat tip to the show for calling it out.)