YAMAZAKI’S NOTEBOOK

Icon

robotics, mostly, admired

Donation Dashboard

   

         Craigslist is a meteor whose impact is sending the lumbering media giants into extinction by taking away their lifeblood, classified advertising. Craig Newmark had a newsletter about things going on in San Francisco that he distributed by e-mail. I imagine it was the sort of thing that once would have been photocopied and tacked-up on bulletin boards in coffee shops, the public library and the doctor’s waiting room, among notices for rooms-to-rent, bicycles-to-buy and plumbers spruiking their services. Then he took the whole bulletin board and put it online. But he put every town’s bulletin boards online and undermined the financial structures of newspapers from the small community paper all the way up to the New York Times.

            No matter how expansive and comprehensive Craigslist becomes it aims to retain the integrity of a community newsletter, doing business within what the FAQ on the website calls “a culture of trust.” There’s been no brand-creation, cross-marketing, and a significant corporate investor is E-Bay, which has also laboured to create a community culture of trust, trying to maintain its connection to being an online host for neighbourhood flea-markets and yard-sales, and relying on testimonials of trust to evaluate buyers and sellers.

            There’s no editorial content on Craiglist. Community news and features has been radically re-imagined by Steven Johnson and John Geraci with Outside.in, which geotags posts and groups them into locations. The ebb and flow of news and opinion can be seen on buzz maps which represent media organisations in grey and bloggers in orange. The Washington Post has teamed up with Outside.in to revamp its City Desk with Outside.in buzz maps.

            This local and simultaneously global phenomenon is the perspective of our age says Steven Johnson who calls it “the Long Zoom”. Joseph Campbell said that the image of the earth shown to us by the Apollo astronauts widened the horizon for mythology, that since that time we’d have to be aware that local and global were inextricably linked states of being. The satellite tracking and imaging technology that was created for the moon missions has now sunk into lowly domestic appliances and communications devices. Outside.in’s navigation system is linked to Google earth’s maps.

            I followed a couple of links from John Geraci’s blog through to assessments of the media. The gist of these reports was that newspapers have a different financial model online because their digital service is accessed in pieces. A reader may perform a search, follow a link, be responding to an e-mail alert and may never even visit the newspaper’s site to read the story. Offline the prestige of the newspaper is in its whole. Investigative journalism may occasionally be dull but it’s the kind of endeavour that brings prestige to a newspaper and is worth the investment in reporting costs. Online an investigative report may bring in few readers and be difficult to match with keyword advertising that readers will click through to. The ideal online story might be a sensational new study of depression, say, that can be surrounded by pharmaceutical advertising that worried readers would surely be likely to click on.

            One of the reports asked Craigslist founder Craig Newmark what he thought was important in newspapers and he replied “investigative journalism”. Well there’s no hope the blogger mused, now that Craigslist is inadvertently destroying the organizations that could deliver investigative journalism. But another blogger suggested that what was needed was not to look at ways of propping up the same kind of media we already have, but taking a fresh approach and experimenting, taking the values of journalism, and seeing how they might effectively work with these new tools, to assess social value not just financial returns. The issues of trust, accountability and responsibility underpin investigative journalism and the project that the Berkeley Center for New Media has undertaken since Craig Newmark made an endowment looks at how the accumulated wisdom of a community can make considered choices about charitable giving. The Berkeley Center for New Media has created a Donation Dashboard that ‘learns’ from the opinions of its users to make suggestions about matching potential donors with charitable organizations.

            Here’s how it works: you are presented with brief descriptions of non-profit institutions and asked to rate each in terms of how interested you are in donating to it. The system analyses ratings in light of others’ ratings and does its best to allocate your available funds in proportion to your interests. Your customized “donation portfolio” is presented in an easy-to-understand pie chart that you can save at the site for future reference. Donation Dashboard, which is being developed by the Berkeley Center for New Media, extends machine learning techniques used by commercial websites to recommend movies, music, and books. Donation Dashboard goes beyond existing charity ranking sites by statistically combining your ratings with the ratings entered by your fellow good Samaritans to compute a portfolio customized to your interests.

            A few years ago the New Yorker’s finance columnist James Surowiecki wrote a book called The Wisdom of Crowds that told of how individuals acting alone, when their actions are aggregated, very often make more considered and correct choices than those following an expert handing opinions and advice down. Donation Dashboard has a conceptual clarity that asks users to base their opinions on two measures, the operational effectiveness of the organization represented as a percentage. The higher the percentage the more your donation goes directly to the mission, the lower the percentage the more your donation is applied to the administrative and operating costs of the organization. And a sliding scale that asks you to rank your interest in the philosophy of the organization from “not very” to “very”. 

            The Director of the Berkeley Center for New Media, Ken Goldberg, is also a scientist and an artist and in telerobotic art projects created through the Industrial Engineering and Operations Research Department at Berkeley has been working with intelligent databases and collaborative filtering programs since before we were used to encountering them on the likes of Amazon.com suggesting other purchases, or ranking the popularity of articles online. Amazon’s suggestions fall within narrow and doggedly obvious comparisons. When I purchase a Nick Cave album I’m given a range of Tom Waits and Leonard Cohen albums to buy. But the ranking system has no way of measuring that I’m more likely to buy Miles Davis’s last album, Doo Bop, because my curiosity had been stirred by Nick talking about listening to a lot of late Miles Davis albums.

            In the material world we’re guided by the opinions and suggestions of others, and are taken into realms that are seemingly unpredictable. Ken Goldberg’s first collaborative filtering program Jester, told jokes that we rated, and then it tried to tell us more jokes we’d find funny. The parameters were complex and finely calibrated and I was often surprised that the program could follow my eccentric and illogical sense of humour, and just as often be surprised that I’d find something funny that I wouldn’t have imagined I’d find funny. Something similar has happened with my first few experiences with Donation Dashboard. I’m generally drawn to operations with lean operating expenses, but there are exceptions. I’m very interested in practical medical programs that provide ongoing services in remote, dangerous locations and I’d want them to have the necessary equipment and security and incur travel costs rotating staff. Donation Dashboard is programmed to pick up on these nuances, and in some cases I overturned previous prejudices, rethinking an organization I’d had a narrow perception of previously.

            The Berkeley Center for New Media has a broad definition of media, including theories and philosophies as media. Donation Dashboard might just stand as a metaphor that helps us think about what we deem valuable and are willing to pay for. Media organizations are increasingly in thrall to advertisers, telling us this is the bitter pill we have to swallow in order to pay for the investigative and community reporting that might not generate revenue. So instead of cultural reports we get lifestyle stories with marketing tie-ins.

            New York venture capitalist Fred Wilson, who has invested in Outside.in and de.licio.us (before it was sold to Yahoo!) has been writing recently on large organizations not understanding the value of the smaller companies and services they buy, gutting or diminishing them, without capitalising on what makes them valuable to the people who use them. I’ve been annoyed, lately, that the New York Times has removed the bookmarking site, de.licio.us from the “share” function in favour of Yahoo! Buzz, a popularity ranking service for articles. The New York Times has been the newspaper I’ve valued most, but it’s astonished me to realize that I value de.licio.us more, that I read the Guardian first now and often buy the Guardian Weekly from newsstands in Sydney, several blogs (particularly Bldgblog )next, and catch up with the New York Times often through Dayna Bateman’s blog, Detritus. She condenses articles in a way that makes them poetry.

              The Donation Dashboard pie-chart, cutting an individual’s donation into several parts, acknowledges that the world we live in is fragmented but that we can perceive it as a whole, symbolically. A community newspaper might not be a single entity any more, or part of a chain, but made up of Outside.in articles and Craigslist ads without the two organisations ever having to have anything directly to do with one another. What’s important is to be able to measure and aggregate trust, to learn from and benefit from the thoughts and insights of others.

Filed under: Ken Goldberg , , , , ,

Leave a Reply