Monthly Archives: April 2005

“the only group that can organize everything is everybody”

Some more thoughts on Clay Shirky‘s keynote lecture on “folksonomies” at the Interactive Multimedia Culture Expo this past weekend in New York (see earlier post, “as u like it – a networked bibliography“).
Shirky talks about the classification systems of libraries – think card catalogues. Each card provides taxonomical directions for locating a book on a library’s shelves. And like shelves, the taxonomies are rigid, hierarchical – “cleaving nature at the joints,” in his words. The rigidity of the shelf tends to fossilize cultural biases, and makes it difficult to adjust to changing political realities. The Library of Congress, for instance, devotes the same taxonomic space to the history of Switzerland and the Balkan Peninsula as it does to the history of all Africa and Asia. See the table below (source: Wikipedia).

Or take the end of the Cold War.. When the Soviet Union disintegrated, the world re-arranged itself. An old order was smashed. Dozens of political and cultural identities poured out of stasis. Imagine the re-shelving effort that was required! Librarians shuddered, knowing this was a task that far exceeded their physical and temporal resources. And so they opted to keep the books where they were, changing the section’s header to “the former Soviet Union.” Problem solved. Well, sort of.
When communication and transportation were slower, libraries had a chance of keeping up with the world. But the management of humanity’s paper memory has become too cumbersome and complex – too heavy – to register every nuance, shock, and twist of history and human thought. Now, with the web becoming our library, there is, quoting Shirky again, “no shelf,” and it’s possible to have more fluid, more flexible ways of classifying knowledge. But the web has been slow to realize this. Look at Yahoo!, which, since first appearing on the scene, has organized its content under dozens of categories, imposing the old shelf-based model. As a result, their home page is the very picture of information overload. Google, on the other hand, decided not to impose these hierarchies, hence their famously spartan portal. Given the speed and frequency with which we can document every moment of our lives in every corner of the world, in every conceivable media – and considering that this will only continue to increase – there is no way that the job of organizing it all can be left solely to professional classifiers. Shirky puts it succinctly: “the only group that can organize everything is everybody.”
That’s where folksonomy comes in – user-generated taxonomy built with metadata, such as tags. Everybody can apply tags that reflect their sense of how things should be organized – their own personal Dewey Decimal System. There is no “top level.” There are no folders. There is no shelf. Categories can overlap endlessly, like a sea of Venn diagrams. The question is, how do we prevent things from becoming incoherent? If there are as many classifications as there are footsteps through the world, then knowledge ceases to be a tool we can use. And though folksonomy frees us from the rigid, top-down hierarchies of the shelf, it subjects us to the brutal hierarchy of the web, which is time.
The web tends to privilege content that is new or recently updated. And tagging systems, in their present stage of development, are no different. Like blogs, tag searches place new content at the top, while the old stuff gets swiftly buried. As of this writing, there are nearly 24,000 photos on Flickr tagged with “China” (and this with Flickr barely a year old). You get the recent uploads first and must dig for anything older. Sure, you can run advanced searches on multiple tags to narrow the field, but how can you be sure you’ve entered the right tags to find everything that you’re looking for? With Flickr, it is by and large the photographers themselves that apply the tags, so we have to be mind readers to guess the more nuanced classifications. Clearly, we’ll need better tools if this is all going to work. Far from becoming obsolete, librarians may in fact become the most important people of all. It’s not difficult to imagine their role shifting from the management of paper archives to the management of tags. They are, after all, the original masters of metadata. Different schools of tagging could emerge and we would subscribe to the ones we most trust, or that mesh best with our own view of things. Librarians could become the sages of the web.
It’s easy to get preoccupied with the volume of information we’re dealing with today. But the issue of time, which I raised earlier, should also be foremost in our minds. If libraries were to shake as violently and often as the world, they would crumble. They are not newsrooms. They are not bazaars. Like writing, libraries create stable, legible forms out of swirling passions. They provide refuge. Their cool, peaceful depths enable analysis and abstraction. They provide an environment in which the world can appear at a distance, spread out on literate strands that may be read in calm and quiet. As a library, the web feels more like the real world – sometimes too much so. It throbs with life, with momentary desires, with sudden outbursts. It is hypersensitive to change. But things pile up, or vanish altogether. I may have the smartest, most intuitive tags in the world, but in a year they might become nothing more than headstones for dead links. It is ironic that with greater access to more knowledge than ever before, we tend to live in a perpetual present. If folksonomies are truly where we’re headed, then we must find ways to overcome the awful forgetfulness of the web. Otherwise, we may regret leaving the old, stubborn, but dependable shelf behind.

wikipedia keeps apace

Barely 24 hours after being selected as the 265th Pope, Cardinal Joseph Ratzinger, now Benedict XVI, has his own Wikipedia article. Actually, Ratzinger did previously have his own page, but it was moved yesterday to the new Benedict XVI address and has since undergone a massive overhaul. The revision history, already quite long, captures in miniature the stormy debate that has raged across the world since the news broke. Early on in the history, you see the tireless Wikipedians wrestling over passages dealing with the pontiff’s early years in Germany, where he was a member of the Hitler Youth (membership was compulsory). One finds evidence of a virtual tug-of-war waged over a photograph of Ratzinger as a boy, wearing what appears to be the crisp uniform and official pin of the Hitlerjugend. The photo was eventually scrapped amid doubts about its veracity and copyright status.
Scanning across the revision history, it’s hard not be to impressed by the vigilance, passion and sheer fussiness that go into the building of a Wikipedia article. Like referees, the writers are constantly throwing down flags for excessive “editorializing” or “POV,” challenging each other on accuracy, grammar, and structure. There are also frequent acts of vandalism to deal with (all the more so, I imagine, with an article like this). Earlier today, for instance, some teenager replaced the Pope’s headshot with a picture of himself. But within a minute, it was changed back. The strength of the Wikipedia is the size of its community – illustrating the “group-forming networks law” that Kim discusses in the previous post, “the web is like high school.”
Not long ago, I posted about a new visualization tool that depicts Wikipedia revision histories over time, showing the shape of an article as it grows and the various users that impact it. For articles on controversial subjects – like popes – it would be fascinating to see these histories depicted as conversations, for that is, in essence, what they are. Any conversation that involves more than two parties cannot be accurately portrayed by a linear stream. There are multiple forks, circles, revolutions, and returns that cannot be captured by a straight line. Often, we are responding to something further up (or down) in the stream, but everything appears sequentially according to the time it was posted. We are still struggling on the web to find a better way to visualize conversations.
It’s also strange to think of an encyclopedia article as news. But that’s definitely what’s happening here, and that’s why Dan Gillmor calls attention to the article on his blog (“How the Community Can Work, Fast”). If newspapers are the “rough draft of history” and encyclopedias are the stable, authoritative version, it seems Wikipedia is somewhere in the middle.
This image sums it up well. It appears at the top of the Benedict XVI page, or above any other article that is similarly au courant.

the web is like high school

Social networking software is breeding a new paradigm in web publishing. The exponential growth potential of group forming networks is shifting the way we assign value to websites. In paper entitled “That Sneaky Exponential–Beyond Metcalfe’s Law to the Power of Community Building” Dr. David P. Reed, a computer scientist, and discoverer of “Reed’s Law,” a scaling law for group-forming architectures, says: “What’s important in a network changes as the network scale shifts. In a network dominated by linear connectivity value growth, “content is king.” That is, in such networks, there is a small number of sources (publishers or makers) of content that every user selects from. The sources compete for users based on the value of their content (published stories, published images, standardized consumer goods). Where Metcalfe’s Law dominates, transactions become central. The stuff that is traded in transactions (be it email or voice mail, money, securities, contracted services, or whatnot) are king. And where the GFN law dominates, the central role is filled by jointly constructed value (such as specialized newsgroups, joint responses to RFPs, gossip, etc.).”
Reed makes a distinction between linear connectivity value growth (where content is king) and GFNs (group forming networks, like the internet) where value (and presumably content) is jointly constructed and grows as the network grows. Wikipedia is a good example, the larger the network of users and contributors the better the content will be (because you draw on a wider knowledge base) and the more valuable the network itself will be (since it has created a large number of potential connections). He also says that the value/cost of services or content grows more slowly than the value of the network. Therefore, content is no longer king in terms of return on investment.
mean girls.jpg
Does this mean that the web is becoming more like high school, a place where relative value is assigned based on how many people like you? And where popularity is not always a sign of spectacular “content.” You don’t need to be smart, hard-working, honest, nice, or interesting to be the high-school “it” girl (or boy). In some cases you don’t even have to be attractive or rich, you just have to be sought-after. In other words, to be popular you have to be popular. That’s it.
SO…if vigorously networked sites are becoming more valuable, are we going to see a substantial shift in web building strategies and goals–from making robust content to making robust cliques? Dr. Reed would probably answer in the affirmative. His recipe for internet success: “whoever forms the biggest, most robust communities will win.”

as u like it – a networked bibliography

This past weekend I attended some of the keynote lectures at the Interactive Multimedia Culture Expo at the Chelsea Art Museum in New York. Among the speakers was Clay Shirky, who gave a quick, energetic talk on “folksonomies” – user-generated taxonomies (i.e. tags) – and how they are changing, from the bottom up, the way we organize information. Folksonomies are still in an infant stage of development, and it remains to be seen how they will develop and refine themselves. Already, it is getting to be a bit confusing and overwhelming. We are in the process of building, collectively, one tag at a time, a massive library. Clearly, we need tools that will help us navigate it.
Something to watch is how folksonomies are converging with social software platforms like Flickr. What’s interesting is how communities form around specific interests – photos, for instance – and develop shared vocabularies. You also have the bookmarking model pioneered by del.icio.us, which essentially empowers each individual web user as a curator of links. People can link to your page, or subscribe with a feed reader. Eventually, word might spread of particular “editors” with particularly valuable content, organized particularly well. New forms of authority are thereby engendered.
Shirky mentioned an interesting site that is sort of a cross between these two models. CiteULike takes the tag-based bookmark classification system of del.icio.us and applies it exclusively to papers in academic journals, thereby carving out a defined community of interest, like Flickr.
“CiteULike is a free service to help academics to share, store, and organise the academic papers they are reading. When you see a paper on the web that interests you, you can click one button and have it added to your personal library. CiteULike automatically extracts the citation details, so there’s no need to type them in yourself. It all works from within your web browser. There’s no need to install any special software.”
Essentially, CiteULike is an enormous networked bibliography. On the first page, recently posted papers are listed under the header, “everyone’s library.” To the right is an array of the most popular tags, varying in size according to popularity (like in Flickr). Each tag page has an RSS feed that you can syndicate. You can also form or join groups around a specific subject area. As of this writing, there are articles bookmarked from 6,498 journals, primarily in biology in medicine, “but there is no reason why, say, history or philosophy bibliographies should not be equally prevalent.” So says Richard Cameron, who wrote the site this past November and is its sole operator. Citations are automatically extracted for bookmarked articles, but only if they come from a source that CiteULike supports (list here, scroll down). You can enter metadata manually if you are are not submitting from a vetted source, but your link will appear only on your personal bookmarks page, not on the homepage or in tag searches. This is to maintain a peer review standard for all submitted links, and to guard against “lunatics.” CiteULike says it is looking to steadily expand its pool of supported sources.
CiteULike might eventually fizzle out. Or it might mushroom into something massively popular (it’s already running in five additional languages). Perhaps it will merge with other social software platforms into a more comprehensive folksonomic universe. Perhaps Google will buy it up. It’s impossible to predict. But CiteULike is a valuable experiment in harnessing the power of focused communities, and in creating the tools for navigating our nascent library. It might also solve some of the problems put forth in Kim’s post, “weaving textbooks into the web.” Worth keeping an eye on.

five million words of public domain restored!

“Once it had walls three miles round, with five or more gates; colonnaded streets, each a mile long, crossing in a central square; a theatre with seating for eleven thousand people; a grand temple of Serapis. On the east were quays; on the west, the road led up to the desert and the camel-routes to the Oases and to Libya. All around lay small farms and orchards, irrigated by the annual flood — and between country and town, a circle of dumps where the rubbish piled up.” (from Waste Paper City by P.J. Parsons)
It was in this garbage dump, outside the ancient city of Oxyrhynchus in modern-day Egypt, vanished except for a single column, that 400,000 classical manuscript fragments were unearthed by British archaeologists in the late 19th century. It has long been thought that the texts, which reside at Oxford’s Sackler Library, represent a vast number of missing pieces from the known classical canon, in addition to thousands of humdrum documents – petitions, land deeds, wage receipts, orders for arrest, registration of slaves and goats etc. – shedding light on daily life in the Greco-Roman world. The problem is that they are largely unreadable, crushed and mashed together, blackened by years of decay, nibbled by worms. Here and there over the years, individual texts have been deciphered, making waves through the academic world. But now, the Oxyrhynchus Papyri can at last be decoded en masse through the use of multi-spectral imaging, a technique developed in satellite photography, which teases texts to the surface with infra-red light. Hailing it as the “holy grail” of antiquarian discoveries, classicists are predicting a major wave of restoration to the received literary canon, including lost plays of Sophocles and Euripides, a post-Homeric epic by Archilochos, and even missing gospels of the New Testament.
Read article in the Independent, via Grand Text Auto.

abandon all hope, ye who enter here

Silly-sounding business jargon and corporate pep talk abounded at the eBooks in Education Conference at McGraw-Hill in New York. The kind of stuff that makes the eyes and mind glaze over. “Fluidity of work flow.” “Content creation.” “Course-centric versus learner-centric.” The “three-legged stool,” of digital publishing… Books are “learning objects.” A teacher is just a “sage on the stage,” as though teaching were an antiquated idea. Youth is a marketing problem to be solved, the latest batch of young professionals in the making, rolling along the conveyor belt. Tim Magner, a stuffed shirt from our dear own Department of Ed., talked about priming kids for the new global economy (he also took the opportunity to mention that we’re doing the same for Iraq). But nothing about the issues at the heart of education. Nothing about a civil society, an educated citizenry, etc. About nurturing critical faculties in an age of information blitz. About advancing the light against darkness. I don’t doubt the good intentions of many of the speakers and attendees (though I do doubt some) but to someone outside the industry, the conference was plainly nothing more than a congregation of hucksters and homogenizers. Only the accessibility folks (the ones concerned with opening up digital media to people with disabilities), and one soft-spoken tech development manager from the University of Nebraska-Lincoln, seemed to exhibit any kind of public spiritedness, or a belief that technology should be made to serve human beings and not companies.
Then there were the numbing waves of PowerPoint, PowerPoint, PowerPoint! Nothing more perfectly exemplifies the muddled thinking, ignorance of design, and all-around mediocrity of the so-called “ebook industry.” Each presenter – with the exception of George Kerscher of the Daisy Consortium (who is blind) – supplemented their talk with the obligatory PowerPoint presentation, which, after a while, becomes a kind of torture. The slides flip one after the other after the other, the bullet points rattle like hail. Incomprehensible charts and graphs slouch across backgrounds of pastel or mock-marble. There’s a lot of teal. A lot of magenta. The slick veneer of the corporate board room washes over you like microwaved cheese, but the occasional tell-tale typo betrays the obvious haste and lack of consideration with which the things are made. In most cases the presenter abdicates entirely and plays human accompaniment to the PowerPoint show, lamely reading aloud as the panels slide past, sort of like the airline stewardess doing her bit with the oxygen mask while the safety film plays behind. After sitting through a few of these, your brain feels like it’s been flattened out with a rolling pin. I kept thinking of Hart Crane’s lines:
“The mind has shown itself at times
Too much the baked and labeled dough
Divided by accepted multitudes.”
I also thought of Edward Tufte‘s wonderful monograph, “The Cognitive Style of PowerPoint” (encapsulated here), a shrewd critique of sandblasted thinking in Microsoft-era America. It’s a little frightening to hear adults talk about the future of education in the scrubbed, frictionless language of the corporate slideshow. If they are teaching by example, then it is a sorry example indeed.
From Tufte:
“Particularly disturbing is the adoption of the PowerPoint cognitive style in our schools. Rather than learning to write a report using sentences, children are being taught how to formulate client pitches and infomercials. Elementary school PowerPoint exercises (as seen in teacher guides and in student work posted on the Internet) typically consist of 10 to 20 words and a piece of clip art on each slide in a presentation of three to six slides -a total of perhaps 80 words (15 seconds of silent reading) for a week of work. Students would be better off if the schools simply closed down on those days and everyone went to the Exploratorium or wrote an illustrated essay explaining something.”
I couldn’t resist including the cover illustration of Tufte’s piece. which gloriously sends up the totalitarianism of PowerPoint..

the dinosaurs are myopic: publishing industry clueless about the future of textbooks

I spent yesterday in the McGraw-Hill building listening to the textbook publishing industry’s ideas about the future of the book. It was grim. The abysmal lack of creativity and insight, the singular focus on “revenue models,” and the utter disregard for the needs of students and teachers, made for a dull, and sometimes disturbing, day.
The ebook offerings ranged from plain old PDFs, to web-based books, to jury-rigged versions of Microsoft office. The only panel that offered a forward-looking vision of the future and interesting ebook software to go with it was the accessibility panel, moderated by George Kerscher–Secretary General, Daisy consortium–who is blind. This panel included a demo of Dolphin Audio Publishing’s EasePublisher a tool that facilitates the creation of multimedia content that unites text, audio and images. Dolphin and Elsevier were the only companies that addressed multimedia and its role in the future book. While McGraw-Hill is offering PDF textbooks because they are, “the easiest, fastest, cheapest solution.” Dolphin is thinking about how to enrich the learning experience for everyone. They found that when students with no disabilities used their multimedia books, they learned more. Apparently the combination of text (reading), audio (hearing the text read by a human), and image (photos, videos or illustrations that illuminate the material) enhances learning. Designing electronic textbooks that exploit this opportunity seems like a no-brainer. Teachers I’ve spoken to and my own experience with students in the classroom suggests that multi-media ignites student enthusiasm. Making PDF textbooks is like driving a Jaguar in first gear. But after 10 years of experience in the field, McGraw-Hill’s Ginny Moffet believes that: “students only care about the grade,” and “the biggest challenge to the [electronic textbook] industry is the high cost of content creation.” Hmmm, what about making a high quality product that everyone wants to buy, isn’t that the problem they should be trying to solve? It’s clear that the job of making an interesting electronic textbook is not being taken up by any of the old giants. Our prediction (and our hope) is that the future of the electronic textbook will not be directed by corporations, but by small start-ups, or non-profit consortiums of schools and academics. Efforts like the non-profit, Virtual High School, are an interesting beginning.

6th avenue agriculture

Even before the head of the University of Nebraska library began bemoaning how the pictures had fallen out of their collection of vintage agronomy ebooks, the Open eBook Forum conference on ebooks in education felt a great deal like a convention of cattlemen gathered to discuss the latest advances in treating animal ailments and increasing their milkfat percentages. The cattle, of course, are the hapless students, handily divided into K-12 & college lots. The publishing industry, with the help of the software industry, is doing their best to milk them for all they can.
The major image that came to my mind, however, was genetically modified corn. Genetically modified corn is theoretically a good thing: you get a bigger harvest of better corn. But! for the good of the masses – so it doesn’t get loose in the wild – Monsanto’s made their GM corn sterile. What this means for their bottom line: the farmers have to buy new seeds every single year. In short, what should be a renewable resource has become corporate property. And this strategy, more or less, is what the people at the Open Ebook Forum were most delighted about having hit upon. They’re selling coursepacks to college kids which expire at the end of the term, ebooks to libraries which only one user can check out at once, and more software to parents so their kids can do their schoolwork. Hopefully, this technology will let you, the efficient new school administrator, get those pesky teachers out of your payroll. Then: profit!
The future of the book looked incredibly bleak from the McGraw-Hill auditorium. One bright spot of enthusiasm: a few groups of people (including Geoff Freed from the CPB/WGBH National Center for Accessible Media and John Worsfeld from Dolphin Computer Access) working on making media more accessible to people with disabilities. Not coincidentally, they were the only people there not primarily concerned with making money off the students.

find it rip it mix it share it

That’s the slogan for the just-launched Creative Archive License Group – a DRM-free audio/video/still image repository maintained by the BBC to provide “fuel for the creative nation.” Other members include Channel 4, Open University, and the British Film Institute (bfi). Imagine if the big three US networks, PBS, NPR and the MOMA film archive were to do such a thing…

ebooks in education

story.100.dollar.laptop.ap.jpg if:book will be spending the day at the Open eBook Forum‘s eBooks in Education Conference at the McGraw-Hill Auditorium in New York.
In the spirit of the conference, here’s an AP article from last week that gives a good overview of Nicholas Negroponte‘s $100 laptop project for developing countries.
“Details are still being worked out, but here’s the MIT team’s current recipe: Put the laptop on a software diet; use the freely distributed Linux operating system; design a battery capable of being recharged with a hand crank; and use newly developed ‘electronic ink’ or a novel rear-projected image display with a 12-inch screen. Then, give it Wi-Fi access, and add USB ports to hook up peripheral devices. Most importantly, take profits, sales costs and marketing expenses out of the picture.”