Monthly Archives: November 2006

an encyclopedia in my pocket

A while back – last March – there was a great deal of excitement over Wikipodia, an open source project to install Wikipedia on an iPod. Wanting a portable Wikipedia, I installed Linux on my brand new video iPod, a necessary prerequisite, but was disappointed to discover that Wikipodia only worked on older iPods with smaller screens. I’ve waited for an update to Wikipodia since then, but the project seems to have gone dark. Probably Wikipodia wouldn’t have been an ideal solution anyway: it requires you to reboot your iPod into Linux whenever you want to look at Wikipedia. You could have an iPod to listen to music or a Wikipedia to read, but not both at the same time.

But a partial fulfillment for my desire to have a portable Wikipedia has come along: Matt Swann has posted a script that puts some of the Wikipedia on an iPod, in iPod Notes format. While it’s much simpler than installing a new operating system on your iPod, it’s still not for everybody – it requires using the OS X command line, although there’s an Automator-based version that’s a bit simpler. (PC versions would seem to be available as well, though I don’t know anything about them – check the comments here.) If you’re willing to take the plunge, you can feed the script a page from Wikipedia and it will start filling up your iPod Notes directory with that page and all the pages linked from it. I started from the entry for book; the script downloaded this, then it downloaded the entries for paper, parchment, page, and so on. When it finished those, it downloads all the pages linked from the linked pages, and it keeps doing this until it runs out of space: regardless of iPod size, you can only have 1000 notes in the Notes directory. This doesn’t meant that you get 1000 articles. Because each iPod note can only be 4 kb long, entries that are longer than 4000 characters are split into multiple notes; thus, I wound up with only 216 entries.

Though 216 entries is a tiny subset of Wikipedia, it’s still an interesting experience having a chunk of an encyclopedia in your pocket. What I find most captivating about approaching Wikipedia this was is that I found myself browsing interesting sounding articles rather than searching them directly. The iPod doesn’t have much input functionality: while you can scroll through the list of entries, you can’t search for a subject, as you usually would. (And with only 216 entries, searching would be of limited utility at best. The Wikipodia project promises full text searching, though text entry is a difficult proposition when you only have five keys to type with.) While you can scroll through the list of entries to find something that looks interesting, you’re likely to get sidetracked by something along the way. So you browse.

To my mind, browsing is one of the primary virtues of a print encyclopedia: the arbitrary logic of alphabetization makes for a serendipitous reading experience, and you often come away from a print encyclopedia having read something in a nearby article that you didn’t intend to read. This is something that’s generally lost with online reference works: links between articles are supposed to make logical sense. This is also a reflection of our reading behavior: if I search for “book” in Wikipedia, I’m probably looking for something in particular. If I’m interested in book conservation issues, I might click on the link for slow fires. If I’m interested in some other area related to books – how to make vellum, for example – I almost certainly wouldn’t. Instead I’d click on the vellum link and keep looking from there. We tend to be goal-directed when we using Wikipedia online: it’s like going to a library and finding the specific book you want. Wandering in a library is an equally valid behavior: that’s what happens here.

Because you’re not looking for a particular piece of information, you do find yourself reading in a different way. Search-based reading is a different style of reading than browsing, which is slower and more casual. This has a downside when applied to Wikipedia: the often atrocious style is more glaring when you’re reading for pleasure rather than reading for information. And an offline Wikipedia inhibits some of the new reading habits Wikipedia encourages. I caught myself wondering how biased the declarations of the Shāhnāma‘s originality w/r/t other national epics were; without recourse to page histories and talk pages I’m left to wonder until I find myself with an Internet connection.

The experience of reading Wikipedia this way isn’t perfect: many links don’t work, and some articles seem to arbitrarily end, some in mid-sentence, some in mid-word. You also realize how many links in Wikipedia aren’t useful at all. If I’m interested in books as a concept, I’m probably not interested in 1907 as a concept, though that is the year that Marc Aurel Stein found The Diamond Sutra, the oldest known block-printed book. Marc Aurel Stein or The Diamond Sutra might be interesting subjects to a book-inclined browser; 1907 isn’t as likely. What you get on your iPod is an arbitrary selection. But there’s something very pleasant about this: it’s nice to have the chance to learn about both Neferirkare Kakai and the Rule of St. Benedict on the subway.

making MediaCommons

Back in July, we announced plans to build MediaCommons, a new kind of scholarly press for the digital age with a focus on media studies — a wide-ranging network that will weave together various forms of online discourse into a comprehensive publishing environment. At its core, MediaCommons will be a social networking site where academics, students, and other interested members of the public can write and critically converse about a mediated world, in a mediated environment. We’re trying to bridge a number of communities here, connecting scholars, producers, lobbyists, activists, critics, fans, and consumers in a wide-ranging, critically engaged conversation that is highly visible to the public. At the same time, MediaCommons will be a full-fledged electronic press dedicated to the development of born-digital scholarship: multimedia “papers,” journals, Gamer Theory-style monographs, and many other genre-busting forms yet to be invented.
Today we are pleased to announce the first concrete step toward the establishment of this network: making MediaCommons, a planning site through which founding editors Avi Santo (Old Dominion U.) and Kathleen Fitzpatrick (Pomona College) will lead a public discussion on the possible directions this all might take.
The site presently consists of three simple sections:

1) A weblog where Avi and Kathleen will think out loud and work with the emerging community to develop the full MediaCommons vision.
2) A call for “papers” — scholarly projects that engagingly explore some aspect of media history, theory, or culture through an adventurous use of the broad palette of technologies provided by the digital network. These will be the first round of texts published by MediaCommons at the time of its launch.
3) In Media Res — an experimental feature where each week a different scholar will present a short contemporary media clip accompanied by a 100-150 word commentary, alongside which a community discussion can take place. Sort of a “YouTube” for scholars and a critically engaged public, In Media Res is presented as just one of the many possible kinds of collaborative, multi-modal publications that MediaCommons could eventually host. With this feature, we are also making a stand on “fair use,” asserting the right to quote from the media for scholarly, critical and pedagogical purposes. Currently on the site, you’ll find videos curated by Henry Jenkins of MIT, Jason Mittell of Middlebury College and Horace Newcomb of the University of Georgia (and the founder of the Peabody Awards). There’s an open invitation for more curators.

Other features and sections will be added over time and out of this site the real MediaCommons will eventually emerge. How exactly this will happen, and how quickly, is yet to be seen and depends largely on the feedback and contributions from the community that will develop on making MediaCommons. We imagine it could launch as early as this coming Spring or as late as next Fall. Come take a look!

google aquires jotspot

Adding wikis to its evolving online office suite.

blogs and time (links for 11.1.06)

Interesting links that crossed my path over the past few days that I haven’t time to post on (and likely never will):

“French publishers join fight against Google Book Search”: Le Syndicat National de l’édition (SNE), a trade association of French publishers, has joined a suit brought against Google by the Le Martiniè re conglomerate in August for “counterfeiting and breach of intellectual property rights” in its book digitization program.
outside.in is a new web service co-created by Steven Johnson and John Geraci that aggregates blog content according to zip code, giving you a regularly updated guide to where you live. It uses a little Google map as a navigation tool — a dynamic table of contents.
US intelligence agencies use wikis: The CIA and other agencies have begun using an internal wiki site called the “Intellipedia” where staff post current events updates and colloborate on intelligence assessments, supposedly to avoid repeating mistakes like Iraq WMD. “‘I think in the future you’ll press a button and this will be the NIE,’ said Michael Wertheimer, assistant deputy director of national intelligence for analysis.”
Clay Shirky on “meganiches”
Wikipedia and the academy: To contribute or not to contribute? Article in Chronicle of Higher Ed. on the fraught relationship between academics and the online encyclopedia. Among other things discusses troubling disparity in quality between science articles and humanities articles. Is there a “two cultures” problem in online scholarly collaboration?
Ehon: The Artist and the Book in Japan: Glorious exhibit at the New York Public Library. A totally different way of thinking about books.

Should if:book serve as a filter and recommender, providing nutritious lists of links like the one above, or purely as a source of original ideas and commentary? If the answer is both, then what should be the ratio of shorter, “pointier” posts to longer, “thinkier” ones? Blogs are agnostic as to the varying size and speed of thoughts — everything goes into the same sinking scroll, soon vanishing into the catacombs of the monthly archives and category pages.
This works fine for news cycle or daily diary-type blogs, but it’s a handicap for a site like ours where longer meditations — the kind that would benefit from longer exposure — are the more common fare, and where extended, multi-post arcs on a relatively small cluster of central ideas are more what constitute the “story” of this blog than any given week’s smattering of entries. As I write, there are several extended conversations taking place within posts that, though only a few days old, are being pushed further and further down the scroll as newer content accumulates. The only hint of their still being active is the “recent comments” link to the right, which is at best an overheard whisper.
Given these constraints, and figuring that it’s the slower moving ideas that matter most, we generally try to avoid posting quick linkdumps like the one above — useful as they might be for annotating our wider web readings and pointing readers to interesting sites — simply because they have the unfortunate effect of pushing the other stuff down. But this only slightly mitigates the still unsolved problem of portraying complex movements of ideas over time on a dinky little blog.
As a side project, we’re thinking about how we could redesign if:book to keep the thinky stuff visible for longer and tied to past related discussions, while also keeping a swift current of useful annotated links and shorter observational posts. This might mean dividing our content into two separate feeds, as on this site.
We’ve also thought about ways to organize content thematically rather than temporally, so what you see at the top isn’t just the newest content, but a cluster of our most important and long-abiding conversations arranged by subject. We’re also considering changes to the individual permalinked pages of posts, perhaps adding dynamically generated links to related posts.
We’ve played around a bit with thematic arrangements on Mitch Stephens’ blog Without Gods. First, just below his banner there’s this tag cloud, which serves as a mental map of Mitch’s writing and interests:

Then there are four side menus with recent posts divided up by general area. “Bonner’s Field” is current events, “Tales of Disbelief” deals with characters in his book, “Thinking Out Loud” is sort of free-form jamming on ideas, and “Book Writer’s Journal” is meta-commentary on the writing process:

I’m also very taken with what this site, an NYU webzine on media and religion called The Revealer. They have a lovely section on the front page that divides articles and blog postings into three distinct tempos, or traffic lanes (which brings us back to the multiple streams/feeds idea):

“Time signature” is something we need to add to our design vocabulary for dealing with evolutionary, never-finished documents. Having multiple rates of movement in a single space can create interesting tensions and provide more points of entry to for the reader. I’m hoping we can put some of this into practice with if:book, and soon.
What are other sites that do a good job of handling time? Any other ideas as to how we might do better here?