I left my digs in Seattle at high noon on Monday and got as far as Missoula on the first day before snow-hypnosis drove me off the highway. I had a suspicion that the first 100 miles of my journey might be the toughest. Or maybe that was hope. They were indeed soggy, and then snowy miles, but much more interesting weather lay ahead I'm afraid. South Dakota... mid November... right. Still, I stopped at the Little Bighorn Battlefield today to pay homage to earlier exemplars of hopeless military adventurism (which side was guiltier of this is left to the perspective of the reader). It is within sight of I-90, and well worth the visit, especially in lonely November.
At one point in Eastern Wyoming, I used my GPS to look for a Starbucks, only to find that the closest one was 282 miles distant. I've decided that I prefer Red Bull to lousy coffee (at least for pharmacologic purposes), so I filled my tank and scored a couple cans. I blasted past Wall Drug (the midpoint of my journey) with hopes of making 1000 miles on the day. But not long after, the weather turned sour again, so here I am in Murdo, SD and grateful to be off the highway and settled in for the night. The driving day started with a pre-dawn close encounter with a deer and ended with a semi that passed me in near blizzard conditions, leaving me in a state of blindness resembling a drive-through car wash, except moving at 40 miles per hour in slush-ruts. Seattle rain is looking very good to me at the moment. So is sleeping in my own bed. I am full of poignant regard for friends and family on both ends of my trip.
Image: 11:55 on Monday, November 13th, loaded and ready to go
This week is National Children's Book Week, and in as much as Marguerite is a serious fan of such, I asked her if she would be my guest blogger for the event. Her contribution follows:
by Faith Ringgold (New York: Crown Publishers, 1991)
I have often thought that anyone
who had no particular reason to browse the children’s section of the
public library was missing an opportunity to discover an absolute treasure
of wonderful books, many of which I would describe as “not for children
only.” Included in this sub-genre are books that are essentially brief
memoirs of the author’s childhood, books that tell fascinating and
often little known stories from the pages of history, books that are
illustrated by acclaimed works of art, and books that are examples of
writing that conveys ideas, feelings, and information in creative, sometimes
quirky, and often inspiring language. Tar Beach
is one of my favorite examples.
Tar Beach began as a story
quilt, a piece of fabric art that now hangs in New York’s Guggenheim
Museum. Reproductions from the quilt are the illustrations for the children’s
book. Both tell the story of a young girl and her family who seek
respite from the summer heat of their New York City apartment by joining
their neighbors on Tar Beach, the roof of their apartment building.
After a delicious-looking picnic supper, the adults play cards while
the little girl, entranced by the lights of the George Washington Bridge
beckoning in the distance, drifts of to sleep and dreams of flying over
that bridge that her own daddy, an ironworker, helped to build, wearing
its string of lights like a necklace. She dreams, too, of flying
over the union house her daddy also helped to build, even though he
couldn’t join the union because of his black and Indian ancestry,
and she dreams of a life where her daddy always has a job, and her family
can have ice cream for dessert every night.
I was originally drawn to this
book for sentimental reasons, as I, too, spent many a summer night in
my early years on my own family’s tar beach, not too far from the
George Washington Bridge. But I became fascinated by the art work,
by the very idea, first of all, that a quilt could tell a story, and
then that the art work of the quilt could be used to illustrate a picture
book. I have since discovered many other children’s books illustrated
with fabric art. Briefly, here are three wonderful examples.
In Memories of Survivalby Esther Nisenthal Krinitz and Bernice
Steinhardt (New York: Hyperion Books for Children, 2005), Krinitz uses
her skills of embroidery to tell the heart rending story of her survival
of the Nazi occupation of her Polish home town.
In The Whispering
Cloth: A Refugee’s Storyby Pegi Deitz Shea (Honesdale, PA: Boyds
Mills Press, 1995), as a young Hmong girl learns the traditional art
of making story cloths, she also finds the means of telling the story
of her parents’ death, a story too painful for words.
States: Piecing Together America, written by Adrienne Yorinks and
50 librarians across the nation, with quilts by Adrienne Yorinks (Washington
D.C: National Geographic, 2005), literally depicts the growth of the
United States, quilt piece by quilt piece, as each state is added to
the original thirteen colonies. Brief descriptions of each state’s
early history, contributed by librarians from each state, add intriguing
bits of information to this lovely work of art.
I could go on, but I hope that
mention of these few titles will entice you to take a closer look at
the many wonderful books sitting on the shelves of the children’s
collection that are definitely “not for children only."
The remainder of my sabbatical here in Seattle is now measureable in double digit hours, and still, stuff is happening. In fact, the week just closing has been among the most exciting of my tenure here. In addition to enjoying the warmth of many well-wishers on the eve of my departure, I've participated in a series of activities that have converged on common themes:
The UW Turing Center invited me to a symposium on ontologies held on Tuesday, November 7. The Turing Center is a research unit at UW that is focused on developing systems that transcend linguistic boundaries to promote discovery and exchange of ideas. The Utilika Foundation, founded by Jonathan Pool, funds the Center with the expectation of progress in promoting cross-cultural and multilanguage discovery. Ontologies are fundamental to this task, affording (at least in theory) a language-neutral structure for knowledge that should help bridge linguistic barriers. You don't have to be an expert in ontologies to realize that this is a difficult problem that is not easily solved. Differences across cultures and languages are far more than simple translation problems, reflecting deep differences in underlying knowledge representation and ways of thought.
The SemGrail meeting (about which I wrote in a previous post) took place the next day, and as it turned out, amplified many of the notions discussed in the ontology symposium, albeit with a somewhat less theoretical (and more skeptical) bent. One of these ideas, a bottom-up approach to building dynamic ontologies using semantic web technologies, seems particularly promising for personal information management environments, or even federations of such that might give rise to emergent shared ontologies that could promote convergent thinking among those working on related problems.
Friday, I met with Pam Kilborn-Miller, a recent MSIM graduate who is working on an idea she describes as the Global Solutions Network. The focus of the effort is to provide an information resource for people and organizations working on UN Millenium Development Goals. How can problems in these areas be matched with:
...information about evidence-based solutions to global challenges both vertically (global, national, local) and horizontally (business, academia, government, etc.).
The problem of cross-language discovery looms large, but perhaps even more difficult is encoding the problems and the solutions in sufficiently general ways that they are discoverable and can be mapped one to the other. Improving air quality in New Delhi is related to global warming, but also (and rather more immediately) to public health respiratory issues. How can these ideas be effectively linked so as to facilitate discovery of potential "evidence-based solutions" related to both?
I don't think attribute-value pair metadata is going to do it, and neither do I believe that top-down, a priori ontology constructs are likely to succeed except in situations of tightly-limited scope. Emergent ontologies again? Pam's problem and Jonathan's hopes seem to map nicely. I am guessing they may find opportunities for mutually fruitful discussions in the near future.
Then this morning, my final Saturday morning Allegro salon, Jon Crump of our SemGrail group showed me his home-built RDF knowledge base of medieval history, built on top of Simile tools, including the Longwell RDF browser. It feels like these technologies and tools, rudimentary though they are, are the seed-crystals for the systems that Jonathan and Pam want to build, that Mark Kelly wants in his Mark-O-Pedia, that we all need to make our collaborative work more meaningful and productive... and shareable.
The convergence of such unanticipated connections is among the greatest delights of my year here. As Pasteur said, "Chance favors the prepared mind." Seattle has rather more than its share of well-prepared minds. ----- Image: Suzzalo Library in May of 2006
I've posted notes from time to time about the Semantic Web Discussion Group (aka SemGrail) that we started over the summer and have continued in the Fall term. Last night was my final time with this group during my tenure as visiting scholar. A great group of folks from a spectrum of organizations and perspectives, last night we had 11 people from 10 organizations that spanned public and private enterprise, venture capital, philanthropic, cultural, biotech, software producers, and educational organizations, all interested in how to make more semantically-aware tools and environments.
The focus of discussion was an extension of previous evenings... in a nutshell, a hybrid Blog/Wiki application that supports, for starters:
a private compendium of personal writings and thoughts
easy publishing and sharing (and the ability to move topics across the private-public boundary)
A-Z look up
both automated and user-driven tagging and clustering
rich linking within and out to other such tools
the ability to share semantics among federated applications to support collaborative work
As Mike Crandall pointed out, we're back to the future, reprising the ideas of Vannevar Bush and "As We May Think".
Oh... and all of this has to be as fast and natural as speaking (or at least writing) -- don't make me interrupt the flow of my thinking to go find links by scraping URLs out of a browser box to link into my writing, for example, as tools do today.
Being able to connect such systems (shared or federated thought-spaces) among trusted associations of people, or just publicly, seems an obvious part of the plan as well.
An approach that seemed popular in this group is the combination of RDF and Wiki software(1). Why? The sense of the SemGrailians is that benefits will accrue in the accretion of knowledge in an an open-ended, organic ontology that grows with the richness of the store of linked musings and notes. To get back stuff that you thought about a year or a decade ago, it may be useful for it to fit into one's personal ontology as well as in an index. But we don't have these structures formalized in our own minds, let alone in our blog/wiki/spreadsheets/document-stores. We'd like to have an emergent ontology that grows with our thinking and writing, and needn't be specified a priori. RDF seems the best bet for achieving this in the semantic Web technological milieu.
This is a lot to ask, yes. I don't want to stop in the middle of my work and fill in a new category template in my ontology. What if I could just mark in some way the concepts or descriptors that I think are likely to be important, and leave them as stubs that I may or may not link, connect, classify, define, reuse and index in the future? Let the system watch out for these markers and do some automated aggregation and linking, and perhaps ask me from time to time whether the accretion of a given set of terms in my writing might indicate some formal attention? Or look for co-occurrences of such in my federation of trusted colleagues? Drummond Reed introduced the idea of plus words, borrowed from his work on XRIs and XDIs(2). Plus Words: terms defined in a dynamic dictionary, dereferenceable by applications. +Social collaboration tools, +emergent ontologies, +Web 2.0 technologies... the idea is to easily mark a term or phrase that you suspect is significant, to allow an application to watch for these terms in past and future writing, in one's own work and the work of others in one's shared personal or professional spheres. A dynamic, bottom-up ontology tailored to (by!) one's own thinking and writing. And shareable.
Ok, this is all very half-baked. I don't know how to do it, I don't know whether it is what we really want. These notes are my limited interpretation of a well-lubricated discussion among some very bright folks searching for better means of collaboration. Consider them rough notes to the future. What I do know is that we're all searching for better tools to improve our work (thought) products, and the current spirit of collaborative tools in Web-2.0-space seems a promising start.
Mark "Informance Artist" Kelly of UW Libraries has helped provide a use-case to focus these discussions. He also, in an entirely different context, brought forth the pun that serves as title for this post. But if I may adapt, we are looking for tools to turn our dead screeds into live, ongoing, linked discussions with ourselves over time and with others wandering in related spheres of thinking. Tools to convert our six degrees of separation into richly articulated degrees of connection (also Mark's turn of phrase). Now we need some tool builders. As some one said last night, "Talk is cheap... the proof is in the implementation." And use.
(1) This group is not the first to have this thought... there are activities underway that explore this notion, including ITeration, and TiddlyWiki seems to have a beachhead in this area as well (not sure if there is any RDF in the latter, but the spirit we're looking for seems to be there).
(2) XRIs and XDIs are standardization efforts ongoing under the Oasis umbrella, for identifiers and data interchange respectively. More on these as I understand them better.
Thanks to Brian Dorsey, who was kind enough to take notes from last night's meeting, and share them with the rest of us. Any mangling or distortion of ideas from the discussion are my own.
----- Image: Early morning fog on Lake Washington near the NOAA facility at Magnuson Park
A week from this moment I’ll be somewhere in the vicinity of
Spokane, a few hours into a long drive to my
(other) home in Ohio. Sometime around Halloween something yanked my
head eastward, and I stopped being here and started seeing most everything as a
last this, a final that. Given the
rains of recent days, I’m afraid I might have seen the last sunshine of my
tenure here already, but true to the viral affection I’ve developed for Seattle, I find that even
the rain has charm for me. Like the
rivers here this week, I find myself full to the brim and spillable at the
Early in my tenure as iSchool Visiting Scholar, I took the List of Four bait. This week seems a good moment to revisit the
Four favorite places to eat in Seattle: (in no particular order)
Joe and Lisa’s dinner table. My neighbors - great food, great conversation.
The Mason’s dining room overlooking Puget Sound. The zeal of converts to the cuisine of the Pacific Northwest, and a taste for the wonderful wines of the region.
35th Street Bistro in Fremont – allegedly also a favorite haunt of Dale Chihouly, but by the time someone like me learns this, he’s probably gone on to other places. This place can make even eating alone a pleasure. (thanks, Mike!).
The Flying Fish is still on my list, but I haven’t been there in a while, drawn more to the lesser known treasures of a city where uncommon devotion to food is commonplace.
Four favorite commercial establishments
The Allegro Café – a great latte, and inexhaustible discourse on love, life, poetry, art, literature, and travel, all decorated with volleys of puns that make Borg and McEnroe seem like pikers.
Magus Books – they know their stuff.
Third Place Books – A café, bakery, pub, social space, and purveyor of books, new and used. Location of our Semantic Web Discussion Group during the summer and fall.
Glazier’s camera store. The place in Seattle to buy art-glass for your camera.
Four happy surprises
I’m still capable of discovery.
The rain isn’t so bad after all.
Ten perfect days showing Marguerite the Pacific Northwest.
People will leave their homes on a rainy Wednesday night (or a sunny Wednesday summer’s eve) to talk about semantic web ideas.
Not enough time in the mountains.
Not enough time in the city.
Not enough time.
Not making better use of time I had (so what else is new?).
Four most important ideas
Say Yes to everything – the basic rule of improvisation (Malcom Gladwell, in Blink).
The virtues of opaque identifiers are largely hogwash.
Web 2.0 ideas (and their cousins, Library 2.0 ideas) provide the hooks for a new kind of librarianship that feels quite a lot like the old kind of librarianship, but webulated.
WorldCat identifiers are potentially the Web’s most useful identifier for curated
intellectual content… now, if we can only capitalize on the potential….
Four books that had the most impact on my thinking:
The Soul of Money (Lynn Twist) You have enough, you’re smart enough, you’re good enough.
Blink – Malcom Gladwell’s exposition of thin-slice thinking.
A Fine Balance – Mistry’s tale of India in the years following partition. Vikram Seth’s A Suitable Boy affords a perspective of the same era from a different class. Great reading, both.
Four CDs I’ve listened to the most
Time, Sex, and Love – Mary Chapin Carpenter.
Modern Times – Bob Dylan.
The Future – Leonard Cohen.
Beethoven Symphonies Nos 5&7 (Carlos Kleiber).
The things I’m
most thankful for in the year just past (no way to stop at four):
Lorcan, who thought my sabbatical was a good idea even when my own doubts made me
want to cling to the safety of home… and Jay, who believed him.
Tony Hey of Microsoft, whose support of this activity was the tipping point I needed.
Michael Crandall, Bob and Marilyn Mason, Harry Bruce, Mike Eisenberg and the many other people at the iSchool who helped make a place for me in the vibrant intellectual life of the University of Washington.
The MSIM crew for helping me to renovate my thinking, and being great companions. In particular, Mike and Geoff and Jason whose generosity of time and ideas lit the fuse of my sabbatical.
Discovering the Allegro Café and the menagerie of astonishing people for whom this amazing place is an intellectual and social home.
A landlady of indefatigable curiosity and dedication to learning, and her brother, my housemate, who each enriched my visit in entirely unanticipated ways. A Craig’s List success story.
Marguerite, who unflinchingly supported my absence, doing the heavy lifting at home. Yes, I’ll walk Buddy for a year.
Chandra and Kody, Efthi and Terry. Randy. Brian and Elizabeth, Mark and Carol and Lauan and others whose names I don’t know. Joe & Lisa. Kevin, Corprew, Sharon, Evelyne, Chris and others in the SemGrail group.
My Mom, who got the three day tour and whose passion for life always inspires me. It’s going to be a great thanksgiving!
----- Image: I rediscovered this image of the Olympic mountains in my frantic effort to organize the 8 or 10,000 images I've generated this year.
Microformats were the topic of tonight's semantic web discussion group at Third Place Books in the Ravenna neighborhood of North Seattle. Of the seven people in attendance, no two were from the same institution, representing biotech, a semantic web startup, various cutting edge IT houses, taxonomy expertise, and libraries.
The Microformats home site is, by consensus of our group, full of bravura and could use some ideological editing. One is tempted to avert one's gaze. Still, there are good design ideas there. The standouts from my perspective:
design for people first, machines second
start with simple, tightly scoped problems
build on existing standards where useful
visible data is likely to be more useful to people, and more accurate in the long run, as errors are more likely to attract correction than hidden metadata, for example.
My discomfiture, as I've suggested in previous posts here and here, is that small, discrete problems aggregate into larger systemic problems, and I wonder whether these independently elaborated solutions will take on a form that is amenable to convergence?
On the plus side, there is lots of evidence that quick, reversible development is far more likely to find a useful target than long term, overly deliberative development, and bad paths are more easily abandoned early on.
On the other side of the ledger, one must wonder how this approach will result in solutions to the complexities of finding right-sized, modular data structures and fitting them into larger architectures. Are we to believe that there is a free lunch here somewhere? Microformats may not be "infinitely extensible and open ended", but reality is.
The general impression of our readers was of a disdain for standardization, and a lack of attention to identification and versioning of formats that will be necessary if this stuff is to achieve broad impact. For all that, there was an expectation among us that the microformat initiative has potential to evolve into a notable idiom for Web data exchange. We're hedging our bets. Buy the O'Reilly ebook (no paper yet... are they hedging their bets too?) on the topic for your Sony Reader, and curl up next to the burning yule log on the tele (not available in all markets). ----- Image: Summer flow in a mountain stream on Mount Rainier (July 2006)
Shirley Hyatt used to tell me that she routinely downloaded electronic text to one of a variety of PDA type devices, extolling the benefits of changeable font size, portability, and being able to carry a number of books in a pocket rather than in a wheel-aboard. The idea never grabbed me, and in the dozen or so years of one or another ebook-reading devices have been around, i've yet to see one that had the slightest appeal, especially against an actual book.
Tonight I saw a contender. One of our semantic web discussion group attendees brought a Sony Reader with him and showed it around. Slim, light, aesthetically attractive. But as its owner Brian said, the web images don't do it justice. It looks like just another instance of a slick, nicely designed tablet display device... yawn.
To see it in the flesh is to realize that the technology has turned an inflection point. The display is 4-level gray scale and sharp. But that isn't the half of it. The display technology is reflective rather than emissive... no bright-light washout or angular attenuation. It has some faults... a slow page refresh rate (about a second), and it lacks the instantaneous usability that iPod has made seem so easy. But to see it is to say...oh... I'll have one of those one day, and like it.
Image: a wildflower in the Union Bay Natural Area by Lake Washington. ID Welcome.