by Jack Kessler, kessler@well.sf.ca.us
July 15, 2010 issue. This file presents an archive copy of the issue of the FYI France ejournal, ISSN 1071-5916, which was distributed via email on July 15, 2010.
Versions of the following have appeared online regularly, since 1992, as a feature of the FYI France ejournal, ISSN 1071-5916, which is distributed for free via email every month except August. Ejournal subscriptions may be obtained via email request to: kessler@well.sf.ca.us
Here this file is one of a number made available -- hopefully attractively, all in one place, and relevant to libraries and online digital information work in France and Europe -- as part of FYI France (sm)(tm), an online service to which anyone can subscribe for 12 months by postal mailing a check for US $45, payable to Jack Kessler, to PO Box 460668, San Francisco, California, USA 94146 (site licenses also are available): please write your email address on the front of your check. And you can pay via PayPal, on the FYI France homepage:
Please email suggestions for improvements to me at kessler@well.sf.ca.us
--oOo--
"The Atomized Library" is an idea, of two US architecture students, to inject some fundamental flexibilities into the idea of "library service".
Several French friends have written to me about this -- some alarmed, some perplexed, some simply curious, some in love -- the concept appears to have caught the French library imagination, in several quarters.
For a design studio, in their final year of architecture at San Francisco's venerable California College of Art, students Duncan Young and Bret Walters suggested that library service be, a) exploded out into the community, and, b) designed dynamically and continuously, "GoogleSearch-like", by the community it serves...
Their college is "venerable" by California standards, at least: at 100+ years, the California College of Arts and Crafts / CCAC, recently name-shortened to "CCA", is the citadel of the post-1906 earthquake Arts and Crafts Movement here, famous for having turned out some of California's finest designers, architects, and artists.
The two CCA students' Atomized Library idea contains two really interesting parts:
-- their concept is fleshed out, and interestingly illustrated, on their storyboards which appear as images on the websites listed below.
The idea of small & scattered branch libraries is nothing new: since the Baths of Caracalla, librarians have "gone where the action is", seeking out locales and activities where users may be found -- also worrying about damage to their precious scrolls or manuscripts or books, from threats ranging from tepidarium steam and bathwater to modern coffee concoctions and peanut butter.
What is new, in the Atomized Library, is the idea of insertion into unexpected corners of the city...
These two students have come up with ingenious ways of varying library spaces, so that some Atomized Library locations might include a small theater, & others a cafe, & still others a specialised music collection, & others none of these but instead a children's room -- and, in all, maybe a printed books collection or maybe not -- and all spaces modular, as architects love to do, plug-in where needed, plug-out where not.
And these little & flexible "libraries" can fit into the strangest places -- the way "pocket parks" do in New York City and San Francisco, or canal-side and railway-elevated walks do in London & Paris & Manhattan -- these small but well-connected Atomized Libraries can enhance tiny corners of surplus real estate, such as alleys & plazas & empty lots & empty buildings, going out to find the customers instead of forcing the customers to trek over to "central campus", or all the way in to now-rarely-visited "downtowns", to find them.
That perhaps is why the French have such interest... Cities in France famously are wonderful mixes of old and new and variety in buildings, and neighborhoods -- and increasingly the megalomaniac among them, Paris, has become choked with travel congestion, so much so that suburbanites there rarely visit the center nowadays except "en passant"... the way families en route to the beach do, "via" Lyon...
One can imagine many interesting nooks in Paris which might use an Atomized Library: the gare, the aerogare, the suburban shopping centre, the downtown-office quartier, the annexe of a tiny church, an eastern-end building or building lot or unused train-yard awaiting development -- not an expensive branch library, but something far simpler, digitally better-connected, flexible, and the modules of which more easily fit into a variety of urban spaces.
The second major prop of the Atomized Library idea concerns flexibility as well: this time, though, less of structure than of programme -- of a feedback mechanism, for involving user behavior and input in the library's design and its services.
These architecture students, like most students everywhere now, not only are in love with Google's "algorithm" but they actually understand it -- at least they understand it in principle, its general role -- while most in the older user-generation, when they hear the word "algorithm", simply flare or run and hide.
So Young and Walters figure library user data, like Google search data, can provide valuable "user survey"-like clues -- to not only user reading habits, but also opening hours preferences, events schedules, tastes in activities -- what library truly interested in library "service" would not want to know when their readers prefer to visit, what resources they would like to use, what library activities and events they might attend?
An informatised system can provide libraries with this data. An Atomized Library information system can pinpoint it. Readers at a tiny railroad station kiosk can tell librarians, precisely, what they would like to see and do there, changing that input as preferences change -- users at a larger, mini-theater-equipped, multi-purpose space, near an elementary school, can state their very different preferences too -- an Atomized Library in a financial district gone-to-seed can get early warnings from its users about hitech startups moving in -- a renovating industrial district can tell its library, directly via usage or indirectly from surveys on its terminals and via remote logins, whether its new "loft district" residents will want a café, or more books, or less books and more terminals.
We have excellent examples of such flexibility now, in many "branch" library systems. Many cities, in France and elsewhere, boast brave projects which have placed small versions of larger central libraries into impoverished neighborhoods, downtown "business" centers, new "loft" developments. But these are large investments, for the library system -- permanent places, with permanent staff & leases & collections. And they change slowly and lack flexibility, just as their central libraries do. The atomized library idea is different: something smaller and far less expensive and, most importantly, more responsive to users' needs, and far more flexible.
Most interesting of all, though, in the Atomized Library idea, is the data... The "GoogleSearch-like" feedback of the system could provide user-survey data to improve the library service itself. Also, though, massive behavioral and opinion and attitudinal data has been found to be invaluable for other purposes -- and it is becoming more and more key to scientific research --
http://en.wikipedia.org/wiki/Massively_distributed_collaboration
-- and for commercial purposes, "le marketing", and for politics, from People Power to Obama.com, and even now as predictors for broad social trends... just as Hari Seldon said, all his caveats against mis-use notwithstanding...
-- for a remarkable demonstration of this use of data, look at Google Domestic Trends --
-- where Google search patterns are suggested as reflectors, or even predictors, of general economic activity. Atomized Library user feedback could use these and other indicators, derived from user data and input, and even provide them to others. Many folks, in addition to libraries themselves, might want to know what categories of people use libraries, and when, and how and for what purposes: not just who they are personally and what they read -- both of those categories "suspect" for civil liberties -- but also other more general data.
France could use all this -- libraries in France could -- but the French tend to be wonderfully-suspicious... The wonderful thing about French suspicions being that they can be useful to new American ideas --
The data-collecting of the Atomized Library, for example, needs vetting by more than one culture. The students here understand this point: they considered Sao Paulo and Montreal and the Randstad -- different cultures do view civil liberties, and privacy, and urban space, very differently -- just as different places have different crime rates, and reading and digital proficiencies, and weather conditions. Young and Walter build flexibility into their study to allow for such differences, but busy implementers have to allow for them too: the streetscape Atomized Library kiosk in Sao Paulo will face different weather, and local customs, and user needs, than it might in Amsterdam.
The best thing to do, to appreciate fully and to evaluate the Atomized Library proposals, is to look at them online: think "flexibility" -- small library service-points, scattered across cities -- designed, maybe even by their users, to change as user-needs change --
Two Notes:
"L'atomisme est une théorie physique proposant
une conception d'un univers discontinu,
composé de matière et de vide..."
* The Library Algorithmic
I am suspicious by nature. It's one reason I like the French, perhaps: suspicion, in France, like beauty and garden snails and so many other things, gets raised to the level of an art-form... When new ideas arrive to the French they are welcomed, but then they get torn apart -- assiduously disassembled, ad nauseam discussed, thoroughly deconstructed -- ideas which survive the Parisian knives will be good ones, or at least they will have been well-considered.
Just so, the idea of a "bibliothèque bis", for the old BN... then the notion of a "caesura", cutting that collection at the not-entirely-arbitrary date of 1945... more recently, the notion that a single US corporation's "secret" method for sorting data should become the sole gateway to all information... French audiences, particularly those interested in libraries, have learned from experience to be very careful of information innovation.
But it makes them a great testing ground. If an idea like the Atomized Library can survive a Paris audience, how much better it will do in Berlin or Bologna or Mumbai or Beijing, where audiences are no less careful but may be somewhat more reluctant to raise objections, early enough, for designs to get effectively and inexpensively corrected.
"The Library Algorithmic"... It sounds wonderful, and it may indeed be wonderful, but only if certain audiences, recently grown shy about the invisible and secret algorithms now governing their daily lives in so many ways -- in information search and retrieval and use, in weapons guidance systems, in oil drill rig operation, in automobile braking -- can be satisfied. Success will be an important measure, but open and detailed and transparent explanation will be too, as will be lots and lots of testing.
* The City is The Library
Best of the ideas Young and Walter put forward with their Atomized Library, though, is a democratic one.
In their feedback mechanism, through which library users might even provide input for library design and operation and change, the city becomes the library...
That can be a very good thing, considering how other information systems have been absorbed into their surroundings and in turn absorbed them, to the benefit of both: the book, developing both sophisticated technique and great influence from its crude incunabula beginnings -- the telephone, once feared as a strange electrocution device -- the Internet, government engineer's plaything become globalized necessity-of-life now for so many -- if the library can meld with its own users, and vice versa, the users going forward might come to read more than just the 20% of the expensively-stored collection they ever get to now, plus they might even get to the point where they the customers efficiently design and operate the library services themselves...
The history of city planning offers many dreams of the City as an organic thing -- a live "hive", buzzing with activity, one particularly creative thinker even suggested -- so much individualized energy, gathered in one place and too often at odds with one another, how much more productive and satisfying if better-connected?
Ebenezer Howard thought it could be done with gardens, Lewis Mumford thought only if tried at a smaller scale, Jean Gottmann though loved the larger, Bucky Fuller thought putting it all beneath a big air-conditioned dome would help... Young and Walter, now in their turn, see the City with Information-Era eyes: perhaps the key, or at least a key, to running our 25+ million population linked-Global-City-system correctly, or at least better, will be more democratic sharing and shaping of its information systems...
What better place to begin than with its libraries, then: that's the treasured "past", where lie the mistakes we must remember or repeat -- also, librarians are the people among us most adept and best-trained at helping nous autres to locate and use information, "face to face" instead of or at least in addition to the "face to faceless" of online information systems -- so let's get all of that on-board, for the brick library and the digital library, both -- The City as Library, yes, good idea.
Bonnes vacances,
Jack Kessler, kessler@well.com
--oOo--
To: Edward Levin,
You said,
> Understanding Google's page-ranking algorithm is not the same as liking it.
Definitely agreed... if I ever met it in a dark alley I'd turn and run, I'm of that non-algorithmic generation myself...
The truly perplexing question I have, regarding GoogleSearch, is how so many users benignly can whisk off the top of the enormous heap whatever happens to float up there, without ever once inquiring, let alone understanding, how it got up there.
> Essentially, that algorithm privileges popular, "well-connected" information, whether or not that information is the most correct or the most useful.
I am not sure that this is all there is to it, nowadays. That was the innovation back when Sergey and Larry were in school at Stanford, yes -- they wrote some now-famous and very interesting papers about it. But since then they have stretched the notion of "algorithm" to include the entire corporate structure down at Mountain View: lots of very different inputs go into those calculations now, I believe.
I am not saying, however, that quantity implies quality, in this. As far as I understand it, "relevance" is the goal -- and as far as I understand that topic it has, per epistemology, been nearly a complete mystery since the Greeks. How modern information science lays claim to having solved that question I don't know -- I studied with a few inventors of the art, and they were far more modest in their claims than data-mining folks are nowadays.
> To structure public information and public access to that information on that model is roughly the equivalent of a library shelving only those books that are most often consulted or checked out.
That certainly would be true for the popularity rankings: circular and question-begging -- as far as the "relevance" ranking certainly and even for the "search and retrieval" steps.
The entire "relevance" enterprise, in fact, is flawed: when the little user comes to the reference desk and demands her "book about horses" librarians do, eventually, figure out what it is that she wants -- but it takes time, and patience, and an awful lot of experience with users -- and the entire process is a neurological mystery, so far as I understand it myself. Yes we do communicate -- no we don't know how that works -- there are some neurologists and philosophers who have given up the search and posited the existence of a little homunculus sitting in the brain pushing buttons, and that may well be the closest we'll ever come to self-understaning there.
> It's a limiting--if not outright dangerous strategy.
Agreed, but then so is driving a car... so is going outside one's home, or staying in...
GoogleSearch is incredibly useful, I myself believe, and lots of fun -- it has provided a very helpful window into our digital information overload situation, albeit a temporary one during this "GoogleSearch phase" of that.
I do have some friends in France, however, who were outraged to discover their cherished Victor Hugo version, of their sacred Revolution, to have been trumped by the Baroness d'Orczy's "valiant British aristocrats saving poor people from Jacobins" version, in GoogleSearch demos at their bicentenaire -- the then-BnF president wrote a book about that. Warping history definitely can be dangerous, either way.
Global cellphone information handling is about to hit, however, and per Moore's Law GoogleSearch is about to be greatly out-distanced by that -- any search strategy which effectively "finds" only the top retrieval among 7 million, as GoogleSearch now does, is going to have real trouble when that pile grows to be 700 million, and lots of Asian and Latin American fingers now very busily are building that much bigger pile.
And then there's the movie industry...
Lots more bits & bytes a-comin', in other words, and GoogleSearch will be a very crude tool for use in mining all that.
> There are services that will, for a fee, manipulate page links to ensure that your website will pop up near the top of a relevant Google search.
Googlers sniff grandly at those... What they don't know is that, as Google ages, ex-Googlers -- right now those are very few in number, as the firm is doing so well, but inevitably there will be some, at some point -- will be recruited to refine the techniques -- so, yes, inevitably that there is and will be some cheating going on. Google may be shocked, shocked to hear that alleged, but I think their leaders anyway are realistic.
> When that sort of manipulation only concerns commercial web sites, it's hard not to simply say caveat emptor. But when you apply that possibility to public libraries, the implications are a great deal more troubling.
As an emptor as well as a lector, I'm concerned with both...
But yes public library users -- and I suggest academic library users as well -- do need special protections, I agree, which don't necessarily have to be extended to buyers.
Those protections need to come from research pedagogy, however -- from librarians, but also from professors -- both, for example, might re-start the process to which I was exposed in my own education, of *being careful about sources*...
No, students should not cite Wikipedia, for instance -- but then students should not cite any encyclopedia, they should use them as valuable starting points, but not ending points -- I was taught this, very firmly, in my public and private educations back in the 60s and 70s -- what has happened to academic standards since?
> And that's before you factor in data mining of personal "user survey" information.
Yeah, we wax narcissistic, nowadays...
We have "news" bureaus now which waste our time by reporting, endlessly, on their own "news" industry -- if I see or hear or read or learn about another bit of in-depth human interest reportage about some "anchorperson" I'll resign from reading "news" altogether... or I'd like to ...
Our privacy laws too -- your point I think -- need revising, as do our fundamental notions about that. The Europeans genuinely are shocked, shocked that US folks seem so casual about the problem: there, with their long histories and memories, they know better than to let snoops of any sort comb through personal data -- here the kids let it all hang out & then some on Facebook.
> No, thanks..
You have no choice -- neither do I -- you know this... The days of Edward R. Murrow and Walter Cronkite are over...
Blame it on Andy Warhol: he predicted that one day we'd all be "famous" for 15 minutes -- well, now we are! That's the pablum which now passes for "information" -- no wonder it takes a miner.
Thanks for your note.
Jack Kessler, kessler@well.com
--oOo--
FYI France (sm)(tm) e-journal ISSN 1071-5916
*
| FYI France (sm)(tm) is a monthly electronic
| journal published since 1992 as a small-scale,
| personal experiment, in the creation of large-
| scale "information overload", by Jack Kessler.
/ \ Any material written by me which appears in
----- FYI France may be copied and used by anyone for
// \\ any good purpose, so long as, a) they give me
--------- credit and show my email address, and, b) it
// \\ isn't going to make them money: if it is going
to make them money, they must get my permission
in advance, and share some of the money which they get with me. Use
of material written by others requires their permission. FYI France
archives may be found at http://www.cru.fr/listes/biblio-fr@cru.fr/
(BIBLIO-FR archive), or http://listserv.uh.edu/archives/pacs-l.html
(PACS-L archive), or http://www.lib.berkeley.edu/Collections/FYIFrance/
or http://www.fyifrance.com . Suggestions, reactions, criticisms, praise,
and poison-pen letters all gratefully received at kessler@well.sf.ca.us .
Copyright 1992- , by Jack Kessler,
all rights reserved except as indicated above.
--hjlm--
From this point you can link / jump up to,
or you can link / jump over to:
Copyright © 1992- by Jack Kessler, all rights reserved.
W3 site maintained at http://www.fyifrance.com
Document maintained by: Jack Kessler, kessler@well.sf.ca.us
Last update: July 19, 2010