Joe Lamantia.com

I’ll be writ­ing about tag­ging, tag clouds, folk­sonomies, and related top­ics over at Tagsonomy.com going for­ward. As Chris­t­ian Crum­lish observed, it’s been quite at Tagsonomy.com for a while, but that doesn’t mean that tag­ging is any­where close to being fully fig­ured out.
To help kick­start the con­ver­sa­tion, I’ve put up two posts since offi­cially join­ing the Tag Team; The Tag­ging Hype Cycle, and Is Tag­ging a Dis­rup­tive Inno­va­tion?.
Com­ments are already flow­ing in — be sure to join the discussion.

A few months ago, I put up a posted called Tag­ging Comes To Star­bucks, in which I attempted to make the point that it’s bizarre when a product’s meta­data *over­whelms the expe­ri­ence of the prod­uct itself in it’s cus­tom­ary real world set­ting*.
My exam­ple was the meta­data encrusted pack­ag­ing of madeleines — “petite french cakes…” — at Star­bucks. Like the famous tooth­pick instruc­tions Dou­glas Adams immor­tal­ized in So Long and Thanks For All The Fish, this is a strong dis­con­ti­nu­ity of expe­ri­ence (though not nec­es­sar­ily one indi­cat­ing things gone awry at the core of civ­i­liza­tion) that implies new cog­ni­tive / per­cep­tual phe­nom­e­non.
New expe­ri­ences and frames of ref­er­ence usu­ally lack descrip­tive vocab­u­lary, which explains why I can’t pin this down neatly in words. But this is surely some­thing we can expect to encounter more in a future pop­u­lated with find­able things called spimes.
The bal­ance hasn’t shifted so far that we’re liv­ing inside Baudrillard’s ‘desert of the real’, but we are get­ting closer with each addi­tional layer of sim­u­la­tion, abstrac­tion, and meta­data applied to real sit­u­a­tions and objects.
After all it is impos­si­ble to inter­act (smell, touch, taste…) directly with these very ordi­nary pas­tries with­out expe­ri­enc­ing the inter­ven­ing lay­ers of meta­data pack­ag­ing.
Madeleines in situ:

The label­ing:

From SLATFATF: “It seemed to me that any civ­i­liza­tion that had so far lost its head as to need to include a set of detailed instruc­tions for use in a pack­age of tooth­picks, was no longer a civ­i­liza­tion I could live in and stay sane.” ~ Wonko the Sane

Get­ting cof­fee this after­noon, I saw sev­eral pack­ages of tasy look­ing madeleines sit­ting in front of the reg­is­ter at Star­bucks. For the not small num­ber of peo­ple who don’t know that shell shaped pas­tries made with but­ter are called madeleines — not every­one has seen The Trans­porter yet — the pack­age was help­fully labeled “Madeleines”.
Prov­ing that tag­ging as a prac­tice has gone too far, right below the word madeleines, the label offered the words “tasty French pas­try”.
Just in case the cus­tomers look­ing at the clear plas­tic pack­age aren’t capa­ble of cor­rectly iden­ti­fy­ing a pas­try?
Or to sup­port the large pop­u­la­tion who can’t decide for them­selves what qual­i­fies as tasty?

OCLC has caught the socially con­structed meta­data fever. A release on the OCLC site titled “User-contributed con­tent pilot” dis­cusses a pilot pro­gram to allow Open World­Cat users to add pub­licly vis­i­ble meta­data, in the form of reviews and descrip­tive details, to exist­ing records.
This looks the lat­est step in the wave of explo­ration of meth­ods and mod­els for putting socially con­structed meta­data into prac­tice that’s play­ing out in pub­lic. (Is this nec­es­sar­ily done in pub­lic? I’m curi­ous to hear thoughts on how this might be done with closed or cloaked com­mu­ni­ties, like IBM’s intranet).
Broadly, it looks like a wide vari­ety of enti­ties are fol­low­ing the stan­dard new prod­uct or ser­vice devel­op­ment cycle with regards to socially con­structed meta­data. A sim­pli­fied ver­sion of this cycle is:
1. Con­cep­tu­al­iza­tion, tech­nol­ogy devel­op­ment
2. Prod­uct devel­op­ment
3. Intro­duc­tion to mar­ket
4. Mar­ket Accep­tance and growth
5. Ongo­ing Mar­ket as con­ven­tional prod­uct
A quick review of known social book­mark­ing / tag­ging ven­tures dis­trib­uted over a num­ber of orga­ni­za­tions sup­ports the idea that each exper­i­ment is at one of these stages.
Some visu­al­iza­tions of devel­op­ment and pro­to­type cycles are avail­able here, and here.
Where’s it headed? I think we’ll see at least forms forms or appli­ca­tions of socially con­structed meta­data sta­bi­lize and become pub­licly rec­og­nized and accepted in the near future, with more on the way that will sur­prise every­one. Those four are:
1. Fee for ser­vices mod­els, pay­ing for access to pre­mium qual­ity pools of col­lec­tively man­aged infor­ma­tion under pro­fes­sional (paid) edi­to­r­ial cus­tody. OCLC could adopt this model.
2. Non-commercial com­mu­nity dri­ven pools of social knowl­edge. This might be delicio.us.
3. Deploy­ment as an enabler or attribute of other prod­uct / ser­vice mod­els. Flickr is an exam­ple of this per­haps.
4. Pub­licly free but com­mer­cial­ized infor­ma­tion min­ing oper­a­tions, deriv­ing sal­able value from for­mal­iz­ing the seman­tic rela­tion­ships between peo­ple, groups, and infor­ma­tion objects. TagCloud.com might fall into this group, or maybe Clouda­li­cious.
5. Some­thing very inno­v­a­tive I will wish I’d thought of when it’s released.
Excerpts from the OCLC release:
“As of Octo­ber 9, 2005, Open World­Cat users are able to add their own con­tent to author­i­ta­tive World­Cat infor­ma­tion about library-held titles. Avail­able under the Details and Reviews tabs, this func­tion­al­ity per­mits those who have located library items through Open World­Cat to return to the inter­face and add eval­u­a­tive con­tent.“
“User-contributed con­tent will help extend the OCLC cat­a­loging coöper­a­tive to include non-cataloging library pro­fes­sion­als and — more impor­tantly — patrons. Their shared par­tic­i­pa­tion in World­Cat con­tent cre­ation and man­age­ment could fos­ter a larger sense of library-centered com­mu­nity and gen­er­ate more inter­est in library resources.”

JP Mor­gen­thal of DMReview.com offers a snap­shot of the process for defin­ing enter­prise seman­tics in Enter­prise Archi­tec­ture: The Holis­tic View: The Role of Seman­tics in Busi­ness.
Mor­gen­thal says, “When you under­stand the terms that your busi­ness uses to con­duct busi­ness and you under­stand how those terms impact your busi­ness, you can see clearly how to sup­port and main­tain the processes that use those terms with min­i­mal effort.“
Not a sur­prise, but how to make it hap­pen, and how to explain that to the business?

1. Cap­ture — In this phase of the process a rep­re­sen­ta­tive for each busi­ness process estab­lishes the vocab­u­lary and their mean­ings required to sup­port that process. For exam­ple, in a sup­ply chain process the rep­re­sen­ta­tive might cap­ture words, such as buyer, trans­port or pay­ment method. In addi­tion to these words the rep­re­sen­ta­tive would explain what these terms mean in rela­tion to the process.
2. Cat­e­go­riza­tion — In this phase, the vocab­u­lar­ies across all processes are orga­nized into a sys­tem… The sys­tem can be a sim­ple tax­o­nomic struc­ture that sim­ply relates process to vocab­u­lary, or it can be a more com­plex onto­log­i­cal struc­ture that cap­tures the rela­tion­ships of words across processes.
3. Lever­age — This is phase where the tech­ni­cal staff imple­ments the vocab­u­lar­ies in the form of a dic­tio­nary or reg­istry. This dic­tio­nary can rep­re­sent a sim­ple lookup facil­ity or it can become an active part of the infra­struc­ture feed­ing the busi­ness rules and busi­ness process engines.

In the same way that infor­ma­tion archi­tec­ture helps take users’ under­stand­ings of the struc­ture, mean­ing, and orga­ni­za­tion of infor­ma­tion into account at the level of domain-specific user expe­ri­ences, infor­ma­tion spaces, and sys­tems, the com­plex seman­tic bound­aries and rela­tion­ships that define and link enterprise-level domains is a nat­ural area of activ­ity for enter­prise infor­ma­tion archi­tec­ture.
Look­ing for some tech­ni­cally ori­ented mate­ri­als related to this level of IA — what I call enter­prise seman­tic frame­works — I came across a solid arti­cle titled Enter­prise Seman­tics: Align­ing Service-Oriented Archi­tec­ture with the Busi­ness in the Web Ser­vices Jour­nal.
The authors — Joram Boren­stein and Joshua Fox — take a web-services per­spec­tive on the busi­ness ben­e­fits of enterprise-level seman­tic efforts, but they do a good job of lay­ing out the case for the impor­tance of seman­tic con­cepts, under­stand­ing, and align­ment at the enter­prise level.
From the arti­cle abtract:
“Enter­prises need trans­parency, a clear view of what is hap­pen­ing in the orga­ni­za­tion. They also need agility, which is the abil­ity to respond quickly to changes in the inter­nal and exter­nal envi­ron­ments. Finally, orga­ni­za­tions require inte­gra­tion: the smooth inter­op­er­a­tion of appli­ca­tions across orga­ni­za­tional bound­aries. Encod­ing busi­ness con­cepts in a for­mal seman­tic model helps to achieve these goals and also results in addi­tional corol­lary ben­e­fits. This seman­tic model serves as a focal point and enables auto­mated dis­cov­ery and trans­for­ma­tion ser­vices in an orga­ni­za­tion.“
They also offer some ref­er­ences at the con­clu­sion of the article:

• Boren­stein, J. and , J. (2003). “Seman­tic Dis­cov­ery for Web Ser­vices.” Web Ser­vices Jour­nal. SYS-CON Pub­li­ca­tions, Inc. Vol. 3, issue 4. www.sys-con.com/webservices/articleprint.cfm?id=507
• Cowles, P. (2005). “Web Ser­vice API and the Seman­tic Web.” Web Ser­vices Jour­nal. SYS-CON Pub­li­ca­tions, Inc. Vol. 5, issue 2. www.sys-con.com/story/?storyid=39631&DE=1
• Gen­ovese, Y., Hay­word, S., and Com­port, J. (2004). “SOA Will Demand Re-engineering of Busi­ness Appli­ca­tions.” Gart­ner. Octo­ber 8.
• Linthicum, D. (2005). “When Build­ing Your SOA…Service Descrip­tions Are Key.” WebServices.Org. March 2005. www.webservices.org/ws/content/view/full/56944
• Schulte, R.W., Valdes, R., and Andrews, W. (2004). “SOA and Web Ser­vices Offer Lit­tle Ven­dor Inde­pen­dence.” Gart­ner. April 8.
• W3C Web Ser­vices Archi­tec­ture Work­ing Group: www.w3.org/2002/ws/arch/

Looks like Dave Sifry at Tech­no­rati has drunk the Clay Shirky Koolaid on tag­ging and social book­mark­ing. Here’s some­thing from Dave’s post­ing State of the Blo­gos­phere, August 2005, Part 3: Tags, that shows he’s clearly joined the acad­emy of received ideas.
“Unlike rigid tax­on­omy schemes that many peo­ple dis­like using, the ease of tag­ging for per­sonal orga­ni­za­tion with social incen­tives leads to a rich and dis­cov­er­able sys­tem, often called a folk­son­omy. Intel­li­gence is pro­vided by real peo­ple from the bottom-up to aid social dis­cov­ery. And with the right tag search and nav­i­ga­tion, folk­son­omy may out­per­form more struc­tured approches to clas­si­fi­ca­tion, as Clay Shirky points out…“
I’m dis­ap­pointed to see this. The qual­ity level of Shirky’s think­ing and writ­ing related to tag­ging is gen­er­ally low; too often he’s so com­pletely off the mark with much of what he’s said about tag­ging, social book­mark­ing, and cat­e­go­riza­tion in gen­eral that his main con­tri­bu­tion is in lend­ing a cer­tain amount of atten­tion by virtue of name recog­ni­tion to a sub­ject that used to be arcane.
There’s lit­tle need to rehash the many, many indi­vid­ual weak­nesses in Shirky’s writ­ings, just one exam­ple of which is his estab­lish­ment of a false dichotomy sep­a­rat­ing struc­tured cat­e­go­riza­tion sys­tems and social tag­ging prac­tices. Broadly, his approach and rhetoric show strong influ­ence from anar­chism, and utopian social the­ory.
From Shirky:
“There is no fixed set of cat­e­gories or offi­cially approved choices. You can use words, acronyms, num­bers, what­ever makes sense to you, with­out regard for any­one else’s needs, inter­ests, or require­ments.“
Fur­ther, “…with tag­ging, any­one is free to use the words he or she thinks are appro­pri­ate, with­out hav­ing to agree with any­one else about how some­thing “should” be tagged.“
Build­ing back on the crit­i­cique of com­put­er­i­za­tion, it’s clear that Shirky uses rhetor­i­cal strate­gies and posi­tions from both tech­no­log­i­cal utopi­anism and anti-utopianism.
Here’s Pro­fes­sor Rob Kling on tech­no­log­i­cal utopi­anism:
“Utopian images are com­mon in many books and arti­cles about com­put­er­i­za­tion in soci­ety
writ­ten by tech­nol­o­gists and jour­nal­ists. I am par­tic­u­larly inter­ested in what can be learned,
and how we can be mis­led, by a par­tic­u­lar brand of utopian thought — tech­no­log­i­cal
utopi­anism. This line of analy­sis places the use of some spe­cific tech­nol­ogy, such as
com­put­ers, nuclear energy, or low-energy low-impact tech­nolo­gies, as key enabling
ele­ments of a utopian vision. Some­times peo­ple will casu­ally refer to exotic tech­nolo­gies –
like pocket com­put­ers which under­stand spo­ken lan­guage — as “utopian gad­gets.“
Tech­no­log­i­cal utopi­anism does not refer to these tech­nolo­gies with amaz­ing capa­bil­i­ties. It
refers to analy­ses in which the use of spe­cific tech­nolo­gies plays a key role in shap­ing a
benign social vision. In con­trast, tech­no­log­i­cal anti– utopi­anism exam­ines how cer­tain broad
fam­i­lies of tech­nol­ogy are key enablers of a harsher and more destruc­tive social order.“
That Shirky would take speak from this stand­point is not a sur­prise; he’s iden­ti­fied as a “Decen­tral­iza­tion Writer/Consultant” in the descrip­tion of his ses­sion “Ontol­ogy is Over­rated: Links, Tags, and Post-hoc Meta­data” at etech, and it’s clear that he’s both tech­nol­o­gist and a jour­nal­ist, as Kilng iden­ti­fies.
Regard­less of Shirky’s bias, there is a big­ger pic­ture worth exam­in­ing. Tag­ging or social book­mark­ing is one poten­tial way for the com­mu­nity of social meta­data sys­tem users to con­front prob­lems of indi­vid­ual and group infor­ma­tion over­load, via a col­lec­tive and nom­i­nally unhier­ar­chi­cal approach to the emer­gent prob­lem of infor­ma­tion man­age­ment across com­mon resources (URIs).

“On aver­age, a new record is added to the World­Cat data­base every 10 sec­onds. Watch it hap­pen live…” Watch World­Cat grow
Accord­ing to the About page:
“World­Cat is the world’s largest bib­li­o­graphic data­base, the merged cat­a­logs of thou­sands of OCLC mem­ber libraries. Built and main­tained col­lec­tively by librar­i­ans, World­Cat itself is not an OCLC ser­vice that is pur­chased, but rather pro­vides the foun­da­tion for many OCLC ser­vices and the ben­e­fits they pro­vide.“
Here’s what went into the sys­tem while I was typ­ing this entry out:
———————
The fol­low­ing record was added to World­Cat on 08/10/2005 9:08 PM
Total hold­ings in World­Cat: 999,502,692
OCLC Num­ber: 61245112
Title: The­o­log­i­cal and cul­tural stud­ies in honor of Simon John De Vries /
Pub­lisher: T. & T. Clark Inter­na­tional,
Pub­li­ca­tion Date: c2004.
Lan­guage: Eng­lish
For­mat: Book
Con­tributed by: SAINT PATRICK’S SEMINARY LIBR
———–
Some impres­sive World­Cat sta­tis­tics from the OCLC site:
Between July 2004 and June 2005:

• World­Cat grew by 4.6 mil­lion records
• Libraries used World­Cat to cat­a­log and set hold­ings for 51.9 mil­lion items and arrange 9.4 mil­lion inter­li­brary loans
• Library staff and users con­ducted 34.7 mil­lion searches of World­Cat via First­Search for research and ref­er­ence, and to locate materials

Also:

• World­Cat has 57,968,788 unique bib­li­o­graphic records
• 53,548 par­tic­i­pat­ing libraries world­wide use and con­tribute to WorldCat
• Every 10 sec­onds an OCLC mem­ber library adds a record to WorldCat
• Every 4 sec­onds an OCLC mem­ber library fills an inter­li­brary loan request using WorldCat
• Every sec­ond a library user searches World­Cat using FirstSearch

For us infor­ma­tion types, it beats the hell out of the old pop­u­la­tion clocks that the U.S. Cen­sus Bureau still runs for the US and the world.
BTW, for the curi­ous, “Accord­ing to the U.S. Bureau of the Cen­sus, the res­i­dent pop­u­la­tion of the United States, pro­jected to 08÷11÷05 at 01:24 GMT (EST+5) is 296,854,475″

The April issue of D-Lib Mag­a­zine includes a two-part Sur­vey of social book­mark­ing tools.
Social book­mark­ing is on the col­lec­tive brain — at least for the moment –and most of those writ­ing about it choose to take one or more posi­tions for, against, or orthog­o­nal to its var­i­ous aspects. Here’s the posi­tion of the D-Lib sur­vey authors:
“Despite all the cur­rent hype about tags — in the blog­ging world, espe­cially — for the authors of this paper, tags are just one kind of meta­data and are not a replace­ment for for­mal clas­si­fi­ca­tion sys­tems such as Dublin Core, MODS, etc. [n15]. Rather, they are a sup­ple­men­tal means to orga­nize infor­ma­tion and order search results.“
This is — no sur­prise from “a solely elec­tronic pub­li­ca­tion with a pri­mary focus on dig­i­tal library research and devel­op­ment, includ­ing but not lim­ited to new tech­nolo­gies, appli­ca­tions, and con­tex­tual social and eco­nomic issues” — the librar­i­ans’ view, suc­cinctly echoed by Peter Morville in his pre­sen­ta­tion dur­ing the panel ‘Sort­ing Out Social Clas­si­fi­ca­tion’ at this year’s Infor­ma­tion Archi­tec­ture sum­mit.
The D-Lib authors’ assess­ment dove­tails nicely with Peter’s views on The Speed of Infor­ma­tion Archi­tec­ture from 2001, and it shows how library sci­ence pro­fes­sion­als may decide to place social book­mark­ing in rela­tion to the larger con­text of meta-data life­cy­cles; a realm they’ve known and inhab­ited for far longer than most peo­ple have used Flickr to tag their pho­tos.
I found some of the authors’ con­clu­sions more sur­pris­ing. They say, “In many ways these new tools resem­ble blogs stripped down to the bare essen­tials.” I’m not sure what this means; stripped-down is the sort of term that usu­ally con­notes a min­i­mal­ist refac­tor­ing or adap­ta­tion that is designed to empha­size the fun­da­men­tal aspects of some orig­i­nal thing under inter­pre­ta­tion, but I don’t think they want read­ers to take away the notion that social book­mark­ing is an inter­pre­ta­tion of blog­ging.
Mov­ing on, they say, “Here the essen­tial unit of infor­ma­tion is a link, not a story, but a link dec­o­rated with a title, a descrip­tion, tags and per­haps even per­sonal rec­om­men­da­tion points.” which leaves me won­der­ing why it’s use­ful to com­pare Furl to blog­ging?
A cul­tural stud­ies pro­fes­sor of mine used to say of career aca­d­e­mics, “We decide what things mean for a liv­ing”. I sus­pect this is what the D-Lib authors were work­ing toward with their blog­ging com­par­i­son. Since the label space for this thing itself is a bit crowded (con­tenders being eth­n­o­clas­si­fi­ca­tion, folk­son­omy, social clas­si­fi­ca­tion), it makes bet­ter sense to ele­vate the arena of your own ter­ri­to­r­ial claim to a higher level that is less clut­tered with other claimants, and decide how it relates to some­thing well-known and more estab­lished.
They close with, “It is still uncer­tain whether tag­ging will take off in the way that blog­ging has. And even if it does, nobody yet knows exactly what it will achieve or where it will go — but the road ahead beck­ons.“
This is some­what unin­spir­ing, but I assume it sat­is­fies the XML schema require­ment that every well-structured review or essay end with a con­clu­sion that opens the door to future pub­li­ca­tions.
Don’t mis­take my piqué at the squishi­ness of their con­clu­sions for dis-satisfaction with the body of the sur­vey; over­all, the piece is well-researched and offers good con­text and per­spec­tive on the antecedents of and con­cepts behind their sub­ject. Their invo­ca­tion of Tim O’Reilly’s ‘archi­tec­tures of par­tic­i­pa­tion’ is just one exam­ple of the value of this sur­vey as an entry point into related phe­nom­ena.
Another good point the D-Lib authors make is the way that the inher­ent local­ity, or context-specificity, of col­lec­tions of social book­marks allows them to pro­vide higher-quality point­ers to resources rel­e­vant for spe­cial­ized pur­poses than the major search engines, which by default index glob­ally, or with­out an edi­to­r­ial per­spec­tive.
Likely most use­ful for the sur­vey reader is their set of ref­er­ences, which taps into the meme flow for social book­mark­ing by cit­ing a range of source con­ver­sa­tions, edi­to­ri­als, and post­ings from all sides of the phenomenon.

While research­ing and eval­u­at­ing user inter­faces and man­age­ment tools for seman­tic struc­tures — ontolo­gies, tax­onomies, the­sauri, etc — I’ve come across or been directed to two good sur­veys of tools.
The first, cour­tesy of HP Labs and the SIMILE project is Review of exist­ing tools for work­ing with schemas, meta­data, and the­sauri. Thanks to Will Evans for point­ing this out.
The sec­ond is a com­pre­hen­sive review of nearly 100 ontol­ogy edi­tors, or appli­ca­tions offer­ing ontol­ogy edit­ing capa­bil­i­ties, put together by Michael Denny at XML.com. You can read the full arti­cle Ontol­ogy Build­ing: A Sur­vey of Edit­ing Tools, or go directly to the Sum­mary Table of Sur­vey Results.
The orig­i­nal date for this is 2002 — it was updated July of 2004.