Tag: mental_models

Discount Code for Indi Young's 'Mental Models' Webinar

December 10th, 2008 — 6:04am

Design­ers, prod­uct man­agers, and any­one who aims to cre­ate rel­e­vant and beau­ti­ful expe­ri­ences would be wise to check out Indi Young’s upcom­ing webi­nar, Using Men­tal Mod­els for Tac­tics and Strat­egy, on Decem­ber 11th. Indi lit­er­ally wrote the book on men­tal mod­els for user expe­ri­ence — read it, if you haven’t yet — and this webi­nar is part of the Future Prac­tice series pro­duced by Smart Expe­ri­ence and Rosen­feld Media, so expect good things for your mod­est invest­ment.
Even bet­ter, our friends at Smart Expe­ri­ence and Rosen­feld Media are offer­ing a 25% dis­count on reg­is­tra­tions, which is good for these tough times.
Use this dis­count code when reg­is­ter­ing: LAMANTIAWBNR

Comments Off | Information Architecture, User Experience (UX)

New Books: 'Tagging' and 'Mental Models'

March 12th, 2008 — 11:00am

If you’re inter­ested in tag­ging and social meta­data, social book­mark­ing, or infor­ma­tion man­age­ment, be sure to check out Gene Smith’s Tag­ging: People-Powered Meta­data for the Social Web recently pub­lished by from New Rid­ers. I reviewed some of the early drafts of the book, and it’s come together very nicely.
Tag­ging takes a very prac­ti­cal approach, and pro­vides an ample set of exam­ples in sup­port of the insight­ful analy­sis. After an overview of tag­ging and its value, the book addresses tag­ging sys­tem design, tags in rela­tion to tra­di­tional meta­data and clas­si­fi­ca­tion sys­tems, and cov­ers the user expe­ri­ence of cre­at­ing and nav­i­gat­ing tag clouds.
Gene likes to build things, so Tag­ging includes a chap­ter on tech­ni­cal design com­plete with sug­gested tools and tuto­ri­als for cre­at­ing your own tag­ging apps.
All in all, Tag­ging is a wor­thy intro­duc­tion to the sub­ject, and a guide for deeper explo­ration.
While we’re talk­ing books, kudos to Rosen­feld Media on the pub­li­ca­tion of their first book, Men­tal Mod­els; Align­ing Design Strat­egy with Human Behav­ior, by the very tal­ented Indi Young!
Men­tal Mod­els is richly illus­trated, filled with exam­ples, lucid, and accom­pa­nied by a con­sid­er­able amount of addi­tional con­tent from the Rosen­feld Media web­site.
Indi has con­sid­er­able expe­ri­ence teach­ing oth­ers the tech­niques and meth­ods behind cre­at­ing insight­ful men­tal mod­els for audi­ences and cus­tomers. Cog­ni­tive / frame­worky meth­ods can feel a bit heady at times (espe­cially how-to’s on those meth­ods), but Men­tal Mod­els is straight­for­ward read­ing through­out, and an emi­nently prac­ti­cal guide to using this impor­tant tool for user expe­ri­ence design and strat­egy.
Men­tal Mod­els is avail­able elec­tron­i­cally as a .pdf for indi­vid­ual and group licenses, or in hard copy; it’s choose your own medium in action.

Comments Off | Reading Room

Why Failed Societies Are Relevant to Social Media

June 18th, 2007 — 10:08am

For reg­u­lar read­ers won­der­ing about the recent quiet here, a notice that Boxes and Arrows will shortly pub­lish an arti­cle I’ve been work­ing on for a while in the back­ground, titled, “It Seemed Like the Thing To Do At the Time: The Power of State of Mind”. This is the writ­ten ver­sion of my panel pre­sen­ta­tion Lessons From Fail­ure: Or How IAs Learn to Stop Wor­ry­ing and Love the Bombs from the 2007 IA Sum­mit in Las Vegas.
I’ve writ­ten about orga­ni­za­tions and fail­ure — Signs of Cri­sis and Decline In Orga­ni­za­tions — in this blog before (a while ago, but still a pop­u­lar post­ing), and wanted to con­sider the sub­ject on a larger level. With the rapid spread of social soft­ware / social media and the rise of com­plex social dynam­ics in on-line envi­ron­ments, explor­ing fail­ure at the level of an entire soci­ety is timely.
In The Fish­bowl
Failed or fail­ing soci­eties are an excel­lent fish­bowl for observers seek­ing pat­terns related to social media, for two rea­sons. First, the high inten­sity of fail­ure sit­u­a­tions reveals much of what is ordi­nar­ily hid­den in social struc­tures and pat­terns: Impend­ing col­lapse leads peo­ple to dis­pense with care­fully main­tained social con­struc­tions.
One source of this height­ened inten­sity is the greatly increased stakes of soci­etal fail­ure (vs. most other kinds), which often means sud­den and dra­matic dis­rup­tions to basic liv­ing and eco­nomic pat­terns, the decline of cities and urban con­cen­tra­tions, and dra­matic pop­u­la­tion decrease. Another source is the very broad scope of the after­ef­fects; because a fail­ing soci­ety involves an entire cul­ture, the affects are com­pre­hen­sive, touch­ing every­one and every­thing.
Sec­ondly, soci­eties often com­mand sub­stan­tial qual­i­ta­tive and quan­ti­ta­tive resources that can help them man­age cri­sis or chal­lenges, thereby avert­ing fail­ure. Smaller, less sophis­ti­cated enti­ties lack the resource base of a com­plex social organ­ism, and con­se­quently can­not put up as much of a fight.
Exam­ples of resources avail­able at the level of a soci­ety include:

  • Lead­ers and plan­ners ded­i­cated to focus­ing on the future
  • Large amounts of accu­mu­lated knowl­edge and experience
  • Sophis­ti­cated struc­tures for deci­sion mak­ing and control
  • Mech­a­nisms for main­tain­ing order dur­ing crises
  • Col­lec­tive resilience from sur­viv­ing pre­vi­ous challenges
  • Sub­stan­tial stores of resources such as food and mate­ri­als, money, land
  • Tools, meth­ods, and orga­ni­za­tions pro­vid­ing economies of scale, such as bank­ing and com­merce networks
  • Sys­tems for mobi­liz­ing labor for spe­cial purposes
  • Con­nec­tions to other soci­eties that could pro­vide assis­tance (or poten­tial rescue)

Despite these mit­i­gat­ing resources, the his­tor­i­cal and arche­o­log­i­cal records over­flow with exam­ples of failed soci­eties. Once we read those records, the ques­tion of how these soci­eties defined them­selves seems to bear directly on quite a few of the out­comes.
I dis­cuss three soci­eties in the arti­cle: Easter Island, Tikopia, and my own small startup com­pany. We have insight into the fate of Easter Island soci­ety thanks to a rich arche­o­log­i­cal record that has been exten­sively stud­ied, and descrip­tions of the Rapa Nui soci­ety in writ­ten records kept by Euro­pean explor­ers vis­it­ing since 1722. Tikopia of course is still a func­tion­ing cul­ture. My startup was a tiny affair that serves as a use­ful foil because it shows all the mis­takes soci­eties make in a com­pressed span of time, and on a scale that’s easy to exam­ine. The Norse colonies in North Amer­ica and Green­land are another good exam­ple, though space con­straints didn’t allow dis­cus­sion of their failed soci­ety in the arti­cle.
Read the arti­cle to see what hap­pens to all three!
Semi Ran­dom Assort­ment of Quo­ta­tions
In the mean­time, enjoy this sam­pling of quo­ta­tions about fail­ure, knowl­edge, and self, from some well-known — and mostly suc­cess­ful! — peo­ple.
“Tech­no­log­i­cal change is like an axe in the hands of a patho­log­i­cal crim­i­nal.” — ALBERT EINSTEIN
“It is not the strongest of the species that sur­vives, nor the most intel­li­gent, but the one most respon­sive to change.” — CHARLES DARWIN
“It is impos­si­ble for a man to learn what he thinks he already knows.” — EPICTETUS
“I have not failed. I’ve just found 10,000 ways that won’t work.” — THOMAS EDISON
“It is on our fail­ures that we base a new and dif­fer­ent and bet­ter suc­cess.” — HAVELOCK ELLIS
“Life is a process of becom­ing, a com­bi­na­tion of states we have to go through. Where peo­ple fail is that they wish to elect a state and remain in it.” — ANAIS NIN
“We read the world wrong and say that it deceives us.” — RABINDRANATH TAGORE
“Who­ever longs to res­cue quickly both him­self and oth­ers should prac­tice the supreme mys­tery: exchange of self and other.” — SHANTIDEVA
“Fail­ure is instruc­tive. The per­son who really thinks learns quite as much from his fail­ures as from his suc­cesses.” — JOHN DEWEY

2 comments » | architecture, Ideas, The Media Environment

Who Should Own How We Work? Collaboration, the New Enterprise Application

May 14th, 2006 — 11:55pm

Col­lab­o­ra­tion is the lat­est ral­ly­ing cry of soft­ware ven­dors hop­ing to embed new gen­er­a­tions of enter­prise class tools and user expe­ri­ences into the fab­ric of the mod­ern work­place. Microsoft, IBM, and other firms expect that con­trol or lead­er­ship in the mar­ket for col­lab­o­ra­tion, whether by own­ing the archi­tec­ture, sys­tems, or other solu­tion com­po­nents, will be lucra­tive. A recent Rad­i­cati Group study (qual­ity uncon­firmed…) of the mar­ket size for enter­prise col­lab­o­ra­tion offered an esti­mate of $1.6 bil­lion now, grow­ing 10% annu­ally to $2.3 bil­lion in 2010.
Beyond the sub­stan­tial money to be made cre­at­ing, sell­ing, installing, and ser­vic­ing col­lab­o­ra­tion solu­tions lies the strate­gic advan­tage of mar­ket def­i­n­i­tion. The vendor(s) that own(s) the col­lab­o­ra­tion space expect(s) to become an inte­gral to the knowl­edge economy’s sup­port­ing envi­ron­ment in the same way that Ford and Gen­eral Motors became essen­tial to the sub­ur­ban­ized con­sumer archi­tec­tures of the post WWII era by serv­ing simul­ta­ne­ously as employ­ers, man­u­fac­tur­ers, cul­tural mar­keters, cap­i­tal reser­voirs, and auto­mo­bile sell­ers. Col­lab­o­ra­tion ven­dors know that achiev­ing any level of indis­pen­si­bil­ity will enhance their longevity by mak­ing them a neces­sity within the knowl­edge econ­omy.
It’s worth tak­ing a moment to call atten­tion to the impli­ca­tions: by defin­ing the user expe­ri­ences and tech­no­log­i­cal build­ing blocks brought together to real­ize col­lab­o­ra­tion in large enter­prises, these ven­dors will directly shape our basic con­cepts and under­stand­ing (our men­tal mod­els and cog­ni­tive frames) of col­lab­o­ra­tion. Once embed­ded, these archi­tec­tures, sys­tems, and busi­ness processes, and the social struc­tures and con­cep­tual mod­els cre­ated in response, will in large part define the (infor­ma­tion) work­ing envi­ron­ments of the future.
And yes, this is exactly what these ven­dors aspire to achieve; the Microsoft Share­point Prod­ucts and Tech­nolo­gies Devel­op­ment Team blog, offers:
“Share­Point Prod­ucts and Tech­nolo­gies have become a key part of our strat­egy for deliv­er­ing a com­plete work­ing envi­ron­ment for infor­ma­tion work­ers, where they can col­lab­o­rate together, share infor­ma­tion with oth­ers, and find infor­ma­tion and peo­ple that can help them solve their busi­ness prob­lems.“
And IBM’s mar­ket­ing is not pitched and deliv­ered in a man­ner as sweep­ing, but the impli­ca­tions are sim­i­lar, as in the overview IBM® Work­place™: Sim­ply a bet­ter way]:
IBM Work­place™ Solu­tions are role-based frame­works to help cus­tomers apply IBM Work­place tech­nolo­gies faster and more pro­duc­tively… These solu­tions are designed to pro­vide ‘short-cuts’ for cre­at­ing a high per­for­mance role-based work envi­ron­ment, help­ing to accel­er­ate time-to-value.“
The Mod­els for com­mu­ni­ca­tion and rela­tion­ships built into our tools are very pow­er­ful, and often employed in other spheres of life. How many times have you started writ­ing a birth­day card for a friend, and found your­self instinc­tively com­pos­ing a set of bul­let points list­ing this person’s chief virtues, notable char­ac­ter traits, and the most impor­tant / amus­ing moments of your friend­ship. The creep­ing ubiq­uity of the rhetor­i­cal style of Pow­er­point (Tufte’s essay here) is just one exam­ple of the tremen­dous social impact of a habit­u­ated model of com­mu­nica­tive prac­tices that’s run amok.
What does the future hold, in terms of enter­prise ven­dor con­trol over every­day work­ing expe­ri­ences? I’ve writ­ten before on the idea that the days of the mono­lithic enter­prise sys­tems are num­bered, mak­ing the point along the way that these behe­moths are the result of a top-down, one-size-for-all approach. I think the same is true of the cur­rent approach to col­lab­o­ra­tion solu­tions and work­ing envi­ron­ments. And so I was happy to see Andrew McAfee of Har­vard Busi­ness School make sev­eral strong points about how enter­prise col­lab­o­ra­tion efforts will real­ize greater suc­cess by *reduc­ing* the amount of struc­ture imposed on their major ele­ments — roles, work­flows, arti­facts, and rela­tion­ships — in advance of actual use.
McAfee sees con­sid­er­able ben­e­fit in new approaches to enter­prise IT invest­ment and man­age­ment that reduce the top-down and imposed nature of enter­prise envi­ron­ments and solu­tions, in favor of emer­gent struc­tures cre­ated by the peo­ple who must work suc­cess­fully within them. McAfee advo­cates allow­ing staff to cre­ate the iden­ti­ties, struc­tures and pat­terns that will orga­nize and gov­ern their col­lab­o­ra­tion envi­ron­ments as nec­es­sary, in an emer­gent fash­ion, instead of fix­ing these aspects long before users begin to col­lab­o­rate.
McAfee says:
“When I look at a lot of cor­po­rate col­lab­o­ra­tion tech­nolo­gies after spend­ing time at Wikipedia, del.icio.us, Flickr, and Blog­ger I am struck by how reg­i­mented, inflex­i­ble, and lim­ited the cor­po­rate stuff seems, because it does some or all of the following:

  • Gives users iden­ti­ties before they start using the tech­nol­ogy. These iden­ti­ties assign them cer­tain roles, priv­i­leges, and access rights, and exclude them from oth­ers. These iden­ti­ties almost always also place them within the exist­ing orga­ni­za­tional struc­ture and for­mal cor­po­rate hierarchy.
  • Con­tains few truly blank pages. Instead, it has lots of templates–for meet­ings, for project track­ing, for doc­u­ments and reports, etc.
  • Has tons of explicit or implicit work­flow– seqences [sic] of tasks that must be exe­cuted in order.

How much of this struc­ture is nec­es­sary? How much is valu­able? Well, the clear suc­cess sto­ries of Web 2.0 demon­strate that for at least some types of com­mu­nity and col­lab­o­ra­tion, none of it is.“
The crit­i­cal ques­tion is then “what types of com­mu­nity and col­lab­o­ra­tion require which approaches to cre­at­ing struc­ture, and when?” As any­one who’s used a poorly or overly struc­tured col­lab­o­ra­tion (or other enter­prise) tool knows, the result­ing envi­ron­ment is often anal­o­gous to a feu­dal soci­ety designed and man­aged by crypto-technical over­lords; one in which most users feel as if they are serfs bound to the land for in per­pe­tu­ity in order to sup­port the leisure-time and war-making indul­gences of a small class of share­hold­ing nobil­ity.
Answer­ing these ques­tions with con­fi­dence based on expe­ri­ence will likely take time in the range of years, and require numer­ous failed exper­i­ments. There’s a larger con­text to take into account: the strug­gle of enter­prise soft­ware ven­dors to extend their reach and longevity by dom­i­nat­ing the lan­guage of col­lab­o­ra­tion and the range of offer­ings is one part of a much broader effort by soci­ety to under­stand dra­matic shifts in our ways of work­ing, and the social struc­tures that are both dri­ven by and shape these new ways of work­ing. And so there are sev­eral impor­tant ideas and ques­tions under­ly­ing McAfee’s assess­ment that social sys­tem design­ers should under­stand.
One of the most impor­tant is that the notion of “col­lab­o­ra­tion” is con­cep­tual short­hand for how you work, who you work with, and what you do. In other words, it’s a dis­til­la­tion of your pro­fes­sional iden­tity. Your role in a col­lab­o­ra­tion envi­ron­ment defines who you are within that envi­ron­ment.
More impor­tantly, from the per­spec­tive of growth and devel­op­ment, your sys­tem assigned role deter­mines who you can *become*. Knowl­edge work­ers are val­ued for their skills, expe­ri­ence, pro­fes­sional net­works, pub­lic rep­u­ta­tions, and many other fluid, con­text depen­dent attrib­utes. And so lock­ing down their iden­ti­ties in advance strips them of a sub­stan­tial pro­por­tion of their cur­rent value, and simul­ta­ne­ously reduces their abil­ity to adapt, inno­vate, and respond to envi­ron­men­tal changes by shift­ing their think­ing or prac­tices. In plain terms, deter­min­ing their iden­ti­ties in advance pre­cludes the cre­ation of future value.
Another impor­tant under­ly­ing idea is the impor­tance of prop­erly under­stand­ing the value and util­ity of dif­fer­ing approaches to sys­tem­ati­za­tion in dif­fer­ing con­texts. McAfee’s assess­ment of the unhealthy con­se­quences of impos­ing too much struc­ture in advance is use­ful for social sys­tem design­ers (such as infor­ma­tion archi­tects and knowl­edge man­agers), because it makes the out­comes of implicit design strate­gies and assump­tions clear and tan­gi­ble, in terms of the neg­a­tive effects on the even­tual users of the col­lab­o­ra­tion envi­ron­ment. For com­plex and evolv­ing group set­tings like the mod­ern enter­prise, cre­at­ing too much struc­ture in advance points to a mis­placed under­stand­ing of the value and role of design and archi­tec­ture.
Fun­da­men­tally, it indi­cates an over­es­ti­ma­tion of the value of the activ­ity of sys­tem­atiz­ing (design­ing) col­lab­o­ra­tion envi­ron­ments to high lev­els of detail, and with­out recog­ni­tion for evo­lu­tion­ary dynam­ics. The design or struc­ture of any col­lab­o­ra­tion envi­ron­ment — of any social sys­tem — is only valu­able for how well it encour­ages rela­tion­ships and activ­ity which advance the goals of the orga­ni­za­tion and it’s mem­bers. The value of a designer in the effort to cre­ate a col­lab­o­ra­tive com­mu­nity lies in the abil­ity to cre­ate designs that lead to effec­tive col­lab­o­ra­tion, not in the num­ber or speci­ficity of the designs they pro­duce, and espe­cially not in the arti­facts cre­ated dur­ing design — the tem­plates, work­flows, roles, and other McAfee men­tioned above. To sim­plify the dif­fer­ent views of what’s appro­pri­ate into two arti­fi­cially seg­mented camps, the [older] view that results in the pre­ma­ture cre­ation of too much struc­ture val­i­dates the design of things / arti­facts / sta­tic assem­blies, whereas the newer view valu­ing min­i­mal and emer­gent struc­tures acknowl­edges the greater effi­cacy of design­ing dynamic sys­tems / flows / frame­works.
The overly spe­cific and rigid design of many col­lab­o­ra­tion sys­tem com­po­nents com­ing from the older design view­point in fact says much about how large, com­plex enter­prises choose to inter­pret their own char­ac­ters, and cre­ate tools accord­ingly. Too often, a desire to achieve total­ity lies at the heart of this approach.
Of course, most total­i­ties only make sense — exhibit coher­ence — when viewed from within, and when using the lan­guage and con­cepts of the total­ity itself. The result is that attempts to achieve total­ity of design for many com­plex con­texts (like col­lab­o­ra­tion within enter­prises large or small) rep­re­sent a self-defeating approach. That the approach is self-defeating is gen­er­ally ignored, because the pur­suit of total­ity is a self-serving exer­cise in power val­i­da­tion, that ben­e­fits power hold­ers by con­sum­ing resources poten­tially used for other pur­poses, for exam­ple, to under­mine their power.
With the chimera of total­ity set in proper con­text, it’s pos­si­ble to see how col­lab­o­ra­tion envi­ron­ments — at least in their most poorly con­ceived man­i­fes­ta­tions — will resem­ble vir­tual retreads of Tay­lorism, wherein the real accom­plish­ment is to jus­tify the effort and expense involved in cre­at­ing the sys­tem by point­ing at an exces­sive quan­tity of pre­de­ter­mined struc­ture await­ing habi­ta­tion and use by dis­en­fran­chised staff.
At present, I see two diver­gent and com­pet­ing trends in the realm of enter­prise solu­tions and user expe­ri­ences. The first trend is toward homo­gene­ity of the work­ing envi­ron­ment with large amounts of struc­ture imposed in advance, exem­pli­fied by com­pre­hen­sive col­lab­o­ra­tion suites and archi­tec­tures such as MSOf­fice / Share­point, or IBM’s Work­place.
The sec­ond trend is toward het­ero­gene­ity in the struc­tures inform­ing the work­ing envi­ron­ment, vis­i­ble as vari­able pat­terns and locuses of col­lab­o­ra­tion estab­lished by fluid groups that rely on adhoc assort­ment of tools from dif­fer­ent sources (Base­Camp, GMail, social book­mark­ing ser­vices, RSS syn­di­ca­tion of social media struc­tures, com­mu­ni­ties of prac­tice, busi­ness ser­vices from ASP providers, open source appli­ca­tions, etc.).
But this itself is a short term view, when sit­u­a­tion within a longer term con­text is nec­es­sary. It is com­mon for sys­tems or envi­ron­ments of all sizes and com­plex­i­ties to oscil­late cycli­cally from greater to lesser degrees of struc­ture, along a con­tin­uüm rang­ing from homo­ge­neous to het­ero­ge­neous. In the short term view then, the quest for total­ity equates to homo­gene­ity, or even efforts at dom­i­na­tion. In the long term view, how­ever, the quest for total­ity could indi­cate an imma­ture ecosys­tem that is not diverse, but may become so in time.
Apply­ing two (poten­tial) lessons from ecol­ogy — the value of diver­sity as an enhancer of over­all resilience in sys­tems, and the ten­dency of mono­cul­tures to exhibit high fragility — to McAfee’s points on emer­gence, as well as the con­tin­uüm view of shift­ing degress of homo­gene­ity, should tell us that col­lab­o­ra­tion solu­tion design­ers would be wise to do three things:

  1. Adopt the new design view­point and focus on design­ing struc­tures that allow col­lab­o­ra­tors to cre­ate value
  2. Spec­ify as lit­tle struc­ture of any kind in advance as possible
  3. Antic­i­pate the emer­gence of new archi­tec­tural ele­ments, and allow for their incor­po­ra­tion under the guid­ance of the com­mu­nity of collaborators

The end result should be an enter­prise approach to col­lab­o­ra­tion that empha­sizes the design of infra­struc­ture for com­mu­ni­ties that cre­ate their own struc­tures. Big ven­dors be wary of this enlight­ened point of view, unless you’re will­ing to respond in kind.

4 comments » | Ideas

New Amazon Features: Translating the Bookstore Experience On-line

January 12th, 2006 — 4:08pm

Ama­zon is offer­ing new Text Stats on “Read­abil­ity” and “Com­plex­ity”, and a Con­cor­dance fea­ture, both part of their com­pre­hen­sive effort to trans­late the phys­i­cal book[store] expe­ri­ence into the online medium. The new fea­tures build on exist­ing capa­bil­i­ties such as Look Inside, Wish­lists, Rec­om­men­da­tions, Edi­to­r­ial and Cus­tomer Reviews, Cita­tions, and Bet­ter Together to cre­ate a com­pre­hen­sive book buy­ing expe­ri­ence. In the same way that book­stores include kiosks to allow cus­tomers access meta­data and other infor­ma­tion on the books for sale in the imme­di­ate envi­ron­ment, Ama­zon is offer­ing on-line capa­bil­i­ties that sim­u­late many of the activ­i­ties of book buy­ers in a book­store, such as check­ing the table of con­tents and indexes, flip­ping through a book to read pas­sages, or look at select pages.
The new fea­tures appear on prod­uct pages for books, as well as other kinds of works. [Try this intro to FRBR for a look at the con­cep­tual hier­ar­chy dif­fer­en­ti­at­ing works from items, and it’s impli­ca­tions for com­mon user tasks like find­ing, iden­ti­fy­ing, select­ing, and obtain­ing items.]
Text Stats may be exper­i­men­tal, but it’s hard to feel com­fort­able with their def­i­n­i­tion of com­plex­ity, which is: “A word is con­sid­ered “com­plex” if it has three or more syl­la­bles.” To point out the obvi­ous, Eng­lish includes plenty of sim­ple three syl­la­ble words — like “banana” — and some very com­plex one syl­la­ble words — “time” “thought” and “self” for exam­ple.
The Text Stats on Read­abil­ity seem a bit bet­ter thought through. That’s nat­ural, given their ground­ing in research done out­side Amazon’s walls. But with clear evi­dence that US edu­ca­tion stan­dards vary con­sid­er­ably across states and even indi­vid­ual dis­tricts, and also evi­dence that those stan­dards change over time, I have to ques­tion the value of Read­abil­ity stats long term. I sup­pose that isn’t point…
The Con­cor­dance fea­ture is eas­ier to appre­ci­ate; per­haps it doesn’t attempt to inter­peret or pro­vide mean­ing. It sim­ply presents the raw sta­tis­ti­cal data on word fre­quen­cies, and allows you to do the inter­pre­ta­tion. Ama­zon links each word in the con­cor­dance to a search results page list­ing the indi­vid­ual occur­ances of the word in the text, which is use­ful, and then fur­ther links the indi­vid­ual occru­ance list­ings to the loca­tion within the text.
With this strong and grow­ing mix of fea­tures, Ama­zon both trans­lates the book­store expe­ri­ence on-line, and also aug­ments that expe­ri­ence with capa­bil­i­ties avail­able only in an infor­ma­tion envi­ron­ment. The ques­tion is whether Ama­zon will con­tinue to expand the capa­bil­i­ties it offers for book buy­ing under the basic men­tal model of “being in a book­store”, or if a new direc­tion is ahead?
Here’s a screen­shot of the Text Stats for DJ Spooky’s Rhythm Sci­ence.
Text Stats:

Here’s a screen shot of the Con­cor­dance fea­ture.

Comment » | Modeling, User Experience (UX)

Mental Models and the Semantics of Disaster

November 4th, 2005 — 3:47pm

A few months ago, I put up a post­ing on Men­tal Mod­els Lotus Notes, and Resililence. It focused on my chronic inabil­ity to learn how not to send email with Lous Notes. I posted about Notes, but what led me to explore resilience in the con­text of men­tal mod­els was the sur­pris­ing lack of acknowl­edge­ment of the scale of hur­ri­cane Kat­rina I came across at the time. For exam­ple, the day the lev­ees failed, the front page of the New York Times dig­i­tal edi­tion car­ried a gigan­tic head­line say­ing ‘Lev­ees Fail! New Orleans floods!’. And yet no one in the office at the time even men­tioned what hap­pened.
My con­clu­sion was that peo­ple were sim­ply unable to accept the idea that a major met­ro­pol­i­tan area in the U.S. could pos­si­bly be the set­ting for such a tragedy, and so they refused to absorb it — because it didn’t fit in with their men­tal mod­els for how the world works. Today, I came across a Resilience Sci­ence post­ing titled New Orleans and Dis­as­ter Soci­ol­ogy that sup­ports this line of think­ing, while it dis­cusses some of the inter­est­ing ways that seman­tics and men­tal mod­els come into play in rela­tion to dis­as­ters.
Quot­ing exten­sively from an arti­cle in The Chron­i­cle of Higher Edu­ca­tion titled Dis­as­ter Soci­ol­o­gists Study What Went Wrong in the Response to the Hur­ri­canes, but Will Pol­icy Mak­ers Lis­ten? the post­ing calls out how nar­row slices of media cov­er­age dri­ven by blurred seman­tic and con­tex­tual under­stand­ings, inac­cu­rately frame social responses to dis­as­ter sit­u­a­tions in terms of group panic and the implied break­down of order and soci­ety.
“The false idea of post­dis­as­ter panic grows partly from sim­ple seman­tic con­fu­sion, said Michael K. Lin­dell, a psy­chol­o­gist who directs the Haz­ard Reduc­tion and Recov­ery Cen­ter at Texas A&M Uni­ver­sity at Col­lege Sta­tion. ‘A reporter will stick a micro­phone in someone’s face and ask, ‘Well, what did you do when the explo­sion went off?’ And the per­son will answer, ‘I pan­icked.’ And then they’ll pro­ceed to describe a very log­i­cal, ratio­nal action in which they pro­tected them­selves and looked out for peo­ple around them. What they mean by ‘panic’ is just ‘I got very fright­ened.’ But when you say ‘I pan­icked,’ it rein­forces this idea that there’s a thin veneer of civ­i­liza­tion, which van­ishes after a dis­as­ter, and that you need out­side author­i­ties and the mil­i­tary to restore order. But really, peo­ple usu­ally do very well for them­selves, thank you.‘
Men­tal mod­els come into play when the arti­cle goes on to talk about the ways that the emer­gency man­age­ment agen­cies are orga­nized and struc­tured, and how they approach and under­stand sit­u­a­tions by default. With the new Home­land Secu­rity par­a­digm, all inci­dents require com­mand and con­trol approaches that assume a ded­i­cated and intel­li­gent enemy — obvi­ously not the way to man­age a hur­ri­cane response.
“Mr. Lin­dell, of Texas A&M, agreed, say­ing he feared that pol­icy mak­ers in Wash­ing­ton had taken the wrong lessons from Kat­rina. The employ­ees of the Depart­ment of Home­land Secu­rity, he said, ‘are mostly drawn from the Depart­ment of Defense, the Depart­ment of Jus­tice, and from police depart­ments. They’re firmly com­mit­ted to a command-and-control model.’ (Just a few days ago, Pres­i­dent Bush may have pushed the process one step fur­ther: He sug­gested that the Depart­ment of Defense take con­trol of relief efforts after major nat­ural dis­as­ters.)
“The habits of mind cul­ti­vated by mil­i­tary and law-enforcement per­son­nel have their virtues, Mr. Lin­dell said, but they don’t always fit dis­as­ter sit­u­a­tions. ‘They come from orga­ni­za­tions where they’re deal­ing with an intel­li­gent adver­sary. So they want to keep infor­ma­tion secret; ‘it’s only shared on a need-to-know basis. But emer­gency man­agers and med­ical per­son­nel want infor­ma­tion shared as widely as pos­si­ble because they have to rely on per­sua­sion to get peo­ple to coöper­ate. The prob­lem with putting FEMA into the Depart­ment of Home­land Secu­rity is that it’s like an organ trans­plant. What we’ve seen over the past four years is basi­cally organ rejec­tion.‘
If I read this cor­rectly, mis­aligned orga­ni­za­tional cul­tures lie at the bot­tom of the whole prob­lem. I’m still curi­ous about the con­nec­tions between an organization’s cul­ture, and the men­tal mod­els that indi­vid­u­als use. Can a group have a col­lec­tive men­tal model?
Accoridng to Col­lec­tive Men­tal State and Indi­vid­ual Agency: Qual­i­ta­tive Fac­tors in Social Sci­ence Expla­na­tion it’s pos­si­ble, and in fact the whole idea of this col­lec­tive men­tal state is a black hole as far as qual­i­ta­tive social research and under­stand­ing are concerned.

2 comments » | Modeling, The Media Environment

Lotus Notes User Experience = Disease

September 22nd, 2005 — 10:13pm

Lotus Notes has one of the most unpleas­ant and unwel­com­ing User Expe­ri­ences this side of a medium-security prison where the war­den has aspi­ra­tions towards inte­rior design and art instruc­tion. One of the most painful aspects of the Notes expe­ri­ence is the default set­tings for font size and color in the email win­dow. The default font size (for Macs) is on the order of 7 point type, and the default color for unread mes­sages is — iron­i­cally — red. The com­bi­na­tion yields a user expe­ri­ence that resem­bles a bad skin rash. I call it “angry red microNotes” dis­ease, and it looks like this:
Over­all, it has an unhealthy affect on one’s state of mind. The under­tones of hos­til­ity and resent­ment run­ning through­out are man­i­fold. And nat­u­rally, it is impos­si­ble to change the default font size and color for the email reader. This is fur­ther con­fir­ma­tion for my the­ory that Notes has yet to escape it’s roots as a thick client for series of uncon­nected data­bases.
After three weeks of suf­fer­ing from angry red microNotes, I real­ized I was lit­er­ally going blind from squint­ing at the tiny type, and went to Google for relief. I found niniX 1.7, a util­ity that allows Mac based Lotus Notes users the abil­ity to edit the binary for­mat Notes pref­er­ences file, and change the font size of the email client. I share it in the hopes that oth­ers may break the chains that blind them. This will only solve half the prob­lem — if some­one can fig­ure out how to change the default color for unread mes­sages to some­thing besides skin rash red, I will hap­pily share with the rest of the suf­fer­ing masses (and appar­ently there are on the order of 118 mil­lion of us out there).
But will it always be this (hor­ri­ble) way?
In Beyond Notes 7.0: IBM Lotus sketches ‘Han­nover’ user expe­ri­ence Peter Bochner of SearchDomino.com says this of the next Notes release, “Notes has often been crit­i­cized for its some­what staid user inter­face. Accord­ing to IBM’s Bis­conti, in cre­at­ing Han­nover, IBM paid atten­tion “to not just the user inter­face, but the user expe­ri­ence.“
Okay… So does that mean I’ll have my choice of dis­eases as themes for the user expe­ri­ence of my col­lab­o­ra­tion envi­ron­ment?
Accord­ing to Ken Bis­conti, IBM Lotus vice pres­i­dent of Work­place, por­tal and col­lab­o­ra­tion prod­ucts, “Through improve­ments such as con­tex­tual col­lab­o­ra­tion and sup­port for com­pos­ite apps, we’ve gone above and beyond sim­ple UI enhance­ment”.
I think sim­ple UI enhance­ment is exactly what Ken and his team should focus on for the next sev­eral years, since they have so much oppor­tu­nity for improvement.

22 comments » | User Experience (UX)

Mental Models: Additional Reading

September 6th, 2005 — 1:58pm

Some addi­tional read­ing on men­tal mod­els, cour­tesy of the Inter­ac­tion Design Encyclopedia.

Comment » | Modeling

Mental Models, Resilience, and Lotus Notes

September 5th, 2005 — 6:05pm

Sev­eral very unpleas­ant expe­ri­ences I’ve had with the Lotus Notes web­mail client dur­ing the past few weeks have brought up some ques­tions about men­tal mod­els; specif­i­cally how users respond to chal­lenges to their men­tal mod­els, and how resilience plays a part in how changes to men­tal mod­els occur.
The IAWiki defines a men­tal model as, “a men­tal model is how the user thinks the prod­uct works.” This is a sim­pli­fied def­i­n­i­tion, but it’s ade­quate for the moment. For a deeper explo­ration, try Mar­tina Angela Sasse’s the­sis
Elic­it­ing and Describ­ing Users’ Mod­els of Com­puter Sys­tems.
In this case, the model and the chal­lenge are straight­for­ward. My men­tal model of the Notes web­mail client includes the under­stand­ing that it can send email mes­sages. The chal­lenge: the Lotus web­mail client can­not send email mes­sages — at least not as I expe­ri­ence it.
Here’s what hap­pens my men­tal model and my real­ity don’t match:

  1. I log in to my email client via Fire­fox — the only browser on the Mac that ren­ders the Notes web­mail client vaguely cor­rectly — (I’m using web­mail because the full Notes client requires VPN, mean­ing I’m unable to access any­thing on my local net­work, or the inter­net, which, inci­den­tally, makes it dif­fi­cult to seem like a cred­i­ble inter­net con­sul­tant.) again, because it’s frozen and crashed my browser in the past ten minutes.
  2. I real­ize I need to respond to an email
  3. I do not remem­ber that the Notes web­mail client is inca­pable of send­ing out email messages
  4. I open a new mes­sage win­dow, and com­pose a chunk of semi-grammatical techno-corporate non-speak to com­mu­ni­cate a few sim­ple points in blame-retardant consultantese
  5. I attempt to send this email
  6. I am con­fronted with a cryp­tic error mes­sage via javascript prompt, say­ing some­thing like “We’re really sorry, but Domino sucks, so you can’t send out any mes­sages using your email client.”
  7. Over the span of .376 sec­onds, I move through suc­ces­sive states of sur­prise, con­fu­sion, com­pre­hen­sion, frus­tra­tion, anger, resent­ment, res­ig­na­tion, and malaise (actu­ally, mailaise is more accurate.)
  8. I swear: silently if clients are within earshot, out loud if not
  9. I switch to gmail, cre­ate a new mes­sage, copy the text of my mes­sage from the Notes web­mail win­dow to Gmail, and send the mes­sage to some eagerly wait­ing recipient
  10. I close the Notes web­mail client, and return to busi­ness as usual.
  11. I for­get that the Notes web­mail client can­not send email messages.

Despite fol­low­ing this same path three times per day, five days each week, for the past five weeks, (for a total of ~75 clear exam­ples), I am always sur­prised when I can’t send a mes­sage. I’m no expert on Learn­ing the­ory but nei­ther lack of atten­tion nor stub­born­ness explain why seventy-five exam­ples aren’t enough to change my model of how Notes works.
Dis­ci­plines includ­ing sys­tems the­ory, biol­ogy, and soci­ol­ogy use a con­cept called resilience. In any sta­ble sys­tem, “Resilience gen­er­ally means the abil­ity to recover from some shock, insult, or dis­tur­bance.” From an eco­log­i­cal per­spec­tive, resilience “is a mea­sure of the amount of change or dis­rup­tion that is required to trans­form a sys­tem.” The psy­cho­log­i­cal view empha­sizes “the abil­ity of peo­ple to cope with stress and cat­a­stro­phe.“
Appar­ently, the resilience of my model for email clients is high enough to with­stand con­sid­er­able stress, since — in addi­tion to the ini­tial cat­a­stro­phe of using Notes itself — seventy-five con­sec­u­tive exam­ples of fail­ure to work as expected do not equal enough shock, insult, and dis­tur­bance to my model to lead to a change my in under­stand­ing.
Notice that I’m using a work-around — switch­ing to Gmail — to achieve my goal and send email. In
Resilience Man­age­ment in Social-ecological Sys­tems: a Work­ing Hypoth­e­sis for a Par­tic­i­pa­tory Approach , Brian Walker and sev­eral oth­ers refine the mean­ing of resilience to include, “The degree to which the sys­tem expresses capac­ity for learn­ing and adap­ta­tion.” This accounts nicely for the Gmail work-around.
I also noticed that I’m rely­ing on a series of assump­tions — email clients can send mes­sages; Notes is an email client; there­fore, Notes can send mes­sages — that make it log­i­cal to use a well estab­lished model for email clients in gen­eral to antic­i­pate the work­ings of Notes web­mail in par­tic­u­lar. In new con­texts, it’s eas­ier to bor­row an exist­ing model than develop a new one. In short order, I expect I’ll change one of the assump­tions, or build a model for Notes web­mail.
Here’s a few ques­tions that come to mind:

  1. What fac­tors deter­mine the resilience of a men­tal model?
  2. How to mea­sure resiliency in men­tal models?
  3. What’s the thresh­old of recov­ery for a men­tal model?
  4. Put another way, what’s required to change a men­tal model?

Based on a quick review of the con­cept of resilience from sev­eral per­spec­tives, I’m com­fort­able say­ing it’s a valu­able way of look­ing at men­tal mod­els, with prac­ti­cal impli­ca­tions for infor­ma­tion archi­tects.
Some of those impli­ca­tions are:

  1. Under­stand the rel­e­vance of exist­ing men­tal mod­els when design­ing new systems
  2. Antic­i­pate and plan the ways that users will form a men­tal model of the system
  3. Use design at mul­ti­ple lev­els to fur­ther the for­ma­tion of men­tal models
  4. Under­stand thresh­olds and resilience fac­tors when chal­leng­ing exist­ing men­tal models

From a broader view, I think it’s safe to say the appli­ca­tion of sys­tems the­ory to infor­ma­tion archi­tec­ture con­sti­tutes an impor­tant area for explo­ration, one con­tain­ing chal­lenges and oppor­tu­ni­ties for user expe­ri­ence prac­ti­tion­ers in gen­eral, and infor­ma­tion archi­tects in par­tic­u­lar.
Time to close this post before it gets too long.
Fur­ther read­ing:
Bio of Lud­wig Berta­lanffy, impor­tant con­trib­u­tor to Gen­eral Sys­tem The­ory.
Doug Cocks <a href=“http://www.labshop.com.au/dougcocks/castalk.html> On Decon­struct­ing Com­plex Adap­tive Systems
Resilience Alliance
Garry Peterson’s blog Resilience Sci­ence

5 comments » | Modeling, User Experience (UX)

Three Contexts for the Term "Portal"

June 27th, 2005 — 3:37pm

I’m work­ing on a por­tal project at the moment for a health­care client, so I’ve heard a great deal about how the con­cept of ‘por­tal’ is so diluted as to be effec­tively mean­ing­less. Fol­low­ing a series of sur­pris­ingly mud­dled con­ver­sa­tions with tech­nol­o­gists, busi­ness types, and end users rep­re­sen­ta­tives around the con­cept for this new por­tal, I real­ized that much of the hand-wringing and con­fu­sion comes from sim­ple lack of per­spec­tive — on the dif­fer­ent per­spec­tives rep­re­sented by each view­point. Ambi­gu­ity or dis­agree­ment about which per­spec­tive is the frame of ref­er­ence in any given dis­cus­sion is the biggest source of the con­fu­sion and fric­tion that makes these projects need­lessly dif­fi­cult.
There are (at least) three dif­fer­ent per­spec­tives on the mean­ing of the term por­tal.
To tech­nol­o­gists and sys­tem devel­op­ers, a por­tal is a type of solu­tion deliv­ery plat­form with stan­dard com­po­nents like authen­ti­ca­tion, an appli­ca­tion server, inte­gra­tion ser­vices, and busi­ness logic and pre­sen­ta­tion lay­ers that is gen­er­ally pur­chased from a ven­dor and then cus­tomized to meet spe­cific needs. Exam­ples are Plumtree, BEA, IBM, etc.
To users, a por­tal is a sin­gle des­ti­na­tion where it’s pos­si­ble to obtain a con­ve­nient and — likely, though not always — per­son­al­ized com­bi­na­tion of infor­ma­tion and tools from many dif­fer­ent sources. Some exam­ples of this sense of the term include Yahoo, MSN, and a well-developed intranet.
To a busi­ness, a por­tal is a bounded vehi­cle for aggre­gat­ing infor­ma­tion and tools to address diverse con­stituent needs in a coör­di­nated and coher­ent way, with low­ered man­age­ment and admin­is­tra­tion costs real­ized via frame­work fea­tures like per­son­al­iza­tion, cus­tomiza­tion, and role-based con­fig­u­ra­tion.
One case where all three of these frames of ref­er­ence inter­sect is with Exec­u­tive Dash­board projects. A dash­board is a por­tal in all three of these senses (unless it hap­pens to rest on a dif­fer­ent archi­tec­ture / tech­nol­ogy stack, in which case I main­tain that it’s some­thing else, so as an IA it’s pru­dent to keep in mind the dif­fer­ing impli­ca­tions and assump­tions asso­ci­ated with each per­spec­tive while deal­ing with their representatives.

1 comment » | Building Blocks, Dashboards & Portals, Information Architecture

Back to top