Play, story and recombination

A bunch of Lego bricks

Dom­i­nant mod­els in IA: space + sto­ry” was one of the notes I took while at this year’s Euro IA Sum­mit. I’ll get into space some oth­er time. Con­cern­ing sto­ry: Basi­cal­ly it strikes me that for a dis­ci­pline involved with an inter­ac­tive medi­um, so often design­ing is likened to sto­ry­telling. I’m not sure this is always the most pro­duc­tive way to approach design, I actu­al­ly think it is very lim­it­ing. If you approach design not as embed­ding your sto­ry in the envi­ron­ment, but as cre­at­ing an envi­ron­ment where­in users can cre­ate their own sto­ries, then I’d say you’re on the right track. An exam­ple I tend to use is a game of pok­er: The design of the game pok­er was cer­tain­ly not an act of sto­ry­telling, but a play ses­sion of pok­er is expe­ri­enced as (and can be retold as) a sto­ry. Fur­ther­more, the com­po­nents of the game can be recom­bined to cre­ate dif­fer­ent vari­a­tions of the basic game, each cre­at­ing dif­fer­ent poten­tials for sto­ries to arise. I’d like to see more design­ers approach inter­ac­tive media (dig­i­tal, phys­i­cal or what­ev­er) like this: Don’t tell a sto­ry to your user, enable them to cre­ate their own.1 Real­ize users will want to recom­bine your stuff with oth­er stuff you might not know about (the notion of seam­ful design comes into play here). When you’ve done a prop­er job, you’ll find them retelling those sto­ries to oth­ers, which I would say is the biggest com­pli­ment you can get.

1. Or to put this in Marc LeBlanc’s terms: Don’t embed nar­ra­tive, let it emerge through play.

Summary of my Playful IAs argument

I thought I’d post a short sum­ma­ry of the argu­ment I made in my Euro IA Sum­mit 2007 talk, for those who weren’t there and/or are too lazy to actu­al­ly go through the notes in the slides. The pre­sen­ta­tion is basi­cal­ly bro­ken up into three parts: 

  1. Future web envi­ron­ments are becom­ing so com­plex, they start to show emer­gent prop­er­ties. In this con­text a lot of tra­di­tion­al IA prac­tice does­n’t make sense any­more. Instead of direct­ly design­ing an infor­ma­tion space, you’re bet­ter off design­ing the rules that under­ly the gen­er­a­tive con­struc­tion of such spaces. In oth­er words, IA is becom­ing a sec­ond order design problem.
  2. IAs tend to argue for the val­ue of their designs based sole­ly on how well they sup­port users in achiev­ing their end goals. I pro­pose sup­port­ing expe­ri­ence goals is just as impor­tant. From there I try to make the case that any pow­er­ful expe­ri­ence is a play­ful one, where the user’s fun fol­lows from the feel­ing that he or she is learn­ing new stuff, is kick­ing ass, is in flow.
  3. Game design is not black mag­ic (any­more). In recent years a lot has become under­stood about how games work. They are built up out of game mechan­ics that each fol­low a pat­tern of action, sim­u­la­tion, feed­back and mod­el­ling. Design­ing play­ful IAs means tak­ing care that you encour­age dis­cov­ery, sup­port explo­ration and pro­vide feed­back on mastery.

Get the the slides, and a list of sources for the talk in this ear­li­er post.

Playful IAs — slides for my Euro IA Summit 2007 talk

After a con­sid­er­able amount of fid­dling with SlideShare I’ve final­ly man­aged to upload a ver­sion of the slides that go with my Play­ful IAs pre­sen­ta­tion. This more or less as I pre­sent­ed it at the Euro IA Sum­mit 2007 and includes an approx­i­mate tran­script of my talk. I hope to get an audio/video record­ing of most of it in the near future as well. When I do I’ll update this page.

Update: I’ve post­ed a short sum­ma­ry of the cen­tral argu­ment of my talk.

Down­load a ver­sion includ­ing an approx­i­mate tran­script (14,5 MB).

I had some great reac­tions to this talk and I want to thank all the peo­ple who engaged with me in dis­cus­sions after­wards. It’s giv­en me a good pic­ture of what areas I should devel­op fur­ther in future sub­se­quent talks. I’m also pleas­ant­ly sur­prised to see that con­trary to what some peo­ple think, the IA com­mu­ni­ty (the Euro­pean one at least) is very much open to new ideas. That’s real­ly nice to expe­ri­ence firsthand.

A lot of peo­ple asked for a list of books and oth­er good sources on the top­ics I cov­ered. Here’s an incom­plete list of stuff I’ve used at some stage to inform my thinking:

If that does­n’t keep you busy for a while, you could always have a dig through my del.icio.us links. There’s plen­ty of good stuff there. Of course of if you ever find any­thing you think would be of inter­est to me, do let me know. Just tag it for:kaeru.

Mirroring mental models — games modelling players

Will Wright demoing Spore at TED 2007

Today I sent in the slides of my Euro IA Sum­mit pre­sen­ta­tion for the pro­ceed­ings. The rough out­line of my talk is done, the most impor­tant thing now is to find the prop­er exam­ples to illus­trate all the fuzzy the­o­ret­i­cal think­ing. That means (at least for me) doing a lot of Flickr pho­to search­es. This time I’ll also be exper­i­ment­ing with using some short video-clips. Games are bet­ter seen in motion after all (and best expe­ri­enced through play of course). Chron­i­cling my think­ing on the sub­ject of play­ful IAs on this blog has been very help­ful in organ­is­ing my thoughts by the way, I’ll def­i­nite­ly try it again the next time I need to do a talk.

On mental models

One idea I man­aged to squeeze into the pre­sen­ta­tion in addi­tion to the stuff I’ve been blog­ging about so far is about men­tal mod­els. I think it was Ben Cer­ve­ny who men­tioned in his Reboot 7.0 talk (MP3) that some of the plea­sure of play­ing games is derived from the grad­ual men­tal mod­el build­ing a play­er goes through. The play­er uses the visu­al lay­er of a game to learn about the under­ly­ing struc­tures. When a play­er mas­ters a game, the visu­al lay­er more or less fades away and becomes a sym­bol­ic land­scape through which he manip­u­lates a far rich­er mod­el of the game in his mind.

From a UX per­spec­tive because usu­al­ly when design­ing web sites and apps we try to adhere to exist­ing men­tal mod­els as much as pos­si­ble to pre­vent con­fu­sion and frus­tra­tion. This is a very valid approach of course. How­ev­er, regard­less of how well done the UX design, there will always be some men­tal mod­el­ling on the user’s part. Best make this as engag­ing as pos­si­ble I guess. This, again, is where games come in.

Will Wright acknowl­edges the fact that play­ers build mod­els of a game but he pro­pos­es to take it one step fur­ther. In an old(ish) talk at Accel­er­at­ing Change 2004 he pro­posed the idea that a game can con­struct a mod­el of the play­er as well. Par­al­lels with online rec­om­men­da­tion engines are appar­ent here. As Wright points out, in games (as in web envi­ron­ments) every­thing can be mea­sured. This way, the expe­ri­ence can be tai­lored to a player/user. He’s apply­ing this prin­ci­ple in the upcom­ing Spore, where game con­tent (cre­at­ed by oth­er play­ers) is dynam­i­cal­ly includ­ed based on inferred play­er preferences.

It can be argued that cer­tain web pro­fes­sion­als are way ahead of the games indus­try in this field. Per­haps there are some inter­est­ing oppor­tu­ni­ties for col­lab­o­ra­tion or career moves here?

The experience of playful IAs

Solving a Rubik's Cube

It’s time for a short update on my think­ing about Play­ful IAs (the top­ic of my Euro IA Sum­mit talk). One of the under-served aspects so far is the actu­al user expe­ri­ence of an archi­tec­ture that is playful.

Bri­an Sut­ton-Smith describes a mod­el describ­ing the ways in which games are expe­ri­enced in his book Toys as Cul­ture. I first came across this book in (not sur­pris­ing­ly) Rules of Play. He lists five aspects:

  1. Visu­al scanning
  2. Audi­to­ry discrimination
  3. Motor respons­es
  4. Con­cen­tra­tion
  5. Per­cep­tu­al pat­terns of learning

Of most impor­tance to my sub­ject is the 5th one. 

Game design, like the design of emer­gent IAs is a 2nd order design prob­lem. You can only shape the user’s expe­ri­ence indi­rect­ly. One of the most impor­tant sources of plea­sure for the user is the way you offer feed­back on the ways he or she has explored and dis­cov­ered the infor­ma­tion space. 

Obvi­ous­ly, I’m not say­ing you should make the use of your ser­vice delib­er­ate­ly hard. How­ev­er, what I am say­ing is that if you’re inter­est­ed in offer­ing a play­ful expe­ri­ence on the lev­el of IA, then Sut­ton-Smith’s per­cep­tu­al pat­terns of learn­ing is the best suit­ed expe­ri­en­tial dimension.

Learning about emergence from games

A game of Go

I’m still try­ing to get a grip on why I think games are such a good ref­er­ence point for IAs and IxDs. I’ll try to take anoth­er stab at it in this post. Pre­vi­ous­ly I wrote about how games might be a good way to ‘sell’ algo­rith­mic archi­tec­tures to your client. Even if you’re not active­ly push­ing your clients to adopt ideas such as on-the-fly cre­ation of site nav­i­ga­tion, soon­er or lat­er I’m con­vinced you’ll find your­self con­front­ed with a project where you’re not asked to devel­op a defin­i­tive infor­ma­tion archi­tec­ture. Instead you’ll be charged with the task to come up with mech­a­nisms to gen­er­ate these pro­ce­du­ral­ly. When this is this case, you’re tru­ly fac­ing a sec­ond-order design prob­lem. How can games help here? 

One of the defin­ing char­ac­ter­is­tics of games are their com­plex­i­ty. A few years ago Ben Cer­ve­ny gave a bril­liant talk on play (MP3) at Reboot 7.0 and men­tioned this specif­i­cal­ly — that much of the plea­sure derived from game-play is the result of the play­er com­ing to terms with com­plex pat­terns. This com­plex­i­ty is some­thing dif­fer­ent from pure ran­dom­ness and most cer­tain­ly dif­fer­ent from a ‘mere’ state machine. In oth­er words, games show emergence.

There are many exam­ples of emer­gent sys­tems. The Game of Life springs to mind. This sys­tem isn’t real­ly a game but shows a remark­able rich­ness in pat­terns, despite (or maybe because of) the fact that it is based on a set of decep­tive­ly sim­ple rules (which appar­ent­ly took its cre­ator, John Con­way, over 2 years to per­fect!) The thing is though, The Game of Life is not interactive. 

A won­der­ful exam­ple of a com­plex emer­gent sys­tem that is inter­ac­tive is the real game Go. It has a set of very sim­ple rules, but play­ing it well takes a huge amount of prac­tice. The joy of play­ing Go for me (an absolute begin­ner) is large­ly due to dis­cov­er­ing the many dif­fer­ent per­mu­ta­tions play can go through. 

So get­ting back to my ear­li­er remark: If you’re con­vinced you’ll need to get a bet­ter han­dle on solv­ing the sec­ond-order design prob­lems pre­sent­ed by the design of com­plex emer­gent sys­tems, games are an excel­lent place to start learn­ing. They are emer­gent first and inter­ac­tive sec­ond, the per­fect twin to the web envi­ron­ments we’ll be shap­ing in the future.

UX and the aesthetics of interactivity

Tetris cookies

I’ve been try­ing to reg­u­lar­ly post some thoughts on the top­ic of play­ful IA here. Pre­vi­ous­ly I blogged about how games could be a use­ful frame for think­ing about com­plex algo­rith­mic archi­tec­tures. Last week I post­ed some thoughts on the appli­ca­tion of game mechan­ics in web apps. There, Rahul was kind enough to point me to the fas­ci­nat­ing blog of ‘Danc’ Daniel Cook, titled Lost Gar­den, where there is one post in par­tic­u­lar that res­onates with my own pre-occu­pa­tions lately.

In ‘Short thoughts on games and inter­ac­tion design’ (which hon­est­ly isn’t that short) Danc Cook looks at some of the ways game design tech­niques can be applied to the inter­ac­tion design of web apps. In sum­ma­ry, accord­ing to Danc Cook game design tech­niques allow you to:

  1. Cre­ate an engag­ing expe­ri­ence that goes beyond sim­ply com­plet­ing a task efficiently.
  2. Sup­port free and deep explo­ration and intro­duce and teach new inter­ac­tions that vio­late conventions.

Some things you should­n’t bor­row from games with­out giv­ing it a lot of thought are:

  1. Spa­tial metaphors
  2. Visu­al themes

These are some of the things most peo­ple think of first as char­ac­ter­is­tic of games but real­ly, they are only sur­face, super­fi­cial, not deter­mi­nant of the actu­al inter­ac­tiv­i­ty of the system.

I think one of the great­est argu­ments for a deep­er under­stand­ing of games by inter­ac­tion design­ers, infor­ma­tion archi­tects and oth­er user expe­ri­ence spe­cial­ists is that they are the medi­um that is all about the aes­thet­ics of inter­ac­tiv­i­ty. It is true that they have no util­i­tar­i­an char­ac­ter, they aim to cre­ate a plea­sur­able expe­ri­ence through sys­tems of risks and rewards, restraints and free­doms, nest­ed feed­back loops and on and on. As a UX prac­ti­tion­er, it can nev­er hurt to have a deep appre­ci­a­tion of the aes­thet­ics of the medi­um you work in dai­ly (beyond sim­ply sup­port­ing user goals, or sell­ing prod­uct, or whatever).

Possibility spaces and algorithmic architectures

A screenshot of Sim City.

One of the con­cepts I plan on explor­ing in my talk at the Euro IA Sum­mit in Barcelona is ‘pos­si­bil­i­ty spaces’. It’s a term used by Will Wright to describe his view of what a game can be — a space that offers mul­ti­ple routes and out­comes to its explor­er. That idea maps nice­ly with one def­i­n­i­tion of play that Zim­mer­man and Salen offer in Rules of Play: ‘free move­ment with­in a rigid struc­ture’. Some exam­ples of pos­si­bil­i­ty spaces cre­at­ed by Wright are the well-known games Sim City and The Sims.

I think the idea of pos­si­bil­i­ty spaces can help IAs to get a firmer grip on ways to real­ize infor­ma­tion spaces that are mul­ti-dimen­sion­al and (to use a term put for­ward by Jesse James Gar­rett) algo­rith­mic. Algo­rith­mic archi­tec­tures accord­ing to Gar­rett are cre­at­ed ‘on the fly’ based on a set of rules (algo­rithms) that get their input (ide­al­ly) from user behav­iour. The exam­ple he uses to explain this con­cept is Ama­zon.

I’ve found myself in sev­er­al projects recent­ly that would have ben­e­fit­ed from an algo­rith­mic approach. The hard thing is to explain its charms to clients and to get a uni­fied vision of what it means across to the design team. I believe games might be a use­ful anal­o­gy. What do you think?

Interface design — fifth and final IA Summit 2007 theme

(Here’s the fifth and final post on the 2007 IA Sum­mit. You can find the first one that intro­duces the series and describes the first theme ‘tan­gi­ble’ here, the sec­ond one on ‘social’ here, the third one on ‘web of data’ here and the fourth one on ‘strat­e­gy’ here.)

It might have been the past RIA hype (which accord­ing to Jared Spool has noth­ing to do with web 2.0) but for what­ev­er rea­son, IAs are mov­ing into inter­face ter­ri­to­ry. They’re broad­en­ing their scope to look at how their archi­tec­tures are pre­sent­ed and made usable by users. The inter­est­ing part for me is to see how a dis­ci­pline that has come from tax­onomies, the­sauri and oth­er abstract infor­ma­tion struc­tures approach­es the design of user fac­ing shells for those struc­tures. Are their designs dra­mat­i­cal­ly dif­fer­ent from those cre­at­ed by inter­face design­ers com­ing from a more visu­al domain con­cerned with sur­face? I would say: at least a little… 

I par­tic­u­lar­ly enjoyed Stephen Anderson’s pre­sen­ta­tion on adap­tive inter­faces. He gave many exam­ples of inter­faces that would change accord­ing to user behav­iour, becom­ing more elab­o­rate and explana­to­ry or very min­i­mal and suc­cinct. His main point was to start with a gener­ic inter­face that would be usable by the major­i­ty of users, and then come up with ways to adapt it to dif­fer­ent spe­cif­ic behav­iours. The way in which those adap­ta­tions were deter­mined and doc­u­ment­ed as rules remind­ed me a lot of game design.

Mar­garet Han­ley gave a sol­id talk on the “unsexy side of IA”, name­ly the design of admin­is­tra­tion inter­faces. This typ­i­cal­ly involves com­ing up with a lot of screens with many form fields and con­trols. The inter­faces she cre­at­ed allowed peo­ple to edit data that would nor­mal­ly not be acces­si­ble through a CMS but need­ed edit­ing nonethe­less (prod­uct details for a web shop, for instance). Users are accus­tomed to think­ing in terms of edit­ing pages, not edit­ing data. The trick­i­est bit is to find ways to com­mu­ni­cate how changes made to the data would prop­a­gate through a site and be shown in dif­fer­ent places. There were some inter­est­ing ideas from the audi­ence on this, but no def­i­nite solu­tion was found.

Strategy — fourth of five IA Summit 2007 themes

(Here’s the fourth post on the 2007 IA Sum­mit. You can find the first one that intro­duces the series and describes the first theme ‘tan­gi­ble’ here, the sec­ond one on ‘social’ here and the third one on ‘web of data’ here.)

Like oth­er design dis­ci­plines, IAs are typ­i­cal­ly brought in to solve a prob­lem. The extent to which the design prob­lem is defined and expli­cat­ed is a huge deter­min­ing fac­tor in the suc­cess of their under­tak­ing. More often than not, an IA would take a prob­lem and run with it, not think­ing whether this is the right prob­lem to solve, or even a prob­lem at all!

This has always seemed like a sil­ly sit­u­a­tion to me. Some of the most enjoy­able ses­sions at the sum­mit there­fore were the ones that dis­cussed ways in which IAs can join in on strate­gic think­ing. This way, we can help dis­cov­er the actu­al prob­lem that needs solv­ing, which gives us a bet­ter chance of actu­al­ly deliv­er­ing a suc­cess­ful and valu­able solution.

Gene Smith and Matthew Milan dis­cussed con­cep­tu­al mod­els (which I’ve been play­ing around with for a while) and the more involved rich map­ping, from soft sys­tems think­ing. Key take­away for me was when mod­el­ling a sys­tem we should also describe its con­text (includ­ing the project itself). Oth­er good stuff by peo­ple of Crit­i­cal Mass (Milan again togeth­er with Sam Lad­ner) was pro­vid­ed in the form of ‘back­cast­ing’, a very visu­al brain­storm­ing method to be used in a work­shop ses­sion with a client in order to envi­sion desired project out­comes and map paths from the cur­rent sit­u­a­tion to those out­comes (notes at The Chick­en Test).

Peo­ple from Avenue A Razor­fish (Gar­rick Schmitt, Marisa Gal­lagher) talked about their frame­work for tying togeth­er lots of dif­fer­ent user research such as click stream analy­sis, search logs, eye track­ing and oth­ers. This remind­ed me of Jared Folkmann’s excel­lent talk at last year’s Euro IA Sum­mit in Berlin.

Final­ly, I attend­ed one nice talk (by James Robert­son) on the val­ue of con­tex­tu­al enquiries, which if noth­ing else has made me all the more deter­mined to try this myself the next time an oppor­tu­ni­ty presents itself.