Maya on Interaction

Error messages are baffling if you take them literally.

Take this one that’s always lurking in the corner of browser windows: “Object expected on line 1.” Though you’d never know this from the message, no ordinary object e.g. fishing rod, tea bag nor any change to line 1 will fix this JavaScript error. If you ask around, you find out that the message means: somewhere – on any line of code – there’s an open curly brace { with no soulmate closing curly brace }.

Turning now to heartache, I’m sort of an open curly brace without a soulmate. It’s a tricky predicament. Although no code will resolve it, the fix is riddled with error messages.

Take “hope to see you soon.” I got this one after behaving like a horrifying goof and asked some earthlings what it meant. As it turns out, mention of future contact at a distant unspecified time means precisely: I intend never to endure the torment of your company again. If you get this error message *do not* wonder how soon or attempt to coordinate a seeing soon.

Because it neither describes a problem nor suggests a solution, “hope to see you soon” is an irrecoverable error. It’s made all the more difficult by the context in which it occurs: A person who you think is the bees knees hijacks your thoughts for weeks. New t-shirts, lipgloss with sparkles, many text messages, panic, nausea and forgetting who you are ensue. In all this fuss, you’re in no shape to deal with confusion.

All this to say: if you must dash out of an admirer’s life, mind your error messages. Follow Jacob Nielsen’s guidelines and give an “explicit indication that something has gone wrong” and “constructive advice on how to fix the problem.” To be thorough, you might use this gadget to put a user interface on your error message and see how it reads.

This is neither explicit feedback nor does it promote recovery.

This doesn’t promote recovery but acknowledges a change in the state of affairs. It is legitimate if unfortunate feedback.

This one describes an unfortunate change in the state of affairs, provides legitimate feedback and promotes recovery.

A street canvasser got me the other day. Street canvassers are committed to expose darkness in the souls of passers-by. They yell out all sorts of difficult questions that insinuate roughly the same thing: Write your credit card number on my clipboard or I’ll carve “I don’t want to help the visually impaired” or “I’m keen to promote drug abuse” on your tombstone.

The canvasser routine is tough but no match for my killer canvasser line: “Look mate, I’m interested in fighting whatever social ill is on your mind. Thing is, I’m not a make-fiscal-decisions-on-the-street sort of lady. I’m a make-decisions-on-a-website lady. So fork over a URL and I’ll take it from here.”

Street canvassers always recede in the dust of my blazing line. But Henrique the Help the Aged campaigner at lunchtime on Old Street was unperturbed. Not a split second after I uttered it, he came back with this one: “I’ll be your website.”

I was floored. For one thing, it was touching. For another, I couldn’t figure out what to do. It was obvious that he didn’t support navigation or keyword search, there was nothing to click, no overview of the information space and no interaction sequence was specified.

The gut-wrenching truth of the matter is that I don’t know how to use these things. In this our age of web personalisation, pipes and collaborative filtering, we want applications to act like beings who get us. Meanwhile, facing Henrique the information-dispensing human, all I could do was stutter and pray for a manual to drop from heaven.

My reaction is characteristic of what Natasha and Tamar have called the Nell problem. It’s unbecoming and I’m resolved to lick it. Esteemed readers, beloved friends: I’ll be your website. Just tell me how. Think of a tip you’d give a Martian (or a mildly disoriented Canadian in the throws of Saturn’s return) before sending her off to the pub. For each bit of advice you post here, I will 1. cherish it and 2. put five quid into a pot for the Help the Aged Enough is Enough appeal. I’ll tack your comments to my bedroom wall and make the donation on the 30th of April. Watch this space for grand totals and spiritual progress.

The new self-checkout machines are all the rage at my neighbourhood Tesco. As in any sensible initiative to replace people with computers, Tesco tripled its human staff load for the occasion. Extra personnel scurry about to iron out kinks and liaise between customer and machine.

Kinks occur because the people at the till or cash register can do much more than follow procedures but self-checkout machines only follow procedures.

Take this procedure:
1. price an item
2. throw it in a bag

And take this situation:
I (driven by early-morning munchies so powerful that I forgot that I am too discriminating to buy croissants from Tesco) put two croissants in a single plastic bag. Since one had pecans in it and the other didn’t, they had different prices. I might have put each in a separate bag but I’m with Ben Bradshaw. (props to Mlle. dron)

At checkout time, the machine asked me to press the button which depicted my product and to indicate how many I wanted. The machine lost its grip on my experience after I priced the first croissant but placed both (since they were in the same bag) on the adjacent scale. After pricing the second, the machine told me to put it on the scale. This request wasn’t coherent because the croissant was there already.

Tesco’s human assistants could see that all was kosher but the machine was in a huff. The weight of my grocery pile had to increase in order for us to move forward. Muddling through, I tossed my keys on the scale. The machine was appeased and a Tesco staff member sanctioned the manoeuvre with a nod.

Still, the case of the mediocre croissant purchase raises a chilling truth: good user experience can’t be specified in a procedure. Users are quirky pecans. Their unaccountable tastes and hang-ups lead to scenarios that we could never anticipate much less optimise in advance. Designing an experience demands an eye to procedure but also a readiness to dive into action – in the midst of the experience - with wily workarounds and situation-specific solutions.

For you good souls that are burning with contemplation and craving action - a thought experiment to blast you out of 2006 analysis paralysis and make you dance like wild monkeys this year:

Think of a donkey who stands at an equal distance from two identical stacks of hay.

Imagine that like you, the donkey is in an analytic rut. He insists that no action is justified without a reason. Not being able to justify a move in the direction of one or the other pile of hay, he starves to death.

So bust an arbitrary move you contemplative cats. Arbitrary action is delicious, nutritious, wild fun and totally rational.

I’m off to celebrate whichever haystack you pick and all the mindblowing things you’ll do with its nutrients.

I’ve been drinking (an unnatural amount of?) this tea.

The box says: “Succumb to this salubrious communion of Rose and Chamomile as it slowly seduces the inner you and irresistibly warms up your heart. For tea lovers everywhere, let the love flow!”

Ingredients
Chamomile, Lime flower, Elderflowers, Marigold petals, Licorice, Red roses, Lavender, Rose extract

I think I succumb.

Sociopaths, I’m OUT of the club.

Photos and hard labour courtesy of Miss Mannella, scarf, warmth and total grooviness by Logan, much-needed vodka courtesy of Laila, equipment courtesy of Lindsay, flight and wishes courtesy of Birgit and Christian, fine chocolate by Tanmay, tiramisu mastery by Dr. Pica, special thanks to John for jogging over against all London odds and to free wireless comerade for managing to make an appearance despite a tumultous night-before.

I’m preoccupied with location. There’s a quality of *being here* that just doesn’t creep into virtual representations or maps of physical space. In order to be meaningful, maps or other virtual representations need to be annotated by people capable of expressing a rich contextual appreciation of a space.In all this, my friends are expressing concern about my competence with regard to spatial navigation.

I submit some ridicule from Dionisis:

“Maya jogged me to death after waking me up with an sms at 9.15am on this brisk Sunday morning. I hope I will live to make it to our next relaxing event. It’s a good thing that we were running in a straight line but she lost her way for approximately 20 minutes so I got some rest walking. I drunk two Lucozades and got a taxi back home but have yet to recover fully”

When I was 9, workers at my Mum’s library went on strike or on vacation, and I stepped in to cope with the extra workload. She put some blank stickers, a rubber stamp and a stack of books in front of me and demonstrated the procedure: peel - stick - stamp. I repeated a million times until an innovative - which means a desperate - impulse kicked in and I changed the procedure: stamp all stickers first - iterate [peel and stick.] Procedure 2 is better because it’s quicker (about 50%) and tidier (no ink bleed on book). My mum was floored and procedure 2 was the talk of the community for a week.

What do you make of this scenario from a knowledge management point of view? Is there some knowledge I exhibited that the regularly scheduled masters of library and information science didn’t have? Uhhh… Is it worth investigating what organisational structure or what lucky confluence of circumstances in the library’s environment prompted the inspired change of procedure? Duhhhh…

No. Unlike the regularly scheduled crowd, whose job it was to follow a procedure, my job was to solve a problem. Being 9, unpaid, unsupervised, and ‘in’ with the management, my circumstances afforded me the audacity to suppose that my insight was valuable and the license to design my work.

Managers inclined to nurture knowledge work in a mum-like way, are strongly supported by management theory since the 50’s. The Japanese production industry is buzzing with methods for coaxing luscious and lucrative ideas out of the cogs. Meanwhile, Amazon, like anyone with any sense, is clawing the walls to see what new modes of exploitation are enabled by service-oriented software.

Artificial artificial inteligence. What’s artificial about Amazon’s brand of artificial intelligence? It uses humans.

This makes good exploitative sense. Turing, G�del, Chomsky, etc. prove that there are some things that humans can know but that computers can’t. Things that are uniquely knowable to people often represent inspired discoveries or feats of creativity - like the double helix or a symphony. There are other things that people can know but computers can’t that are dull and low-level, like CAPCHA tests - when you have to decipher letters and numbers from those images that online services use to prevent automated subscriptions. So guess which kind of knowledge Amazon’s leveraging?

Right! It’s the “Is there a pizza parlour in this photograph?” sort of knowledge that you can aspire to cultivate if you want to join Amazon and its customers in the neurotic scurry map and serve location-specific information.

While spacial coordinates are fixed and bounded, their coffee-serving, film-projecting and petrol-dispensing occupants are not. To maintain its status of trusted referrer, Google maps doesn’t really have a margin of error - if it buggers you with enough, or even one false positive (yessiree, there’s a circus right over yonder. Round up the kids and get the engine croaking…) they compromise the credibility of every single bit of information they have about what’s where. And while machinery can monitor fluctuations in space, only its human components can tell you weather some conglomorate of matter in a space constitutes a pizza parlour.

I’m delighted to see your name in the to field of a service-generated message, and I would even read the contence of the greeting with interest if I didn’t have to click the link below to declare my address live and spammable.

Even though I’ve only resorted to it once, I can understand that mass-messaging makes life easy, and maybe there is something about 16 pt Comic Sans MS that heightens the tenderness of your written wishes. It’s ok, my love for you is not contingent on your personal e-attention. But you are testing my love by precipitating spam.

Read the privacy policies (lurking, in very uncelebratory 9 pt arial at the bottom of hyperactive e-greeting pages) before you divulge my address which I try hard to safeguard from vile precipitators of info overload.

To be honest, I am enraged by your greeting approach. Here is my usual (but highly personalised) treatment of the sticky issue:

“Thanks so much for your holiday wishes. It’s great to hear from you, and makes me curious about life and new developments.” [I remain curious because you’ve put me in a situation where I have to expose myself to danger in order to get the details.]

…Specific questions about recent travel, happy or contentious family matters, business initatives… [I love you enough to consider happenings that are significant to you.]…

“Oh - P.S., my suspicion about those e-greeting card services is that they are sneakier than they seem.” [I’m not a know-it-all, just sober and downtrodden by spam.]

…Then I breifly describe what I think the greeting card pimps are up to…

“Not sure if spam is as much of a problem for the rest of the gang, [Make no mistake: you are causing malaise for everyone on your mailing list] but speaking personally, it’s such a thrill to hear from you, that plain text more than does the trick!”

Sometimes what I think is fantastic diplomacy comes out as blazing sarcasm. Tell me if my message seems harsh in the context of furry graphics and fonts. Maybe someone should build an e-greeting response site: Choose a color scheme, cute animated things, and a public service message: “You meant well but did you know…” that’s neutral and impersonal so just as unlikely to offend as its trigger message is to heartwarm.

Usability is a virtue of literature not just websites.

A client and friend once told me that he is plagued by systems that are well-conceived but ill-understood.

The applications he built were rich in the functionality that his users ached for. To his perpetual distress, however, they continued to ache long after the application was delivered.

If, like him, you paid my bosses 90 Canadian dollars per hour for my usability sermons, I�d churn out insight that�s purely intuitive and that could easily have come to you in the shower. I�m sure I�d attack your navigation scheme. In particular, I�d ask why you chose to put terms or phrases on navigation links that make sense to you and your work cronies, but that give users no inkling of what they could expect to find once they clicked.

Websites, if they�re to persuade us to divulge our credit card numbers, need to take every opportunity to instruct. It would be nice if academic articles did the same.

Digesting academic information systems literature, the usability witch in me rages �quit hogging bandwidth and get to the goddamn point.� I probably need a new line more suitable for paper journals, but I think line for scholars is the same as for web designers: use plain language.