Refusing to be seduced by a golden Sun kissed day and a bustling University Street Fair a stone’s throw away, over 100 professionals, academics & enthusiasts spent their Saturday in discussion on mobile UX topics such as Gestures v/s button; Siri User Research; Fixing small group communication; Strategies for going mobile; Why music apps suck; Winning a date with Windows 8; Crowdsourced UX to heal kidney disease….and these were just a handful among 23 presentations made at the Seattle Mobile UX Camp 2012 ‘Unconference’ held yesterday at the Information School –UW (that’s right… it was right in our backyard….at Mary Gates Hall).

I hoped to attend more than just the four sessions I was finally able to, but the one that stood out for me; also the first one of the day was by Nika Smith – User experience researcher at Blink interactive titled: ‘A Siri-es of unfortunate events – User Research on the first month of Siri ‘

As part of a longitudinal study tracking usage behavior among 11 brand new iPhone 4S users, she recounted their emotional response to Siri (one of the top features they were excited about and couldn’t wait to try) as they tread through three broad stages of Siri exploration – a journey that began from optimistic curiosity fueled by excitement, preconceived notions, and heightened expectation brought about by the media buzz and television commercials slowly veered towards frustrated exploration and eventually rested at acceptance and reconciliation.

It was interesting to learn how participants engaged with the shiny new feature and their resultant emotional states which in Siri’s case fluctuated between wonder, ecstasy, self-doubt, despair, learned helplessness, and hilarity.

The relationship with Siri began as a fun engagement and although it misunderstood queries 50% of the time, participants expected that over time the results would improve as Siri gets to know them better by retaining details of their wants and preferences. Some even thought of it as a friend – someone they could talk to non-stop – referring to the feature as ‘she’ rather than ‘it’.

As days went by a few figured out what it was good at (directions, maps, and basic simple commands). Some did not give up attempts to get it to work on their terms; they expressed that they ‘yet need to learn’ how to effectively use the application, while those more emotionally connected went to the extent of berating themselves for not being able to get Siri to understand them.

Assessing these failure states, Smith noted that Siri did not do a very good job at error recovery to assuage these frustrations. For instance, if instructed to open ‘my notes’, and Siri knows that there is an application called ‘Notes’ Siri won’t offer an alternative that says “I cannot open ‘notes’ but I can open ‘a’ note”. For the music app Pandora; Siri would open the stock price of Pandora which got some participants to realize that apps – even native ones – that did not originally come with the phone could not be opened nor could it search a schedule of an event that was not already logged in your calendar.

As they moved closer towards reconciliation and acceptance, participants acknowledged that even though Siri was cool and fun; the system was broken. They also feared that inaccuracies in the application could have them being caught in an embarrassing situation if Siri misunderstood them (For e.g. erroneously calling an ex) and public use of Siri was probably avoidable because it understands commands only when they are simple, unidirectional, require no follow-up and stated precisely – almost unnaturally – making private use a much better idea. Besides having a robust location intelligence as well as integration with maps and calendar, it also worked well for informal emails and messages as Siri does not recognize alphabet case and punctuation.

The Q&A ensued further discussion on the perils of ‘over-expectation’ induced via marketing and how design and interaction embedded into a consumer application or technology ought to strive harder at ensuring that applications and technology should make the user ‘feel smart’ rather than ‘stupid’ as they navigate a system. While voice has not been perfected as yet, I think that this study overall speaks to the ‘degree of intelligence’ and user-friendliness that users have come to expect from consumer technology-based applications and with a growing beta release culture in this area bears particular attention.