Adaptive Path Blog

Sunday I lead a workshop with my friend and former Yahoo! colleague, Mirjana Spasojevic – currently at Nokia Research Center in Palo Alto - at the Ubicomp conference in Innsbruck, Austria. We had a good turnout of people (14 total) with a mix of folk from both academic and industry backgrounds

The goal of the workshop was to harness the collective mobile wisdom of the group and create 4-6 mobile user experience design principles. We started out the day with short introductions and launched into discussing possible themes from which we could base the principles.

The themes that emerged were:
- Mobile phones and changing social rules
- Where does the data live?
- Relevance: Personalization and location-based services
- Divided attention
- New mobile interaction models

Next, we divided into groups of 3-4 people and spent the afternoon discussing the theme and shaping it into a design principle. Here are the themes and accompanying design principles. Admittedly, these discussions are reflective of some wandering conversations and may read a little “wonky” – but there is some really good stuff in there.

Theme 1: Mobile Phones and Changing Social Rules
Examples of changing social rules with regard to mobile phones:
- acceptable to talk to yourself on the street (when using a Bluetooth headset or speaker feature on the mobile phone.
- addictive “Crackberry behavior”
- always perceived to be available now that you have a mobile phone
- busy button doesn’t help – negative connotations

The lion’s share of the conversation for this group focused on the theme of expectations. Social rules are based on our understanding of expectations - expectations about ourselves and how we want people to engage with us, and expectations of others and how we want them to engage with us.

We talked about how there are different expectations for the various communication channels. When you call someone, you expect them to answer. If they don’t, you leave a voicemail and expect them to get back to you. The rules for email and text messaging are slightly different – often the social rules are personal and reflective of the relationship the sender and receiver have – or the social contract they share.

There is an interesting tension between inference and plausible deniability. When someone doesn’t answer their phone, there are a set of likely explanations: the receiver is busy, not within reach of their phone, etc…

Conflict seems to occur when expectations are not in alignment. Technology such as gps location or IM status message of “busy” adds a layer of complexity – the sender can know more about the receiver’s state and adjust their expectations accordingly.

Design Principle 1:
Design an appropriate level of ambiguity – tell users something about state, but not everything. Allow for “states” that are not fully revealed so that people can manage expectations.
- Allow people to do what they need to do to retain social contracts
- Knowing is not everything. Sometimes white lies are necessary and desired.
- Systems should support different levels of profiles and different levels of engagement to reflect the variety of expectations people have for various relationships.

Theme: Personalization and location-based services
This group’s discussion started off with a comparison of camera phones and the mobile web. Why did camera phones become a mainstream feature on phones while the mobile web continues to struggle to find a widespread audience. Sure, there are technology constraints that could be the cause. However, the group also added that we’re not that good at predicting and understanding people’s relationship to information.

How do we better predict what people will want? How do we do it without being pushy and invasive. The building blocks of these location-based services are context/location, state of mind, and user motivation. We can use the technology to predict location – but we need to find ways of understanding motivation and state of mind.

The group then discussed how these new mobile location-based services are like a new friendship – tenuous. You can’t be too pushy too soon. Then Dean interjected: Being annoying is sometimes okay… if it works. Pushy people are good at getting what hey want. If the goal is compliance in the short term, pushy can good. Point taken - i see where he is coming from, but disagree. Annoyance is yucky and should be avoided.

Next, the team discussed the idea of enhancement. We discussed the Starbuck’s – iPhone – iTunes service. interesting:
- Being in a Starbuck’s infers state of mind
- The value to the user is in the distillation of large amounts of information into a simple interaction and fulfillment of a need/desire.

Ultimately, the group felt the real value is in enhancing existing experience, not hijacking the experience.

Design Principle 2:
“I’m not in the mood”: Services should enhance the experience and provide added value. The phone can determine location, but mood and motivation are key

Mood and motivation are hard to predict. Location can give some insights into mood (Starbuck’s example). Status (location+emotion+motivation) is everything.

Theme 3: Mobile applications as interventions
This team took on the theme of attention resources. They discussed that a user’s attention is often divided when using mobile applications and services. Therefore, they framed the discussion around mobile applications as interventions.

Next, the team discussed that there are three basic phases to consider in when designing a mobile “intervention applications”:
- Sensing, (data collection and context)
- Delivery notification
- User response

When thinking about the sensing phase, consider the value of the application to a user – is there personal and/or global benefit? This may give insight into how much attention and energy the use is willing to give during this phase.

Users ultimately need to have control over the notifications. Users will ultimately want to be able to override the notifications. Deliver notification data that is contextually appropriate as possible. Consider the environmental and social contexts. Interventions should not be like your annoying friends.

Design Principle 3:
Mobility implies changing contexts and changing interruptability. Consider the three phases of mobile “intervention” applications:
- Sensing, data collection and context
- Delivery notification
- User response

Theme 4: New mobile interaction models
This team discussed the importance of effort and time when considering interaction models. Context is king; it determines cost (to use) and user value. The challenge and opportunity of mobile is that context is highly variable and hard to predict.

There was then some debate as to who ultimately bears responsibility in managing attention and intention? Producer (sender) or receiver. Notification is also a question. For example: is it socially appropriate for me to expect that you know my blog posts?

The conversation then turned to a discussion around broadcasting services such as Twitter, Radar, Flickr, microbloogging and lifecasting and the power of mobile as a capture device.

It was also discussed that it is easy in such cases to get too much information on a mobile device and the Twitter + SXSW example was cited. People signed up, chose to follow lots of friends, but then eventually turned off the service because they were inundated with “tweets” from too many friends. Was the problem a flaw in the design of Twitter or that people didn’t fundamentally understand the “rules” of social networks on mobile.

Francis felt that users should have no barrier and that ultimately the burden for providing controls should be embedded in the system. Dean disagreed (slightly) and explained that it is often difficult to predict what will happen and design appropriately.

Design Principle 4:
It’s not just about designing for a user: it’s about designing for a user embedded in a context. It’s about recognizing the different roles that people have.

Think multiple channels. The mobile platform is sms, mobile web, voice, applications. Depending on the context of your application, expectations are different.

Thanks to all the participants for a day of engaging conversation and contributions to the mobile UX principles.
Daniel Harris
Tony Lee
Fu Yu
Jaejoon Hwang
Taejin Jeong
Alexander Meschtscherjakov
Tim Sohn
Paul Aoki
Francis Li
Mattias Rost
Dean Eckles

This entry was posted on Monday, September 17th, 2007 at 8:18 am and is filed under Conference Commentary, Mobile, Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

This post is licensed under a Attribution-Noncommercial-No Derivative Works 3.0 Unported License.