UX@UW

On the afternoon train down to Portland– just beyond the usual giddiness I feel on the train heading with Kevin, a good friend and super-talented UX colleague, to one of my favorite cities w/ our bikes safely stowed in the luggage car– there was a tightness in my gut as I looked over the conference schedule for one last time to designate which sessions I would focus my attentions on over the next three days. I hate this part. A feeling of helplessness takes hold of me as I read over the carefully crafted, heavily perfumed session summaries that flirt with my expectations and judgement, rendering them dazzled, bewitched and totally useless to the task at hand. Those soft, insinuating phrases, accented with voluptuous, suggestive buzzwords that threaten to burst with meaning from the tight bounds of brevity– all shimmying for my attention with winking promise of relevance and meaning. However, as I am drawn in, they turn coy and I am left to wonder if there is anything of substance behind the beckoning words.

The success of my conference attendance hangs in the balance! As I know from past experience, picking the wrong sessions can ruin a conference experience. From a UX perspective, I have a clear goal and I just need the facts about what the presenters intend to present, no tarted-up descriptions or advertising required. I am a captive, engaged audience. If I’m interested in the facts of what you will present, I’ll be there. If not, then I’ll hit something else. From the presenter’s perspective, they should be concerned with getting their message out to an audience that will benefit from it and, most importantly, be engaged in what they have to say. The needs are compatible; to present the simple facts to allow the audience to make the best decision about whether the session will be relevant and interesting to them. There is no logical reason for them to try to compete for people who are more interested in one of the other subsequent lectures. Of course, if it really is this simple, then why do we consistently get these ridiculous summaries that often contain few clear facts and are simply misleading most of the time? This is a total mystery to me. Maybe a misguided marketing effort by conference promoters? Maybe butts in seats for previous conferences is an actual evaluation consideration when concert promoters are selecting conference presenters? I’m perplexed. I know this is a ubiquitous problem… but it really shouldn’t be.

Overall, this proved to be a small concern. Either due to blind luck or to a ridiculous abundance of excellent presenters, I ended up with some remarkable sessions on my dance card, many of which left my head abuzzing with new ideas that I couldn’t wait for a break to chat with someone about. As I am writing this up, I find that I have filtered my takeaways through personal interpretation, conversations w/ other attendees, and in discussions within the ACA UX team since I’ve been back. I’ll try to reference the sessions in which an idea germinated below via footnotes.

My personal takeaways from WebVisions 2012:

Working with teams

• Experiment with remote whiteboarding. When working with remote designers or clients, using screen sharing and OmniGraffle to get/give real-time feedback on wireframes. Could be interesting to use w/ Dropbox, as then you can truly collaborate on the same file. [1] [3]
• Essential design critique tips. No blessings, brainstorms, egos or silent people allowed. Focus on WHY. [7]
• Get developers coding early. As soon as functional reqs, audience and context have been identified, you can work together to whiteboard/illustrate your system model using the nouns and verbs uncovered in your contextual analysis and functional requirements. Nouns become objects and the verbs define the API. By working in this visual model, both devs/design can communicate effectively to nail down the system model early on. Then the evs can get started building the basic model while you work on designing the UX. [3]

Design

• Embrace complexity. UX Design is about making the complex seem simple/intuitive to the user. [14]
• Be ridiculous. Don’t be afraid to put ridiculous things in front of users. You are guaranteed a reaction and that is almost always informative. [16]

Mobile Design

• Responsive Web Design (RWD) is all about layout. We need to move beyond this and make sure that we are designing the experiences and interactions that will benefit our users. [4]
• Do the proper context/user research to understand what users need. Don’t just do things because they are possible.
• APIs are where the power for designers is, not in layout!
• Maybe a good way to use RWD is for mobile devices only… where you have a web front-end that takes care of phones and tablets, as their capabilities and limitations are similar. Then have a different web experience for PCs. [1]
• Keep the PC/Mobile system in mind when designing. An example is in supporting shopping tasks (PDF). PCs and mobile devices usually complement each other, with a large percentage of users initiating a task on PC or Mobile and then finishing the task on the other device. Understand the system to understand what should be supported on which device. [4]
• Think about changing the UI for the users’ context. If they are in a specific building, the UI can be one thing that helps them with building-specific tasks… if they are in their home, the UI can change to help them w/ home-specific tasks. [16]
• Crowdsourcing solutions. Hackathons are a good way to pull community resources around a specific issue/problem. Contests result in a wide spread of problems/solutions and aren’t focused enough to be worthwhile. [11]

Working with content

• APIs are the key. Freely restyle, restructure content for different contexts/devices. Content should be broken into the smallest pieces possible, so that it helps in reuse. [13]

Working w/ Data

• Stand on your head. When using data, try to look at the same data from different perspectives (using different sources, visualizations and tools) in order to fully understand it. [6]
• Numbers can lead you astray. Quantitative data is only one part of the story. It must be understood w/in the context of qualitative data. Do NOT let numbers drive your decision making process. [8]
• Know your data. Know exactly why you are collecting it and what you are measuring with it. Tie data back to Objectives, which tie back to functionality. Define measures and know at what points you would respond to the data. [8]
• Know your context. Know what your data should look like when the objectives are being met and users are happy AND know what your data should look like when the objectives are not being met and users are unhappy. [12]
• Own your data. Fight against misleading measures and inappropriate appropriation of your data. [8]
• Know your audience. Think about the consumers of your data and what they are looking for. That will influence how you present your data. [12]
• Understand their strengths and play to them. Since most visualizations are relational, know what humans are good at comparing (super good at line length/height. Not so good with intensity, sizes (volumes) or quantities.
• When designing visualizations: [12]
• Use the right visualization for what you are communicating.
• Assume normality and highlight abnormality
• Design for what is less relevant on the page, rather than just finding ways to make things bolder
• People will learn to read complex charts if it is in their best interest to do so
• Never lie with visuals. These should always clarify information, never muddle it.

Maps

• Google may not be the best choice for maps. They are beginning to ramp up pay-for-use and advertising schemes. Lots of open source options that actually allow for more designer/dev control by separating out the data/tiles/api (controls). Complete info for how to do this [5]

Tools/Readings to dwell on

• Twitter as professional tool. Decided to use WebVisions as a reason to give Twitter a fair shake. Now, it is finding a place in my professional life. Simple, powerful stuff! @jcivjan #wvpdx [1]
• Breakout JS: Interesting framework to create controllers that span the virtual and real worlds. [4]
• Kendo UI Mobile: Powerful framework for using w/ PhoneGap/Cordova to create native apps. Automatically tweaks interactions to customize for android vs iOS. Cost is high, but free for non-commercial (for now). [10]
• Reading: Zombie Apocalypse of Devices [4]
• Data analysis tools: [6]
  • Webtrends tool suite, especially Analytics 10
  • Google Analytics
  • chartbeat (very much for real-time data)
  • Omniture
  • App Annie (app store integration to get download data for your app)
• Good UX/data blog: mauvyrusset.com [8]
• Great Data Blog: blog.intercom.io [12]
• Cool app sandbox. PDX is a model for gov’t and technology. They are using this site to test/generate apps. [11]
• Tools for building visualizations (HTML is the way to go) [12]
• Highcharts is a great JS library ($49 per developer)
• Flotr is popular
• D3 is immense
• Rickshaw is based on D3, but easier

Sessions you should schedule some time to curl up with:

• Measures of Success: How to Quantitatively Measure Your User Experience. A fascinating and truly innovative approach to use quantitative data to help make design decisions. It didn’t hurt that Richard Dalton framed this information in a sleek and absolutely engaging presentation.
• Data Visualization and Dashboard Design. Another master in the art of presentation, Des Taylor, distills pretty much everything you should know about data visualization into 45 minutes.
• Discussing Design: The art of Critique. A thoughtfully articulate and thorough presentation by Aaron Irizarry + Adam Connor.
• When Google Maps Gives You Lemons, Make Lemonade. Mind-expanding presentation by Wm Leler. Reminded me to question the things we take for granted everyday (like Google Maps). That doesn’t happen often enough.
• Embrace the Anywhere, Everywhere Web. A timely rant from Tim Kadlec

Sessions I regret spending time with:

• Workshop: The Device is More Than a Display by Gino Zahnd and John Bragg. A total bait and switch from my perspective. Ended up being a group exercise in UCD.
• Identifying your Audience: Data Gone Wild by Jordan Dossett. My fault, as I didn’t read the summary carefully. I wasn’t expecting a session geared towards people with absolutely no experience gathering user data.
• Whither Twitter? by Laura Fitton. I cringe at misleading data and people obsessed with celebrity. This talk was not for me.