Recent Blog Posts

The trend towards podcasts and "video articles" and other media broadcast over IP seems to continue unabated, but I have lots of problems with them. I suppose in the strictest sense I have the time, but I don't consider most of my information consumption to be that single tasking.

In the car, I'd rather listen to NPR or talk to Alison, and mostly I am annoyed by the low information density and linear, single-layer nature of all broadcast-like media. With text, I can skim, I can read at my own speed, I can stop and dive into something, I can back up and read it again, I can even copy text for later or for someone else (or quote it in a blog entry). Even reading books, I'll stop and search on the computer (or my phone) for relevant information, and send emails to friends about items I just found. Hypermedia can make me even happier, with deep dives into related topics by following a link to a wholly other article (and if not provided, just do a search).

See full-size image

With audio or video, I have to go at the broadcaster's speed, and live with their detail level. I cannot follow links, and cannot even search really, unless I stop and go somewhere else. Yes, I do this even with TV (PVRs are wonderful), but there is a discontinuity to the consumption process that is unavoidable. At TV-watching-time, that tends to be my focus activity. When focusing on other things, it can be difficult to follow a narrative with arbitrary interruptions.

Mobile usage of the same media just exemplify, and amplify, the problems I encounter on the desktop computer. The first thing that made me start using any internet, nav or multimedia features on my mobile was multithreading. Because I can hardly get thru a song, or look up something on a map without getting a call or text, or needing to look something up in another program.

But aside from (often, and only recently) being multi-taskers, mobiles are not really designed well for interruptability. Only one device I have used really has a decent system of pausing media when a call comes in, and resuming when it's done. But that only works for the built-in applications, and it's very binary; the audio stops, then restarts immediately.

In our ideal world, the mobile is not just a tiny little computer/camera/mp3-player/tv, but is designed to take advantage of it's unique attributes, and avoid it's pitfalls. Design for glancing and interruptability is one of these key factors. So what do I see as a solution to my vague griping?

Short term:

• Deeper, broader content. Especially, better linking. Far too much mobile content is a dead-end, for no good reason. Offer links to other parts of the site or service, or to other information sources entirely. Think outside the box as far as what other content can mean; why does your music player not have the ability to get lyrics from the internet?
• Always offer multiple versions of content. If there is a video about a product, transcribe it. A few services do this, but mostly on the desktop web; NPR has close to perfect transcripts of every article on every show, posted fairly soon after broadcast.
• Always tell the user what they are about to view, and offer those options. I am never happy when browsing around a website and following a text link loads a "video article," with no text version even as an option.
• Use better software. There are some really neat ways of viewing content that already exist (hence, its short term, since nothing needs to be developed), but are inexplicably niche products. The FileMagnet, with it's perfectly-tuned tilt-for-scroll control, is perhaps the first reader I have ever used that felt totally natural. There is a long ways to go yet on input and output technologies, but at least the state of the art could be employed.

Longer term:

• Allow all applications to be aware of device features, then they could all react appropriately to incoming calls, or any use of other applications. This used to be totally pie-in-the-sky since no one wanted to play nicely, but Bondi might help get us all there.
• Hypermedia outside of text. How would you like to be able to "click" on an audio phrase, or listen to the rest of that music track in the background of news story?
• Dual-channel communications. Accompany audio/video with text content, which can contain links and even additional visualizations of content. This seems the ideal mobile use pattern for any content. Look up the weather, and you hear a talking-head weather report while a chart displays on the screen. Use whichever you like. Aside from my convenience, think of the appeal to low-literacy communities (and related: youth, aged, low-vision...).
• Monetize it. Without a good financial reason for existing, nothing will happen on a large scale. I think, however, that the above capabilities for multi-channel communications or audio/video hypermedia have a lot of opportunities for promotion, cross-selling, stickiness, advertising. Just consider TV commercials: why can't you click to get more info about a product, to call the number on screen, or set up to record the show they are advertising? No matter how appealing something is, friction kills clickthrough and sales.

Okay, this isn’t relevant to most of you out there in the world, but it’s good for you local folks.

Yes, there are web, mobile, industrial, graphic, and other designers in Kansas. More than you think. But we’re at dozens of little companies, with lots of freelancers and distance workers. So we don’t know each other. Let’s fix that.

Inaugural Lawrence Design Professionals meeting
Thursday, December 11
8 am
First Watch (Iowa Street)

What to do next? Please:

1. Forward this message to anybody you think would be interested
2. Fill out your information in the Google form, if you haven’t already
3. Send me an RSVP indicating whether you can make it (so I can warn the restaurant!)

The latest Carnival of the Mobilists is up on the Mippin blog.

And now that I’m back from a big trip, We’ll be posting a bit more often.

You’ve probably noticed that not everybody reads RSS feeds, even one as (potentially) relevant as ours. The use rate may be as low as 11%, though that data is somewhat questionable. I’ve especially noticed that mobile designers, product managers, and marketers in large companies don’t read blogs. As a result, they are unaware of the huge array of resources available to do their jobs.

If you know someone like this, please take a moment and suggest that they
sign up for our newsletter! With the numbers above, we should have at least twice as many subscribers for our newsletter as for the feeds, but the numbers are reversed from that prediction.

Each newsletter contains highlights of the month’s blog content, occasionally supplemented with more content and usually clustered in ways that weren’t obvious while we were writing the original articles. It comes once a month, so really doesn’t fill up the mailbox, and content is always highly related to mobile design.

And we don’t use the email address for anything else, unless of course you ask us to.

I keep fussing about details, because no matter how much wow factor your service has, missed details can erode your brand over time. Yesterday at the Mobile Device and User Experience, Scott Weiss (now of Human Factors International) talked about how the great transitions and visual design of the iPhone provided a “halo effect” that delayed users from noticing and becoming frustrated with the inconsistent back button placement, difficult text entry, the fact that the device never does learn anything about text input.

I just tried out the ABBYY Business Card Reader for S60 so I didn’t have to bring all the cards home. I’ll spend the money on the app, but I’m not sure why it consistently puts FIRSTNAME LASTNAME into the first name field, with nothing in last name. It seems a particularly easy bit of code, so what gives?

It’s way too easy to pick on pretty much any device and any service on getting details right. Scott does a great job of pointing out iPhone issues (though my chief irritants have to do with browser behavior), but what about Nokia? Well, here are some issues I found irritating. When I swap out the SIM on my E71 and launch messages and try to view one … nothing happens. I have to reboot it again. And then, there is the lack of keyboard shortcuts, especially on my QWERTY device. Not to mention, I learned that my device had predictive input at an industry conference, not having noticed it in the device.

By the way, XT9 is terrific! It learns words immediately. Love it so far.

Another example of missed details is the latest Gmail client. Two major problems for me:

• I keep getting cognitive dissonance when it announces that I have 6 messages in my Inbox but I can’t find them; I must go Options > Refresh to actually see the new messages


• It doesn’t realize I’m using a QWERTY device and thus my delete “shortcut” is # (two key presses) and there is no shortcut for archive

Why are these details wrong? It depends.

1. Not considering target audience and their goals, activities, needs, desires. The G1 is so developer-focused that small hands will actually cramp while typing.





2. Insufficient development & testing resources, and after all something had to give (think about the abrupt transition from Flash to Windows Mobile on many devices especially HTC)


3. Legacy code (I’m looking at you, Motorola and Nokia)


4. Insufficient knowledge of the domain (it’s shocking how many companies approach us and do not have an approach for dealing with device diversity


5. Piecemeal design, like a rambling home added onto by many owners in many architectural styles, without a consistency of purpose (like parts of the iPhone UI and also like the location of my very first job, the Winchester Mystery House


6. Product management processes that rate bugs on a scale of 1 to 5, then get all of the severity 1 and 2 bugs and some of the severity 3 bugs done before launch. Oh, and typically the worst UI bug can be is a 3.

How do you fix it? It’s hard. If you’re designing a platform, try to make the presentation layer flexible, including screen and even functionality. This will allow a bit more time before things start getting clunky. Rethink your paradigms every once in a while; don’t assume that a great user experience 8 years ago remains great. Features have been added, content has scaled, device capabilities have shifted, input mechanisms changed, and user expectations have evolved.

Set standards for user experience testing. Usability must score a certain level, perhaps benchmarked by the competition, before the product can launch; a bad task score must be launch gating. Don’t just test high-frequency things. And don’t just test usability: test learnability, speed to expert use, satisfaction by expert users. And definitely test the things that drive revenue and costs. Measure how well the experience matches with your brand goals. You do have brand goals, don’t you? A story about what you provide?

Get a second opinion. Even if you can’t invest in a lot of testing, get knowledgeable but outside resources to play with

One of the discussions I hear occasionally from user experience types is how to get good user-centered design into actual products. Most products come from companies whose core competency is something other than design; they are product or software or finance-centric. And that's fine.

A common refrain is, as recently discussed here is that UX should never be a department, but must be ingrained in the company. The theory is that if it's one department then no one else will bother to work on it.

There's only one problem with that concept: no one but us UX folks have this discussion. There is no reason good design, or user-centered-anything will occur without an organizational structure to guide them to it. You can be sad about this, but I've learned to live with the truth that this is how corporate entities work.

That's not really a bad thing, though. People do like to work together. Policy and procedure get followed, at least in principle, so your activities will be performed. And when you understand the systems, you can work within them. I have been contacted a lot (when I worked at a Fortune 50 company) by people who had only a vague notion of what our team did, but had heard good things about us from others who had executed projects.

If you are toiling away in the basement with no respect, you might just be shaking your head and laughing that this could ever exist at your company. I, for one, have been on a team that moved from an obscure corner to a 60-person team with responsibilities in CEO decks. To get there you need to:

1. Exist – You have to have a department. However small it is, make sure it's a single team with a name that means something. If you have a choice, and you might later on as your company reorganizes itself, where is the best place to be? I am not sure. I suspect getting a C-level team would be good (with your boss, the Chief User Experience Officer) but how likely is that. I've worked in or with teams in Marketing, Product Development, Operations and IT (and probably others) and none were perfect or terrible. But do understand the culture in which you are operating; IT and Marketing have different measures of success.
2. Get in the process – Assuming you are talking about interactive design, software development processes rule the world. Whether this is good, or any process is well-designed, is a whole other disucssion and not something you will be able to influence right away. You need to get in the process, formally. This will take a couple steps alone:
   1. Understand the process – Most people do not understand software development processes. However, most developers don't either, or have forgotten over time. Get some books, hire the right people, and talk to the process managers (they probably exist). You can use the process to your favor if you know it well. Also note, many software processes (though not many web processes) have an interaction design phase, that is just poorly understood.
   2. Sell your position – You need to be able to talk to VPs, product managers and process boards. You'll need all the knowledge of why UX is important, why this will improve development, why this will help the company bottom line and be able to do it off the top of your head. You might not be able to do this; when building a team, often the manager is a key job for the core designers to find. Get someone who can do this for you.
   3. Build the process – Once a line item is accepted as being your responsibility, you aren't done. There are still options as to how it gets worked, and those will determine how much influence your design actually has.
   4. Defend your step – You have to keep working as you said you would, maybe even push the boundaries, and explain the value all the time. Being complacent or skipping out of some projects can cause you to be left behind, even if the official paperwork says you have a seat at the table.
   Getting signature authority is best; no project can proceed without your department's approval. But this will not happen day one, and with it comes extra responsibilities. Can your team support these efforts with it's current headcount?
3. Market yourself – As implied above, it's largely political. You need to sell your job role, and keep selling it over time. Teams change, processes and priorities change and new products emerge all the time. You need to make an active effort to market the value of your team all the time. This freaks out a lot of people, who think marketing internally is evil, and confuse marketing with advertising. Look up some definitions, and think about how you approach product design; expressing core values is not that different for your team.

If you still think this is just cynical and short-sighted, so be it. My long-term goal when at the big company was still to get everyone thinking about user experience (and customer experience, and related topics) and I've seen it start working. Design leadership and examples of how you can help, are good ways to start moving your company in the right direction.