Interaction

Everyone’s hyped up about getting a Google Wave invite. I have one and I don’t see what the fuss is about. Yes, I’ve seen the YouTube video, and yes I’ve watched the Developer Preview video too. It looks great and all, but seriously folks – it’s one complicated thing… next to email.

And this is why Google Wave isn’t about email. The same way computers aren’t (just) about calculators.

What it is is real-time collaborative messaging with historical playback. Which, granted, is very good for real-time anything. Except that real-time in human terms is highly contextual.

Because people have better things to do besides sit in front of their computers all day long, they are much better off carrying Blackberries and having push email rather than collaborative messaging. In my opinion, push email works just as well – because you don’t need to be in front of a computer to communicate with your colleagues.

Wave is also a tool. Meaning that the interface itself is not the message. In order for users to synthesize information, they have to make sense of it in a collaborative sort of way, which means paying attention to whatever is happening in that messaging window, which takes up at most 1/2 of the screen. It looks and feels like, a video player – but for messages. Think of IM on steroids, not email.

Email works because it’s mind-dumbingly easy. And because of that, nobody needs to take 3 week lessons on how to use email – and that includes your grandpa. And because it’s so atomic, it can easily be transported as a message to any other atomic messaging forms – SMS, blog post, forum post, IM chat, whatever.

Waves, on the other hand, are almost asynchronous streams – anyone can start writing content at anytime into a wave, and hope that it doesn’t make a mess of the conversation. Last I checked, people still pay attention to body language and subliminal messages, both of which are devoid in Wave but affects the way people communicate collaboratively and socially together – i.e. turn-taking and all that jazz.

Which part of a wave is atomic? You have to spend effort to calculate and decide. Then you have to find a way of “clipping” that message – and how do you make it portable? Can you cut and paste portions of your wave onto your blog? SMS? can you print it out?

Like email, most of what is published gets fixed in time. But a Wave, is a collection of content played out over time. Just like how photographs put together over time become videos. But with the added complexity of multiple sources of photographs, it becomes more and more difficult to separate and organize content in a Wave.

And that’s why Google Wave is not about email.

Sites and Articles Relevant to this Post:

• It’s Just Fancy Talk
• EasierToUnderstandThanWave

boon :: Oct.13.2009 :: interaction design, technology, tools, web :: No Comments »

I’m currently in a team of 3 people working on a redesign of a website. To me, this feels like my first “real” design project, one where I’ve initiated without any requirements for software implementation. Instead, we began to ask ourselves who are our users, and what exactly are we trying to communicate to them?

User-Centred Design – Using What Works For Me

I felt that a lot of “formal” usability methods didn’t help me much, here. I felt somewhat trapped by procedure and process, thinking about best practices for wireframes, personas, user journeys, etc. Having been exposed to too much user experience jargon, I realized that user-centered design isn’t necessarily synonymous with visual design. In fact, each stakeholder seems to have a different understanding of who the user is, even if the differences are slight.

What ended up working for me was to put down the questions I felt the users would ask (while using the interface/application) onto little bits of post-its, and looked for patterns like organizing those questions into related categories. I organized this in several ways:

“User Journey”-type Things and “Site Map”-type Things

One way of visualizing the questions was to list them out in the style of a “user journey”, which were lists of questions laid out in a temporal fashion (e.g. question A leading to question B as a user traversed deeper into the application). This helped me understand which questions to address first, and which to address later. It helped me prioritize my workflow, and make decisions on which interfaces to work on first. It was already obvious that we needed to focus on designing the homepage, but there were other pages that weren’t immediately obvious, such as a page to help users to “learn more” about the product – that page only came about after looking at the questions from the user journeys.

Another way was to group the questions that were related to one another, and label them as categories. That helped me to visualize the overall site layout, because I absolutely hate building software in the dark – and that includes building on software requirements without understanding the whys to the point I’m convinced of its business and user needs.

At this point, I hadn’t used any formal methods I had learnt from a class, website, or book. It was good that I understood who I was designing for, and I wouldn’t have been able to do this without classes, websites and books – but the formal methods were pretty useless. I just used what worked for me.

We Used Fireworks, but I can’t Recommend it to You

It’s important to note that two of us jumped onto paper and pencil and did wireframe sketches, and then designed a hi-def visual of the homepage on Fireworks (which I now really love) – only because we were three people and we couldn’t quite picture the outcome in our minds. At the same time, we needed our boss to make a decision and he has a lot of emotional attachment to the look and feel of the site, being the founder and all. So it was important to get those designs out.

I don’t think I can recommend the same process to other designers – it really depends what you’re designing for. Those fireworks designs took up several days and a lot of it was thrown out of the window as we iterated from one design to another. But I think that’s necessary in design, sometimes – because the interface is what users will end up seeing and using, again and again.

Sometimes I read articles from other designers and it all sounds really cool and makes you feel that there’s some kind of order or process in the design of an interface. But I find these things hard to use because it’s the designer that makes a difference, not the method. It’s about knowing what you know that works – a process or method can’t save you. They’re just tools.

boon :: Oct.04.2009 :: interaction design, tools, user experience :: No Comments »

I’m at the start of a new phase of our development, and we’re at the point where we need to generate ideas for our interfaces. I’ve observed that wireframes aren’t always well-received by people who don’t understand them – clients who think they need more color, developers who want to code over it, etc. If the only resolution we have to this problem is educating the public, I think we’re in for a very long ride.

This might be because wireframes aren’t obvious enough to people.

You heard that right – if people are misunderstanding your wireframes for a finished product, or your sketches for a placeholder for a WYSIWYG dev tool like Visual Basic – there’s clearly something inappropriate about the tool’s use.

I believe some of us take wireframes for granted, such that we see no other way of implementing a product. While wireframes work well for some designers, I sometimes think they take up time and that there might be better, faster, more efficient ways to build software.

Also, wireframes aren’t great for super-dynamic AJAX-heavy interactions. And they aren’t great if you want to break out of the 2D mode, and work with more dimensions (how do wireframe a keypress, or an animation, or audio?).

I remember the first time I used tons of wireframes on a project, and ended up realizing that we wasted all that time building hi-fidelity prototypes just because I felt it was the right way to design an application. Time was wasted testing the prototype because we didn’t get THAT much insight out of it, simply because it was so blatantly obvious – we should have just tried it on ourselves before testing it on other users.

I’m probably stepping on some UX no-go zones – aren’t we supposed to test apps on real users, and not ourselves? Well, I think there are some things which are so blatantly obvious you needn’t waste the users’ time to fix or test. Testing via paper prototypes a la Carolyn Snyder isn’t always going to save you time. There are much faster ways to get it done, if you think through it a little.

I’m not advocating against wireframes. I’m just advocating against getting locked down on a specific method – against “methoditis” – you know, when you start believing that one method is better than another just because.

Basically, you might want to re-think wireframes if they:

• confuse the people you’re communicating to
• take up time when something else can be so much faster
• cut down too many trees

Here are some examples or ideas of doing without wireframes:

• http://gettingreal.37signals.com/ch11_Dont_Do_Dead_Documents.php
• http://code.new-bamboo.co.uk/polypage/ (a tool developed by Clearleft and New Bamboo)
• method-acting (see Bill Buxton’s Sketching User Experiences for an example a designer “being” the device)
• plasticine? feel free to add your own examples.

boon :: Sep.21.2009 :: interaction design, tools :: No Comments »

What I like about A/B Testing using Google Website Optimizer:

• It helps provide real, direct data (as opposed to sheer guesses, or “eye-balling” secondary sources of data)
• It helps you build strong skills in testing
• It’s fairly straightforward to setup
• It’s fairly flexible – allows for HTML markup, etc.

See how 37 Signals puts A/B testing to the works here.

See Carsonified’s blog post about doing A/B Testing with Google Website Optimizer on Wordpress here.

For a quick overview, watch the video from Carsonified below:

How to do A/B Testing with WordPress

boon :: Aug.03.2009 :: evaluation, resource, tools, web, websites :: No Comments »

I’m employing a diary study for my MSc dissertation on image search, and I realized I was doing it all wrong today when I met with my supervisor and project sponsors. Well, okay I wasn’t entirely clueless but I did make some pretty bad assumptions about diary studies, and these were the lessons I learnt.

1. Don’t assume people remember what they noted in their diaries

I started off telling my participants that all they needed to do was fill in their diaries with descriptions detailed enough so that they would be able to recall the activities. I told them the study would last two weeks, based on some ballpark figures I discussed with my supervisor, and I told them that I might give them a call every now and then to “see how they were doing”. In a recent meeting my project sponsors and supervisor suggested that I should call them daily, because there’s a high chance that the participants would forget a lot of the rich detail that you want to get in a diary study interview.

I had to correct my slip by contacting every one of my participants to ask them permission to call them on a regular basis, and thankfully the ones who have gotten back to me so far have been totally cool with that because they were acquaintances or friends. But it was just bad foresight on my part.

2. Be as specific about your population as possible

When I sent the word out that I needed participants, I had a lot of friends were willing to participate because they had they time and they didn’t mind doing it (plus there was a £40 incentive). I didn’t think too much about the demographics of the population and I let in two participants who were edge-cases – a photographer and a graphic designer. In my study about domestic image search behavior, they’d stand out in a crowd. Thankfully I had other participants who were more “normal”, but I would’ve saved myself the embarrassment had I spent a bit more time thinking about who I should’ve been recruiting.

3. Research requires testing too

I assumed that people would quickly understand how to fill in the diary forms, because they were relatively straightforward. I was promptly advised to call in and check with my participants so that I could get a quick “feel” for the differences in behavior and data, and adjust the study process accordingly. Some participants, for example, may be more active internet users than others, for example – so they would need less “checking up” since they would have a lot more stuff to report after a few days of diary-filling. I initially thought that calling them they day after I had spoken to them was too soon, but there are ways to inform the participant of how often you need to get in touch with them.

4. Don’t believe everything you read in books

I guess it’s not all method-driven and form-filling. There’s a bit of an art to this, so I guess it takes a bit of practice to get a hang of it. My supervisor/project sponsors were forgiving and kind enough to guide me in the right direction, but I could’ve saved myself by asking more questions, talking to the right people, and not just basing my ideas off books.

update: I’m still looking for a few more participants, preferrably male aged 40 to 65. If you’re interested, please email me at boon [dot] chew [at] gmail [dot] com. For more info, click here.

boon :: Jun.25.2009 :: tools, user experience :: No Comments »

I’m working at a small startup company and we’ve just got ourselves onto Yammer and I’m really getting into it. I use Twitter as well for my regular stuff, but Yammer has me posting all sorts of ideas and thoughts and feedback about the work we do.

We basically use it to state what we’re currently working on, respond to each other’s comments and ideas/rants, post up screenshots of stuff we’ve seen (inspirational sites, examples, etc.), ask questions, etc.

It’s like IM, but I think it works better because it’s like a blog as well. Yammer sends you a summary of the day’s posts, so that you can catch up with what’s going on. I find that it works not just for remote-working solutions but even for team-related tasks because you can be working on different things but not actually see each other’s work.

I like the fact that it’s closed – so that no dumb-ass spammer can find my profile and send a request to follow me.

I also like the fact that I can upload stuff directly into my posts. That way my readers can quickly open up a screenshot, etc. Sure twitter has twitpic, but that requires opening up a separate browser window, etc. With Yammer, all I had to do was use Fireshot to grab the screen capture and save it onto my hard drive, and get Yammer to upload the file from there.

I’m not a big fan of the clients they built (PC/Mac, etc). But it’s functional, and I can use it. The interface doesn’t look and feel as nice as Twitter’s, but I guess again, I’m not really complaining.

I don’t know of any other folks using Yammer, but if you are I’d love to hear what you think.

boon :: Jun.21.2009 :: social web, tools :: No Comments »

Update: Rey alerted me to Yahoo! YSlow

This looks interesting:

http://code.google.com/speed/page-speed/

Curious to know how effective this is towards usability/interaction design.

Blurb from site:

Page Speed is an open-source Firefox/Firebug Add-on. Webmasters and web developers can use Page Speed to evaluate the performance of their web pages and to get suggestions on how to improve them.

Page Speed performs several tests on a site’s web server configuration and front-end code. These tests are based on a set of best practices known to enhance web page performance. Webmasters who run Page Speed on their pages get a set of scores for each page, as well as helpful suggestions on how to improve its performance.

By using Page Speed, you can:

• Make your site faster.
• Keep Internet users engaged with your site.
• Reduce your bandwidth and hosting costs.
• Improve the web!

boon :: Jun.12.2009 :: tools, web :: No Comments »