Notebook

Fastco interviewed Chrissie Brodigan, design and UX researcher at GitHub to talk about how they're using deprivation to study people's reaction to removing elements from a product.

The gist of the technique is to work with a set of participants who develop a pattern of use around a feature or design characteristic and journal their experience. After some time, at a point when they've had time to be accustomed to that thing, they take away or alter it, and observe through the journal what the users' reactions are to its removal.

It seems like A/B testing with a greater focus on qualitative measurement. Deprivation could be a powerful factor and tool for assessing value of features on both new and existing products. Designers and developers get to measure the features by removing them, and seeing how upset their users get within a controlled group. Removing features from an existing product is difficult, but the upside in measurement could be that features that have a cost to the user or vendor might turn out to be unnecessary, or the research could lead to improvement.

This is technique is new to me, but I'm not unfamiliar with the pain of having features removed for me, or the backlash you can experience when you change something as a designer.

I don't like public speaking. I love talking to people one to one and in very small groups. But public speaking and me don't mix, and if you've ever seen me speak, you probably agree. In the handful of times that I have spoken publicly, I think only two were really worth coming to see and hear.

I believe the only reason the worthwhile talks were not utter failures was because I had something to share that I think people wanted to hear, and could get something out of. They really had little to do with me. I think those talks had more to do with what the audience could do with what I shared. My passion about the topic let me forget about myself until I was just basically just giving people something they wanted instead.

Kathy Sierra is one of the people whose articles I like to read and presentations I like to watch, because I come away with ideas that I feel I can do something with. There's something almost pragmatic to me about what she shares. Don't get me wrong though, she is a charismatic, funny, and super smart speaker. But her post about public speaking and presentation skills is useful to me as someone who hates public speaking, and always feels bad turning down the invitation.

Here's the message: If you imagine that your role in speaking is to be the delivery mechanism for the message/story/whatever, you become the UI for the User Experience of that exchange between you and the audience.

Ultimately, I think this thought sums up what the role is about:

It's not really about the user experience they have during your presentation. Like your presentation, their experience of it is also just the enabler for something bigger. Because what matters most is NOT the UX but the POST-UX UX. What happens after and as a result of the user experience? ... What happens after what happens happens?

When they walk away from the user experience, then what? Are they different? Are they a little smarter? Are they a little more energized? Are they a little more capable? Are they a little more likely to talk to others about it?

Maybe there's hope for me after all? I like this idea of the post-UX-UX, when it comes to products or presentations. At the end of the day, while the delivery and execution are very important, your presentation skill holds much less weight than how you've impacted the user.

And this brings me back to why I got into working in this field to begin with. I started professional life working on libraries, and wanting to help people find information that helps them do things that fulfill needs in the real world, outside of the experience of using the library. What I do still comes down to that, and I guess public speaking should be no different.

I'm still afraid of being the awkward kid at the head of the class, even if I know the material. But maybe next time, I'll just imagine myself as the UI, and the rest will fall away.

If you take a lot of screenshots on a Mac like me, you probably have a good workflow for doing it using something like LittleSnapper/Ember, Awesome Screenshot or Skitch. The biggest problem I've had is that after I take a screenshot, I often want to do something with it. I can do this by just getting Cloud to watch and upload files passively, but I've always wanted to save to my own server after just issuing one command to capture the page.

I love Alfred. Launchers like Quicksilver and Alfred are what I think of as the best minimal UI apps that I use on the Mac. I finally got around to searching for a way to capture a full scrolling screen from it and do something with the output.

I found an excellent Alfred workflow by user moom on the Alfred Forum that uses a shell script to run Webkit2Png in the background. It's awesome. I just copy the URL of the page I'm on, summon Alfred with a few keystrokes, and then it gets captured, optimized, and uploaded to my server in the background. I made a video of the workflow showing Alfred, Hazel, ImageOptim, and Transmit working together with lots of spit and chewing gum in case anyone else has this need.

It's a lot of steps to set up, but if you have this need, you'll appreciate it. ;) You can modify the steps for your workflow to move it to note in Evernote, to Cloud, or wherever your heart desires. Download the workflow here.

If you're interested to know how I modified the workflow, these are the options I used:

python webkit2png -F -W 980 --delay=5 "$(pbpaste)"
ls *.png | grep -v icon.png |
xargs -I {} mv {} ${HOME}/Sites/media.konigi.com/snaps/2013/

-F -W 980 outputs a full scrolling screen at 980px wide. You can see what the options are by running this command in the terminal: webkit2png --help.

Jackson Fox and Jeremy Fields did an excellent 2-part write on the use of "infield" labels at Viget. There are several techniques for placing labels inside the input box. For varying reasons we may want to avoid the label on top or side positioning, but as Jackson points out, the infield placement can be problematic.

Part 1 talks about the problems using infield labels. Part 2 shows how they deal with the problem by using a visible tooltip that doesn't obscure the input, with a jsFiddle snippet to play with the code. You can download the jquery plugin from Git.

Not a lot of bookmarking this month.

SIX UX is a gallery of transitions and animations shown as 6 second videos. Andreas Popp combined Vine app videos with tumblr to showcase otherwise ephemeral movements found around the web. Low fi and awesome. Check it.

Jakub Linowski, creator of Wireframes Magazine just started writing a newsletter called Good UI on how to design sites that are easy to use and also yield high conversion rates.

He's looking at how to employ practical ideas that satisfy the business side as well as the people using it, which I'm sure is something we all want to do on sites that sell a product or service. He plans to publish the newsletter monthly. Check it out.

I've been looking for ways to energize my creativity in between the longer stretches of time spent on difficult, routine, and less creative work. The thing that has been doing it for me lately is working with static html on some of my smaller production sites.

I was playing around with static html tools for a little side project I've been toying with. I was tinkering with Stacey and Kirby. Awesome tools for static PHP page building. I was also playing around with CodeKit, which looked like the most promising tool for little projects. But I figured it was all just playing and there wouldn't be a real static site project for me to work on for a while probably.

Then last month we launched UX Apprentice using WordPress. We were on a stable server that hosts some pretty massive sites, so I was happy. But then, as web hosts always do, it went down for a short period. Some rare issue on load balancers was the cause we were told. Downtime happens. But then while discussing this issue I said, "Fuckit. This site is so simple, let's make it static and put it on an S3 bucket."

A week earlier I was playing with Hammer for Mac, the latest static site builder I had found. Among my options, including just doing it all by hand, this was the simplest of them all. I looked at what I would need to port the site to Amazon's S3 service for websites with no fuss. All I needed to do was suck the WordPress site down and convert it to static HTML.

I know this is small potatoes compared to the bigger things people are doing out there, but I got a kick out of doing this.

Here's how I did it, in case anyone out is considering doing the same:

1) Download
SiteSucker pulled the site down for me and converted it to static HTML. You can do it on the cheap with wget.

2) Create a Hammer Project
Look at your downloaded site, and clean out the cruft. Open Hammer for Mac and create your project. Open the source folder.

3) Organize files, use HTML editor to convert PHP to Hammer-smart stuff
I then moved my img, css, and js folders to the new project's assets/ folder. When Hammer builds, it reads your entire project and then knows where everything is so you can use the @path directive to reference files only by name, and include styles and css simply by name without paths using the @stlylesheet and @javascript directives. I cleaned all files up to use this.

I did the find/replace grunt work using Coda on my CSS and html files. Replaced the PHP that I use in my header to use the @include directive and the $title variable. My SCSS files got picked up and compiled with no problems. I then used the @path helper on images in my SCSS.

About two hours of this clean up and it built with no errors.

All my SCSS was compiled, my includes worked, and my CSS and JS were optimized (combined and minified).

4) Testing
I set up my hosts on my machine so I could test locally in the browser so all the TypeKit would work after I added that domain. I also learned that you can't test Typekit with a url using the file:/// protocol. Bummer. I already use VirtualHostX which makes simple work of managing multiple hosts on your local machine. It edits your httpd-vhosts.conf and hosts file. I had bought it originally because the web sharing option went away in OS X, and I'm too lazy to edit files just to turn PHP on.

Then of course, I tested on all my browsers and tested the responsive layout. We were ready to go.

5) Publish
Publishing on S3 is pretty simple. A normal person just needs to set up a bucket and use an S3 aware FTP app like Transmit and you're done. We ended up commiting it to a repository that is watched by a server that kicks off updates to S3 for pages that have changed.

That was the whole mini-relaunch that happened quietly after we launched UX Apprentice. We were live on WordPress for a week, and now we're hosted on an AWS instance and I expect the uptime will be awesome—as good as any S3 site. We do the same for our main balsamiq.com site and every time I see our blogs report downtime, I feel this wave of relief knowing our static pages on S3 are happily being pushed out without issue.

Anyhow, that made for an exciting day last week. It's energizing to feel frustrated by something and quickly turn something around that's better. And for anyone that's doing UX work that's looking for a decent HTML prototyping workflow and knows enough SASS and JS, Hammer may be the thing for you.