Ubuntu 11.10 and Old UI Design

Over the weekend I updated my development VM to the latest Ubuntu release, 11.10. The big change in this release is the Unity UI as the default. My understanding is that the release prior also had Unity as the default, but it was only enabled on machines that had 3D acceleration.

My overall impression: I like it.

Long ago Garret gave me a quick demo of Quicksilver on the Mac. Beagle was still the search engine of choice and apps like Banshee were very much in beta. It was the sort of tool that when you became familiar with it could easily become indispensable. It was a theme that we used at the time when we were redesigning the Novell Linux Desktop start button menu. The idea was that you’d get something very similar to what Unity is doing now.

Fast forward to today and the world of mobile and it is interesting to see how the ideas we were building off of have become more reasonable to the masses. Now the app of choice on OS X seems to be Alfred. Mobile applications don’t have a button so to speak but use a lot of the same type of search technology but instead of finding things, it attempts to help complete thoughts in order to reduce the need to type.

Going back to Unity, what I like is that the concepts of Quicksilver and Alfred are first class citizens. I also like how the maximize works as it feels a little like the tiled window managers. The UI gets out of your way and you can focus more on the applications.

I should also mention that because I’m using it within a VM, that I’m probably not get the full experience. I usually only have 3 apps running, Emacs (of course!), a web browser (Firefox) and thg (the tortoise hg tool). All of these apps have their own models for containing different projects and organizing their sub-content, which might reveal why I care more about the UI of the application than the UI of the shell.

On a side note, 11.10 sets Chrome (Chromium actually) as the default browser. This is kind of funny because I had been using Chrome pretty exclusively for a while, leaving Firefox. When Firefox switched over to the short release cycle with its major release I gave it a try. I dedicated a day or two to using it and the big feature that sold me on it was the tab grouping. You can group your tabs and switch between tabs via a keyboard shortcut that works kind of like Expose. The nice thing about it is that I can once again start typing and it will highlight the tabs according to my search terms. The result is that I can dedicate a set of tabs for things like “work”, “music” and “reading” and switch back and forth as necessary using the keyboard. The grouping also allows me to avoid distractions by keeping things like reader or social networks on a whole other tab group.

I’ve always tried to avoid the mouse and while it is often the best tool for the job, there is something to be said for finding better models. Typing a word may not seem very efficient but I can type “fir” and hit enter to pull up my browser without having to negotiate where my cursor is or what icons are available. As we continue to evolve our systems to include larger datasets and most likely more space for UIs, having a way to be accurate may very well trump the flexibility of the mouse and pointing.