Tags: patterns and widgets
Recently Jensen Harris, Lead Program Manager of the Microsoft Office User Experience Team, gave a very excellent presentation to BayCHI Jensen’s Office UI Blog”>(get it here).
I have also been following closely Jensen’s Office UI Blog with great interest. At this point though I have a few questions on a theoretcial level:
First off, when I look at the changes that the Office 12 UI is going through, my first visceral response is “Yeah!!!! It’s about time.” I also want to add that it takes a lot of courage in this environment to make such dramatic changes as those that are to be unleashed onto the world.
The story in Jensen’s presentation is compelling and convincing. He describes in enough detail to give us a sense of the grueling analytical processs they went through to even begin to take on this task. It is a case study of sorts for all designers to closely review and take what is relevant for your work.
Jensen says that they had all this data to collect, but he didn’t say the source of that data at all. Is this something that all installs of Office 2003 send out?
My question as the title suggests is around a fundamental tenant (so I am under the illusion) that it is good to keep interface objects more static, so that it is easier to remember where they are when you need them. I was taught that this concept is muscle memory.
Because of the contextual nature of the application, the “Ribbon” as it is calledis never static. It is constantly changing depending on what objects you have selected. it could be said that the items in any given context are always in the same place, so that it is not breaking that rule, but I was taught that it is ok to have options that cant’ be used now, so that when i do need them I’ll be able to find them. They just have to clearly indicate that they can’t be used now.
Of course MS already got rid of that idea when they created the expand/collapsing menus in Office 2003. At least I could expand if I wanted to, but who did that, right?
I have not had opportunity to play with Office 12 yet. I’m not on the beta (though I’d love to be). I did have the opportunity to use Windows Vista, which takes on a similar approach to having a highly contextualized menu system. Different, but similar enough to extrapolate from. I have to say that at first I was incredibly thrown by this interaction model. it was distracting and took a very long time to get used to.
What I had figured out was that the decision for these types of risks around learnability is linked directly to the posture (Cooper) of the user’s relationship to the software being used. In both the case of Vista and Office, the posture is one of sustained (sovereign; Cooper) use. As Jensen says in his presentation, “People spend more 1-on-1 time with Office than they do with their spouses.”
Usability, and most evaluation models used today, are not really adept at testing for usability across different postures. They are usually run against a single user-base who are first experiencing the application. Even site visits later in the product lifecycle process do not get a real understanding of the long term effects of different user contexts.
There are two reasons that possibly these changes will be successful:
1. Scalability of the products complexity mandates this and maybe muscule memory needs to be balanced and not considered a rule for all contexts.
2. The context is well understood. We have been at this for well over a decade with lots of opportunity to capture datapoints through subtle change. It seems well-founded to make such a change now.
3. The web – Yup! the Web affords us new ways of thinking about GUI. Not everything we do on the web is nearly as uniform as the applications that we use on our desktops and we use the web as much if not more than desktop applications. There isa growing balance between what is effecting what.
What are your thoughts?