Linked by Kroc Camen on Thu 31st Dec 2009 14:13 UTC
Microsoft BetaNews writes: "Microsoft executives and product managers -- Chairman Bill Gates, above all of them -- showed great technology vision for the new millennium. The company was right about so many trends to come but, sadly, executed poorly in bringing too many of them to market. Microsoft's stiffness, perhaps a sign of its aging leadership, consistently proved its foible. Then there is arcane organizational structure, which has swelled with needless middle managers, and the system of group competition".
E-mail Print r 2   · Read More · 94 Comment(s)
Permalink for comment 402168
To read all comments associated with this story, please click here.
RE[3]: "Ribbon"
by Bryan on Sun 3rd Jan 2010 06:35 UTC in reply to "RE[2]: "Ribbon""
Member since:

Hmm. There's a part of me that says "just walk away at this point", but why not...

First off, I agree: users shouldn't ever have to worry about the engineering considerations behind a product. In fact, a software product can be considered successful largely to the extent that technical considerations are able to fade into the background, allowing the user only to focus on whatever he or she wants to accomplish. That being said, in practice, that point of view ends up being more of an aspirational ideal than an achievable reality. Inevitably, the context in which a product was designed, as well as many design decisions themselves, will be telegraphed to users once it is delivered.

For example, consider the process-per-tab architecture adopted by Chrome/IE8. Should users care about the differences between threads and processes? Of course not, but as the web moves from static pages to rich applications which are more prone to crash, they will appreciate the resiliency offered by the process model, even if they never understand the details behind it. Design decisions matter to users even when they don't care.

Likewise, any software engineering project takes place within limitations of time, budget, and technical skill. The choices made on these and other variables will greatly affect the resulting product in terms of what is delivered and when. While it may be convenient to think of Microsoft as a company having virtually unlimited resources, it isn't very realistic. The Office team has try to work to deliver functionality within a given time and budget. Given that the ribbon was meant to rectify the limitations of the old interface, it simply didn't make sense expend further effort on something that was no longer serving its purpose. Resources that would be allocated to that effort would be better served elsewhere on the product--areas that would deliver tangible value to users.

Second, I am under no illusion that Microsoft is a benevolent charity that wants nothing but the happiness of its customers. They're a publicly traded company, and therefore legally obligated to increase profit over time. However, it is naive and cynical to the point of small mindedness to assume Microsoft doesn't care about customer satisfaction at all. To reiterate: Microsoft wants users to be as effective and productive in Office as possible--even if only because satisfied users are less likey to consider alternatives if they're getting sufficient value. And even if you insist that the company as a whole is too dysfunctional to be capable of that kind of rational thought, there's no reason to believe that the individual engineers don't take pride in their work and desire to build a product that reflects that.

Your point about constantly changing formats seems a bit out of the blue. The move to XML formats was a one-time thing, and the binary formats have been stable since Office 97. There was a lot of churn throughout the '90s, for reasons both technical and (sadly) competitive, but that doesn't seem relevant at this point.

Concerning the "smart" autohide menus, you're right, but you seem to make the unreasonable leap of implying that any attempt they make at improving the experience is doomed to similar failure. Microsoft understands why that feature worked so badly, and the lessons they gleamed were applied in the design of the new interface. The presentation Kroc linked to is a great resource: Harris covers much of the evolution and reasoning behind the ribbon. If you were to watch that (or browse the blog I linked to earlier), you'll have a good grounding for understanding what Microsoft was working towards with the design. If you want to argue that the ribbon is a failure as an interface, you need to have something less vague and less limp than cries of "it's not what we're used to".

To suggest that the issue was simply that Microsoft did a bad job leveraging the existing UI mechanisms in their products betrays a poor understanding of the problem. Each of the applications in Office has hundreds of commands--a few have over a thousand. Such a broad spectrum of functionality stretched the use of menus and toolbars beyond their limit. The interface would often end of sprawling along each edge of the window, and creating meaningful 32x32 pixel icons of hundreds of commands is a challenge to say the least. Many of the top feature requests for Office were for things that were already in the product, but simply couldn't be found by users.

In contrast, the ribbon consolidates most of the interface along the top edge of the window, revealing additional functionality for various objects (tables, images, equations, &c.) in a consistent and predicable way. Not only are many of the icons larger, allowing them to communicate more information, but the vast majority are all clearly labeled--most of the exceptions are font and paragraph formatting commands, which are pretty much universally understood at this point. Moreover, if you hover over a command, you'll get a tooltip providing any keyboard shortcuts and a short description. And for more complicated commands, such as Excel's conditional formatting, you can hit F1 and be taken directly to the full help entry. A large part of what makes the ribbon such a success over the previous interface is that users spend less time trying to understand the interface and more time exploring how features can be leveraged.

As for OOXML, that's a whole different sh_tstorm, but I'll take a brief stab at it. First, the politics involved were deplorable, but I will point out this was something both sides contributed to. But the end result is that the default formats for Office documents are open standards that can be implemented by anyone, and I think that's better than the alternative. Really, the biggest advantage of OOXML over ODF is that it enables the older binary formats to be reliably converted to an XML format with full fidelity; ODF wasn't designed with those formats in mind, so conversion may have been problematic in many cases. ODF may well end up becoming the de facto standard in the long run, but in the meantime, OOXML is an important transitional format if nothing more.

Finally, your point that there is no one right answer for UI design is well taken. However, just because there may be more than one right answer doesn't mean there isn't at least one wrong answer. For an application as broad as the ones in Office, the evidence seems to strongly suggest that menus and toolbars are no longer the right answer--or at least no longer the most correct answer among those that have been tried. That was less true 10 years ago, and completely false 20 years ago, but today I think that's true. It is also true that some people can be incredibly productive in programs such as vi and Emacs, but Office is targeted towards a broad audience and needs to be optimized for enhancing the productivity for as many of them as possible. For Office, the ribbon has proven to be a far better mechanism for doing that than the system of menus and toolbars it replaced. I base my opinion not simply on *my* personal experience, but on a deliberate effort to understand and compare the merits of each approach. (The ribbon was introduced when I was writing a senior thesis on user interface design, so it caught my attention.) This isn't a matter of individual preferences as it is cognitive and perceptual limits that are fairly consistent across the population.

...Christ. I swear, I'm pretty reserved in person. :-)

Reply Parent Score: 1