1) http://www.google.com
2) http://www.google.com/advanced_search?hl=en

The default UI is, of course, #1. The problem with enterprise applications is that we expose advanced UI to all users, even the users who need to do simple task (like approve/reject/submit) in the application.

The ideal approach is to design two UIs…one for advanced user and one for ‘business’ users. The fact that business users already are familiar with one ‘complex’ UI i.e. Exchange, Lotus Notes, etc, the application designer can take advantage of that and provide the UI for business users in Lotus Notes/Exchange. This way the business user never has to leave his/her Inbox to do simple task in the enterprise applications.

If you think about it, Google does the same. Most people don’t go to google.com to do search…they simply search Google from the search bar in their browser.

1) http://www.google.com
2) http://www.google.com/advanced_search?hl=en

The default UI is, of course, #1. The problem with enterprise applications is that we expose advanced UI to all users, even the users who need to do simple task (like approve/reject/submit) in the application.

The ideal approach is to design two UIs…one for advanced user and one for ‘business’ users. The fact that business users already are familiar with one ‘complex’ UI i.e. Exchange, Lotus Notes, etc, the application designer can take advantage of that and provide the UI for business users in Lotus Notes/Exchange. This way the business user never has to leave his/her Inbox to do simple task in the enterprise applications.

If you think about it, Google does the same. Most people don’t go to google.com to do search…they simply search Google from the search bar in their browser.

The reason enterprise application designers don’t go the ‘extra mile’ and create a second UI that is integrated within the UI that business user already lives in is probably because of resource constraints and lack of integration platforms between enterprise application platforms and Lotus Notes/Exchange.

I am with SAP and we are trying to bridge this gap with Alloy (for Lotus Notes) and Duet (for Exchange).

Reducing pilot error and workload is a high-stakes affair, so one might expect that interface design lessons learned in that arena may say something about design. I think they do say something just like what you did about the new Office: put things where people have been trained to expect them or risk problems.

All those cockpits share certain elements, because moving them around could be catastrophic, so your statement that “The risk of this is something to consider but shouldn’t deter you from redesigning a product to make it truly more efficient” requires more qualification.

In fact, many things we (people) do in life are based on seemingly arbitrary patterns set by various past events. Sometimes change costs are too high to make a change worthwhile, and that math is not always rational.

Hiding complexity may or may not be a good idea, depending on the situation, and doing it is easier said than done, as the effort creates its own complications.

I return to my original point: complex tasks require complex interfaces, all other things being equal. The goal of good interface design should not be in reducing complexity per se, but on shaping the learning and labor demands of the system to best match the rewards provided for each user.

In my mind, that means that the interface should teach and learn, both implicitly and explicitly. Software has been focused on the former, but needs to move toward the latter.

I have been impressed with Authoria’s approach- while not ideal as executed today, they are thinking along the right lines and I’d like to see more of that goodness in our own offerings.

That said, from years of experience actually having to sell these things on the street, I think the most important thing to remember is that the job comes first. If a tool cant do the job, all of its design features mean less than nothing.

I think flight simulator/cockpit interfaces are at an extreme end of the spectrum just like Google’s search interface is at the extreme simple end of the spectrum. Most software falls in between those two. What you think of as a complex business application can certainly have a simple interface. One reason that it is hard to think of an example is because in really good interfaces the complexity is hidden so you might not realize that a simple interface is on top of a complex system.

Who says complex applications require complex interfaces ? I said that the more complex the activity, the more complex the interface will need to be (one way or the other).

Instead of pointing to simple activites supported by simple interfaces, how about some examples of complex activites supported by simple interfaces ? Oh….there arent any that come to mind ?

Click on this link to examine the work of some of the world’s best designers in trying to present the very best interface to handle a complex set of tasks, and find one that is radically simple…..oh yea, when life or death is on the line too …..

http://www.flightsim.com/cgi/kds?$=main/howto/getreal3.htm

Saying that complex applications require complex interfaces is just an excuse to not fully think about human behavior. Having the right people who specialize in this field (such as interaction designers) is necessary to create simple interfaces to complex applications. The result might be that the most common goals are very easy to accomplish while less frequent goals are more difficult. This is perfectly acceptable.

If you look at the new Microsoft Office UI you will see that they used exactly this approach… and they were right to bring in the experts at Adaptive Path to help them do it.

However, many things in life are not simple, and recruiting and applicant tracking fall into the not-so-simple activites. Wishful thinking extends to the desire to push a button or type a word to complete the job, but that is not happening right now.

When you look at even the very latest interface design for something like say…..a Boeing 787 dreamliner…..you are still going to have hundreds if not thousands of function points.

When the real-world reduces everything you have to do into one simple step, you can bet that most software will reflect that change.

Now instead of the frame presented by Anu, which is really a waste of time, we should be talking about how to embed training for both the job and the technlology directly into interfaces.

Actually that frame is more than a waste of time- its destructive because it leads people to buy underpowered systems all the time because they look easier to use.

Of course, this does not speak to the notion of supressing anything a user is not needing at a given moment, which does reduce clutter. Recruiting however, being a people business, will tend to require multi-tasking and some situational ambiguity, which means leaving stuff on the screen.