A step back for a leap forward
Sometimes you just have to step back in order to go ahead.
This is what I thought when I finished reading the article Gestural Interfaces: A Step Backwards in Usability by Norman and Nielsen.
In this critique, the current interface design of touch-screen devices are really severely criticized. Norman and Nielsen even described this as a “usability crisis”. The title and the first few paragraphs are likely to convince you that the interfaces of iPhones and Android phones we use are so miserable and hopeless. But after reading the rest of it, I think, actually, things are not that bad.
The Problems
First of all, don’t get me wrong. I didn’t mean to say the problems that are pointed out by Norman and Nielsen don’t really exist.
Frankly speaking, as a iPod Touch user, I’ve encountered quite a number of the frustrating experience mentioned by the article, especially the lack of undo. Most of the apps on iOS don’t have anything like that. When you’ve deleted something, a piece of note for instance, with a fancy animation, it’s gone forever. (Well, there might be some way to restore the note, but I don’t even know one after almost a year of usage.) [*]
Regarding Android, I found the menu button is a truly mysterious design. It seems like a compromise between developer-friendliness and user-friendliness. With menus, inexperienced app developers could stuff all the features (which they don’t know where else to put) into the menus instead of building up a terrible interface. But this is a rather strange trade-off to make. What if some genius designer manages to come up with a wonderful interface design that doesn’t need a menu? What should he do with the menu button? I don’t think it’s a clever idea to make the button available on all Android phones just for potentially lazy developers.
Consistency or diversity
The article emphasized a lot on the consistency and standardization of interfaces. Surely there’re so many inconsistent designs because they don’t have a common interface schema. However, I wonder, even in the world of desktop applications, do we really have some standard for all these softwares?
WIMP, you might say. But how many non-WIMP applications are there? A lot, and quite a few of them are carefully designed with great usability. Particularly, PC games have hugely different styles of interfaces and WIMP widgets are not widely used. But gamers don’t find it more difficult to enjoy themselves and spend considerable amounts of time on it. Don’t forget that game UIs are designed for complex tasks and intensive usage. So, designers can do a good job without strictly following all the standards, whereas they can also make terrible UIs with WIMP.
The opposite side of consistency is not chaos. It’s diversity. We are in the middle of them. Standardization for the consistency over all apps is not the silver bullet to solve the problem with gestural interfaces. We won’t do the trick without a good balance. With consistency, we’ve got better overall usability and less learning cost for users. With diversity, we’ll have more excellent innovations and thus usability could improve through continues evolution. It might be correct to say we are in desperate need for consistency for now. But we should not overestimate the power of standardization in the long run.
How standards become standards
I have a very curious question: after all, how does a standard come into being?
We have plenties of engineering (industrial) standards. TCP/IP, USB, SATA, etc. These standards are written by someone and acknowledged among manufacturers. We’ve got very clear specification for them and we need to exactly follow every word in order to make our own products compatible with others.
Are there similar standards for design? This must be a joke. I didn’t heard of any organization like IEEE-SA, IETF or W3C who makes standard documents for design firms (though there’re styles or genres of design).
What we have for usability are actually guidelines or principles for interface design, not engineering standards. Guidelines do not work the same way as the standards. Although many people know about WIMP, there’s no written specification for “WIMP GUI Standard”. WIMP became the de facto “standard” with the joint efforts of all the people who make interfaces.
Before WIMP, we’ve got pioneers who broke the rules of “command-line interfaces” and practitioners who definitely don’t have idea what the term WIMP is. They build up different interfaces with the limited resources they have. The usability improves with the gradual changes until one day the acronym “WIMP” is invented. That is to say, we have WIMP UIs even before we have the term “WIMP”. And it’s not built by someone who wants to invent a “standard”. On the contrary, it’s built by people who are not content with average UIs at that time and who think differently.
Why we have to back off a little bit
Design guidelines are not born to be standards. We need explorations before we discover the path to better usability.
An example is web design. We have the logo on the top left corner in almost all modern websites and it’ll always take you to the homepage when clicked. That defines a de facto standard, but who invented this? No one, otherwise he/she might get a SIGCHI Lifetime Achievement Award for this (just kidding). Interestingly, this standard does not apply to desktop applications, which means, the early adopters of the new standard are exactly the ones who refused to follow established conventions, in the same way as the creators of WIMP.
So, why are we so mean to the adventurers who might bring us to a whole new world of amazing experience? This question could be a bit too emotional, I must say, but it’s reasonable: Even though a majority of the new interfaces will eventually fail, we still have to make this step backward, and open our mind to possibilities in order to come up with new usability guidelines that are optimized specifically for touch-screen devices.
The next step
I’m not here to persuade everyone to give up the old standards. The established standards are still important. It’s not because we must always follow them, but in the sense that they can still guide us on our way of exploration. We should carefully examine the problems that are pointed out by Norman and Nielsen, and work out solutions for them. Thus, we won’t get lost before we find the new path.
I believe we’ll have a bright future of gestural interfaces in the near future. The experience will be far better than desktop applications we have now. (People love touch-screens!) What we’ve established as principles now are probably just small parts within future frameworks for HCI.
To make this happen, we have to back off a little bit. That is what peoples do when they face a gap that you have to jump over. We step back, then take the leap.
[*] Actually there IS an iOS feature called “undo/redo typing”. You can shake your device in some text-editing mode, then an undo/redo option should came out. But how could someone ever know that simply from the interface?
