Usability expert Jacob Nielsen was interviewed by Builder AU. When asked about the 3D interfaces of the future Nielsen says “As long as we’re stuck with a 2D computer screen that means the information needs to be in a 2D medium. Complete article.
Jakob Nielsen Interview
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
I’m certainly no fan of Nielsen. He’s a strict utilitarian and minimalist… and in my opinion, utilitarianism is for people who don’t know how to have fun.
“As long as we?re stuck with a 2D computer screen that means the information needs to be in a 2D medium.”
That makes about as much sense to me as “As long as we have uniplane displays information must be presented in a planar, non-overlapping manner”
He hits the nail on the head with his description of usability issues of open source software. As a general rule the usability of proprietary software is substantially better.
Want to make a usable product? Consult your users. You _know_ how your product works, but is it obvious to them?
Don’t throw too many features into your product, or it will become unusable.
Open source software must put more effort into innovation than into cloning, or the commercial industry will stay ahead.
Don’t put 3D interfaces on a flat surface. It’s only going to complicate things.
Customers don’t like litigation. If your customers feel they are at risk by using your product, they may well leave.
That’s a basic summary of the article. It’s good to hear an expert saying things I have been thinking and saying.
He’s a strict utilitarian and minimalist… and in my opinion, utilitarianism is for people who don’t know how to have fun.
Nielsen (http://www.useit.com/) is a (web) usability expert, not trying to give any advice on designing cool, fun or artistic looks, like, say, David Siegel (http://www.killersites.com/) is. There’s room and need for both, and no reason why you couldn’t combine both kind of advice in one product. But if you want to learn about (web) usability, rather read Nielsen than Siegel.
“As long as we?re stuck with a 2D computer screen that means the information needs to be in a 2D medium.”
What’s wrong with that idea? I feel that quasi 3D UI like Sun’s Looking Glass is more about trying to achieve something that looks very cool than trying to achieve better usability. Maybe things like Looking Glass could provide some real usability improvements in future too (besides of the 3D effects only eating much more system resources and slowing performance down…), but there haven’t been too much proof of that yet available AFAIK.
He hits the nail on the head with his description of usability issues of open source software.
That’s too true, I’m afraid…
…I tried to find Siegel’s old site in order to give a chance to compare Siegel’s and Nielsen’s ideas better, but couldn’t find it first. Here it is:
– David Siegel: http://www.dsiegel.com/index.shtml
(and Nielsen: http://www.useit.com/)
Siegel is certainly more fun than Nielsen. but guess who knows more about web usability? (maybe that also has a bit to do with the fact that Siegel’s current personal web address http://www.dsiegel.com/ links to a a very plain looking, “Nielsenian” web site instead of the old “artistic” http://www.dsiegel.com/index.shtml site…)
It’s silly to say that because we’re limited to 2D displays that we should limit presentation to two dimensions. Paper has been 2D for a very long time, and that doesn’t stop engineers from doing perspective diagrams.
Actually, if you really want some advice about web usability, look for Pete Pirolli. He works at Xerox Parc (as a HCI expert) and is way more respected than Nielsen. http://www.parc.xerox.com/istl/members/pirolli/pirolli.html
Nielsen tends to concentrate upon the “ten top tips” style of usability advice rather than looking in detail. A glance at Pirolli’s work will show you that HCI and usability is way from being simple.
“Many of the next generation operating systems and interfaces such as Apple’s Mac OSX, Longhorn, and Sun’s Project looking glass have 3D interfaces. What are your thoughts on the 3D interface?”
Gee, what is this guy on? Is OSX really similar to project looking glass? I don’t think 3D means glass looking scrollbars and drop shadows.
Nielsen tends to concentrate upon the “ten top tips” style of usability advice rather than looking in detail.
Not exactly true…:
http://www.useit.com/jakob/publications.html
But the Top 10 tips lists are useful for those who need and want short summaries (only).
if you really want some advice about web usability, look for Pete Pirolli. He works at Xerox Parc (as a HCI expert) and is way more respected than Nielsen.
To me the two guys don’t seem to compete with each other with their views but rather emphasize quite similar principles, though maybe study things from different view points sometimes.
“Is OSX really similar to project looking glass? I don’t think 3D means glass looking scrollbars and drop shadows.”
I think he’s confusing 3D card acceleration with 3D interface. I don’t know what else he might be thinking.
He hits the nail on the head with his description of usability issues of open source software. As a general rule the usability of proprietary software is substantially better.
You might find he would say that proprietary sofware is not substantatially better but that there is more proprietary software that is better than open source software. Most of the software in the usability hall of shame is proprietary.
Nielsen (http://www.useit.com/) is a (web) usability expert, not trying to give any advice on designing cool, fun or artistic looks, like, say, David Siegel (http://www.killersites.com/) is. There’s room and need for both, and no reason why you couldn’t combine both kind of advice in one product. But if you want to learn about (web) usability, rather read Nielsen than Siegel.
I contend there’s room for both usability and style, and OS X certainly has done an excellent job in that respect.
Nielsen has maintained one of the ugliest web sites on the Internet (http://useit.com/) and consequently I could care less about his design advice, as “usable” as it might be…
First of all, there ARE 3D displays available to the public today, so the 2D argument is out. Second, a 3D interface in general would be backtracking, since 4D interfaces are being created via human interface devices. Laser interfaces, optical keyboards and projected devices are all in working order, they just need to finalized and developed for personal use.
Trying to simulate 3D interface in an 2D environment is a bit like trying to produce 2D graphics with ASCII graphics, i.e. with a 1D, CLI interface. Even ASCII graphics do work sometimes (and they certainly look entertaining…) – but it’s often better to leave it at a few suitable cases than to try to make the whole CLI interface filled with some useless and distrurbing ASCII graphics, right?
For example, 3D design and modeling in 2D desktop environment is, of course, a very well working idea. (Perhaps also 3D UI could have more useful implemenations in that sort of tasks?) But I’m not sure how it would enhance usability much if your desktop has some 3D gimmicks like rotating program windows in 3D on the screen. It may look very cool, futuristic and geeky first especially if you’re into technology for technology’s sae, but what are the real benefits that could help a normal desktop user to get her tasks better and easier done?
Nielsen doesn’t say (nor I) that 3D interfaces always suck, what he does say is this:
“There are cases where 3D works well like construction, engineering, medicine and so on but most of the things we do with computers are more information orientated.”
4D interfaces are being created via human interface devices
Hmm…, to me 4D interfaces sound a an advertising term only (after all, the real world is only 3D, and computer screens are 2D). Does 4D UI implement traveling in time too.., or does the 4D UI use some superb new cyber-telepathy and read your mind and intentions even before you press a button…?
From a usability standpoint how do you rate open source software compared to proprietary software?
The interviewer as well as Nielsen are both shitheads. How can the licensing affect the usability of a software product, and even if it could, can you dare to judge _all_ FOSS software in one sentence?
The question should have been “how do you rate some well-known open source packages compared to their proprietary equivalents?” or “how do you rate GNOME?” or something like that.
And if the expert thinks he has so much expertise to be able to judge all of FOSS with one word, that’s also quite strange.
Reading on, his longer answer to the questions seems to have some valid points, but if you look at popular projects such as GIMP, GNOME or OpenOffice.org, reality does not look like his description of open source development models. The main reason for this is corporate interest, companies are paying developers to achieve consistency and write documentation. The only big FOSS project which comes to my mind where his arguments may be right is KDE.
There are even open source projects without commercial support where the users about its usability and – yes, features. People seem to like features, just take VirtualDub.
Nielsen has maintained one of the ugliest web sites on the Internet (http://useit.com/) and consequently I could care less about his design advice, as “usable” as it might be…
Think a bit… You mean you couldn’t care less. Remember that.
Think some more and you will hopefully find that your statement is stupid.
You _know_ how your product works, but is it obvious to them?
Yes… but most users don’t know what they want. Or they may have the most very vauge ideas, but they have no ideas whats possible.
That’s where the designer comes in. They know what’s possible AND what the user wants.
“The only big FOSS project which comes to my mind where his arguments may be right is KDE.”
Indeed, KDE is an excellent example of what Nielsen is talking about here… technical brilliance but ignorance when it comes to usability, polish and visual elegance. I am afraid this sort of attitude is firmly tied to the cultures of many such projects.
I sometimes wonder what the goals of the developers are? I find it hard to see that, for example, KDE could compete with commercial products such as MS Windows or OS X (or even with Gnome in the long run), given that the goal is to create a viable alternative for the average computer user.
But then maybe the goal of most of the contributors is just to create someting for themselves and learn something in the process?
Looking Glass will be great when holographic touchscreens are available.
Yeah, Nielsen’s statement about the FOSS usability was not totally fair. I wonder if he knows about the GNOME HIG, for example? He should have focused his statements about FOSS usability better, compare individual projects etc. – though that may be a bit difficult in a quick interview only.
GNOME is a good example of a FOSS project where usability is prioritized very highly (though many users often tend to disagree with the implementation…) However, many things still remain to be done even in GNOME to make it more easier for common, even novice PC users (I don’t mean to say that Windows and Macs are always easier to use).
Also, OpenOffice, though relatvely easy to use, and quite feature rich, could improve in its usability a lot in my opinion. It is not always veryy intuitive to find out features you are looking for. (An example, how to remove a hyperlink from a text string in OOo Web Writer? I still haven’t figured it out – but maybe I’m just too stupid for such fine software…) IMHO, it would be useful if the Open Office project could, for example, adopt GNOME HIG guidelines.
Anyway, both Open Office, GNOME or the GIMP – and even KDE… are actually among the very positive examples of FOSS usability. But consider those many individual projects and FOSS program GUIs (e.g. *nix media player GUIs…) that don’t follow any consistent usability guidelines but some changing and “creative” whims of their developers only. Everybody knows that those kind of individual(istic) free software projects form quite a big bunch of FOSS.
It’s a blessing that we FOSS users and developers have at least the GNOME HIG to improve FOSS usability and consistency. Some small problems that the GNOME HIG may sometimes have is a very small price to pay for all the big benefits it has on FOSS software usability.
“First of all, there ARE 3D displays available to the public today, so the 2D argument is out. ”
There have been stereoscopic displays for about 15 years, but you have to wear special glasses for most of them. I believe there are now some lenticular displays around, but I don’t see them advertised in the PC magazines. In principle, a lenticular display can be 3D, like a hologram, but this needs a very high resolution which no flat screen yet made can reach.
Most of the OS proposals are not for 3D or even stereoscopic (2-imsge) displays, but for simple perspective on a flat screen, as used in many computer games.
Whether a perspective view is any use for file management or launching programs seems doubtful to me. I would prefer a screen display divided into clearly defined and labelled rectangles, each of which gives a selected action.
Newtek’s Lightwave (ironically a program for creating 3D images) is an example of a good computer interface, IMO.
It’s silly to say that because we’re limited to 2D displays that we should limit presentation to two dimensions. Paper has been 2D for a very long time, and that doesn’t stop engineers from doing perspective diagrams.
No, but you don’t represent information on paper, or more accurately, store information on paper in 3 dimensions. It’s physically impossible, and that’s the point he’s trying to get over.
@ Julian (and others):
> The interviewer as well as Nielsen are both
> shitheads. How can the licensing affect the
> usability of a software product…
In so far as most FLOSS software is driven by a team of developers and lacks any UI experts. Unless you count the self-proclaimed.
> …and even if it could, can you dare to judge
> _all_ FOSS software in one sentence?
Looks like you don’t have a problem with judging all _proprietary_ software in one sentence.
Hypocrisy.