Jef Raskin Talks Skins
gwernol writes "Jef Raskin, one of the original Mac design team and a distinguished figure in the world of user interface design has given an interesting
interview over on OS Opinion. He talks about the tradeoff between interface consistency and customizability, and particularly the impact of skinnable applications on usability. Interesting reading, including some harsh words for "guru UI designers" like Steve Jobs..."
Jef Raskin is making one VERY big assumption.
He assumes that 'experts' can come up with one 'grand-unified-ui' which works the best for everyone, ignoring the fact that peoples minds interpret things in many different ways.
He should try working with mentally disabled children for a while, it is a BIG eye-opener as to how different people minds can be in their interpretation and reaction to a given stimulus, and is a little undersood area.
These 'UI' experts who apparently know us well enough to design the 'one true' interface are chasing an impossible dream, IMHO. You only need to look at how many people love/hate XPs default look, or apples aqua, for examples. I personally cannot develop efficiently without multiple desktops to support my many open windows, yet I know other excellent developers who will NOT run more than one app at a time and run it fullscreen.
This is the equivalent of trying to design the perfectly efficient kitchen, it will never happen.
I suspect a lot of the problem is that the 'common' desktop ui's out there don't really skin very well, the underlying system is too limited. X windows is the exception to this, as it only exists as seperable layers, allowing a much fnier control of it's functionality (via KDE and Gnome, for example)
The most consistent UI I've ever used was under OS/2, and IBM did a LOT of development on that, I wish windows would catch up, but it was far from perfect for me, and I bet the majority.
Customisation is required for ANYTHING we interact with in a major and complex way, computers are probably the biggest example of this yet, thankfully their customisability is growing.
"No customizations for you!"
In this interview, Jef Raskin comes off as rather arrogant. He seems absolutely convinced that there is an objective, scientific, Best Way for everything about interfaces.
I'm not convinced. One person might actually work better with white text on a deep blue background, or whatever. I can think of other examples.
With Mr. Raskin it is all-or-nothing: if you work for him, you don't get to customize anything, unless you convince him that you really have a better idea (in which case he switches too, and everyone else who works for him has to switch too.
His supporting arguments didn't impress me much either. A "Preferences" dialog makes an app consume more resources? Not enough to matter, I'd say. That's like saying that putting foam cushions on a car seat makes the car heavier.
The absolute gem of a quote, though, was this one:
Of course, there are no really well-designed interfaces out there good enough to prove the point that you don't need preferences. Any programmers who want to help build one with me, drop me an e-mail.
Maybe he can actually create an interface so amazing, so perfect, so right that no one would ever be able to improve upon it. I won't hold my breath, though.
steveha
lf(1): it's like ls(1) but sorts filenames by extension, tersely
Now I have been UI programming for years but thinking back on it all the command line applications are by far the most productive. Have you ever seen a data entry person using cobol programs on a mainframe? That type of interface is generally for business applications the most productive I have ever seen. Desktops are for people that like to play, if you want to see efficient and consistant interfaces there is no need to look further than the good ole command line.
Got Code?
For me, I would prefer to stretch my "subconscious" processing of alternative, skinned UIs so that I can respond to a greater number of computing/information presentation. I may not be able to autonomically process a given interface as quickly as somebody who is familiar with the standardized UI, but I will be able to adapt to other UIs.
:)
I would say that's a resonable opinion for someone who wants to get a diverse computer use experience, but then again most computer users (of which slashdoters are a small small minority) just want to get in, get things done, and get out.
Have you ever used the trial version of WinZip? The trial dialog that pops up before you can use the product swaps the OK and Cancel buttons intentionally every time it starts up. Even if you use WinZip 10 times a day every day your concious mind comes into play every time you use this application. What's the task at hand? Accepting a dialog. Why should you have to think about this?
The same mental lag happens for novice users working between UIs (or skins)... bringing the concious mind into the picture is not helpful to people who need/want thier entire concious mind available for the task at hand, not accepting a dialog.
Hey, Jef: I use Solaris (and even SunOS 4(!) for some of the legacy systems) at work. I also use Windows NT at work. I have Windows 98 at home for games. I have a Linux box at home for networking. I have an iBook on which I run MacOSX and Yellow Dog Linux.
So, Mr Jef "Allowing users to customize their desktops creates a hodgepodge of interfaces" Raskin, I have a question: Have you considered that my life is already a hodgepodge of crazy and different intefaces, and that skins and themes actually let me make it MORE regular?
(Yeah, I know he's not really reading this. But I needed to vent somehow. If Sun, Apple, MS, and Red Hat will suddenly all agree on a common UI, I'll drop my need for customization.)
"I'd bet my last dollar that web browsers are the source of more wasted time at work than all UI 'issues' put together." I think most pople that frequent /. have easy jobs.
Saying Java is nice because it works on all OS's is like saying that anal sex is nice because it works on all genders.
I really don't understand the point that Raskin's trying to make here. That a UI shouldn't even allow changing colours? That seems pretty draconian. That people might change your settings behind your back, so there shouldn't be any settings? That seems pretty incoherant. That a system shouldn't automatically save settings without some way to undo the change easily? Good advice, but it doesn't seem to have anything to do with the question that was asked.
What is the point of this anecdote, and why is it here?The entire concept of "consistency" in interface design is misguided from the start. The issue of "practicality" is an important one but it's certainly no less important than the issue of human individuality. If everything in life were designed for practicality above all else, and if everybody were forced into using all the same products for the sake of consistency, I'm not sure I'd really want to go on living. The issue of interface consistency is no different than the issue of whether or not we should all be forced to drive Nissan Sentras and paint our bedroom walls off-white. Please, somebody kill me if that happens.
This whole argument also completely ignores the fact that the user interface is increasingly moving off the desktop and onto the net - advanced Flash web sites and even html sites can have their own user interfaces that have nothing to do with the OS or programs you're running and that have their own learning curve to deal with. In fact, for my job, I'd say I deal more with web-based interfaces than I do with my OS interface. Is this guy really going to argue that every single web site, no matter what the content, should have the exact same interface for consistency's sake?
I couldn't really care less if somebody has to spend 10 minutes acclimating themselves to my own preferences on my own computer, and I accept that I will probably have to do the same if I use someone else's (how often *do* people use each other's computers, anyway?). I don't personally see what the big freakin' deal is, especially if skinning allows me and everybody else around me to feel just a little bit more expressive, creative, and downright human in what's increasingly becoming a sanitized and overly regulated world - especially at the office. If I want to put a friggin' Final Fantasy X wallpaper on my office computer, it isn't up to this guy to tell me a plain white background would somehow make me more "productive".
First of all, goto wasn't removed from any of C, C++, or Java. Secondly, who the hell moves around and uses everyone else's machines all the time? That's what we have multiuser OSs for.
Third, I have a problem with this idiotic "logic":
(paraphrased from the interview)
"one time, I set some guy's font color to red and his background to red. He couldn't see anything. Therefore, all interface customization is bad."
That's like saying that because I can paint a fake door onto someone's wall and cause confusion, paint should not be manufactured.
It must be fun to spend a day with Jef.
User interface design is important. But when Raskin says you ought to design the interface first and then design an OS to support it, it's terribly clear that he has been looking at one problem for so long that he sees everything in terms of it.
If I said we ought to design a really, really good steering wheel, and then design a car to support it, everyone (except perhaps Mr. Raskin) would say I was a fscking moron. And they'd be right.
It's not a failure of the OS engineer if the UI sucks, it's a failure of the UI designer. Computers and operating systems are built to provide raw power for people to do things, and the job of the UI designer is to present that power to the user in a way he or she can understand and apply it. It obviously isn't easy, but to hear Mr. Raskin tell it, if he gave you a Lamborghini with a square steering wheel, it's the fault of the guys who designed the engine.
Pure hooey. I'll start taking these UI creampuffs seriously when one of them finds an intuitive and simple way to provide me with a GUI with even one feature as elegant and powerful as the CLI pipe and redirect symbols. Until then, they're just arrogant toy designers.
Proud member of the Weirdo-American community.
What an unfortunate opinion piece. For someone extolling the virtues of a scientific approach to HCI, it's almost criminal to simultaneously engage in fatuous argument. It colours the topic by association. Why is it bad argument?
The other flaws in the argument fall into one of these two categories. For example, that there is only a choice between (1) a single customisable interface across a platform, and (2) other users of a machine being stuck with an unfamiliar interface. As other commenters have noted, this is silly. The clear sensible option is per-user preferances, and ideally ones that migrate with the user.
Lastly he argues that adding interface customizability enlarges applications. This is certainly true. He naturally fails to mention that this is a trade-off against the possibility that the user of an application may be able to work more efficiently as a result, if they make the interface more suited to themselves.
As with all these sorts of things, it's tedious and sometimes difficult to sort out the fallacious arguments from the valid, spot the omitted facts, and distinguish between truth and spin. There are some valid points buried in that rubbish, but the Jef and the interviewer do their readers no favours by using such points as support for an argument that is deceptive and unreasonable.
How about some intellectual honesty? Or is that just too much to ask?
Raskin's ideas are good for companies that use throw-away employees. These people don't stay in one position long enough to maintain their own preferences and become customized AND efficient, so if every interface is identical, then there are less problems learning the UI. Positions that aren't disposable tend to customize much more while retaining efficiency because they have the time, the skill, the intelligence, AND the permission to do so.
Always going forward, 'cause we can't find reverse.
I find it very interesting that only comments critical of Raskin's view have been modded up. I actually tend to agree with him, and I am disappointed to see that the controlling sentiment on /. is so one-sided.