Slashdot Mirror
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..."
skin is good. it is by far my favorite sin.
mmm.... skin
Was it right to be so critical of Mr. Steve Jobs though? Mr. Steve Jobs inventented the Macintosh Apple, and it is still the best O.S. ever. Maybe if you are just as successful, you can be critical. Even so, it's good to not like it my way so much. I know I don't.
I can see paying for this. Indeed.
Windows XP SP2 told me to install third-party software that prevents viruses and protects stability... I chose Ubuntu
Java. It seems to me that everyone is beginning to talk about cross-platform issues now that Microsoft is hitting us with .NET.
But we've been talking about this kind of thing for years, haven't we? Consistency and usability across _all_ platforms and applications?
XP: Hi, I'm Windows XP, and I am innovative and cool because I have a customizable UI and can do skinning, or whatever you call it. Java: Hey, no cutsies! I was here first!
It's all about the game. There is nothing else. http://watchingthewatchers.org
~MadScie
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.
Problem statement: If everyone uses a different skin, then you're lost if you use someone else's computer.
Obvious solution: Select your favorite skin on the other computer, do your work, restore to the original skin before you leave. This could be as quick as typing a number in a box "My favorite skin is #372 optionflags 17".
What the article says: Obvious solution not mentioned. It is assumed that you HAVE to deal with someone else's skin choice. Your subconscious chokes. Bad bad bad. UI expert conclusion: no more skins.
*sheesh*
Good article. The conclusion that uniform, scientific UIs will increase productivity seems pretty strong. The time lost in trying to understand a skinned interface cannot be too large, at least not at the mild complexity of computing we face today. Also, the uniqueness of a skin is what may help reduce repetitiveness that leads to inefficiency. I might argue that learning a variety of modified UIs is a good way to improve one's schema of UI design. Do we want to create a uniform environment that makes users unable to quickly adapt to different situations? Our minds can only handle so much information regarding a given schema, yet with practice, our representation of what fits the definition of a proper UI can be expanded. 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. There is really no argument against the author's claims; better coverage of the underlying issue--how much users want to balance breadth and depth of UI knowledge and processing--would have been really neat. Anybody have some good research links to point to?
this does nothing to moz, but those stupid facist ie twats get fucked in the anus like they should... keep it up my good man.
"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
From the article...
If play is the objective, then anything that gives you the jollies is fine. Just don't impose it on others.
Pretty much says it all, doesn't it? And it just seems like such common sense. If your home PC is primarily used for playing Starcraft, go ahead and plunk a Starcraft theme on it. It would be less appropriate to use a Starcraft theme on, say, the computer your company uses to process accounts receivable.
Really, do we need an "in-depth exploration of the human psychology" to get this common-sense point across? Sure, if you get too used to the tacky theme on your home PC, you run the risk of someday thinking, "Huh.. where the hell is this shortcut by default, anyway?" And, yes, that would technically make you less productive. But, let me tell you, more tragic things have happened. I don't think skins are going to go away because of the danger of temporary moments of confusion.
"Beware he who would deny you access to information, for in his heart he deems himself your master."
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?
Hmmm... lots of random assertions, plenty of highly opinionated jabs at people and companies, but not much actually said about the nature of the problem and no interesting solutions at all. He has, perhaps, one point. The rest is vague claims of a lack of "science" in other people's UI design. If that's the best he can do, I'm certainly not buying one, let alone two books he wrote.
In our group, I strongly encourage everyone to keep their fscking mitts off of my machine. That way nobody gets hurt.
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.)
What a boring, stupid article. It's sole focus rests on productivity, when there is so much to do between sitting and sitting, reading and being stirred with ideas and abhorence. There is litle to be said for art, lest it underpin productivity.
Fuck the bottom line. Rest on your heart.
The man seems rather arrogant to me, insisting that customization will reduce productivity and therefore is a Bad Thing(TM). With this kind of logic we should insist on only right-hand mouse settings since 90% of the people who use the computer are right-handed and would be less productive if they had to change the settings when they sat down at some left-handers computer.
Fact is, people generally customize their computers to make things easier on themselves, and to make the machines more pleasant to work with. Nothing wrong with that. Most machines are used by a single person anyway, and if that person moves on then the next will customize it according to his or her preferences. I can't see how the Ultimate UI will radically improve performance; this 'Ultimate UI' would probably just end up annoying a whole lot of folks who don't like what the 'specialists' think is the key to greater productivity.
Now, if the man was serious about improving productivity in the work place he'd abandon this topic altogether and lobby to ban web browsers from company computers. I'd bet my last dollar that web browsers are the source of more wasted time at work than all UI 'issues' put together.
Max
My god carries a hammer. Your god died nailed to a tree. Any questions?
See, this demonstrates the fundumental problem with "scientists" who view their science as foolproof. The problem with discarding human opinion is that a human will eventually have to use and adjust to whatever you construct. If you have crafted a "perfect" GUI by some magical formula, there is still no way to assure that everyone will a) like it b) fine it "perfect" or c) think it useful at all.
Now, I'm not a big Steve Jobs fan...but let's set that aside for a moment. Sure, it's nice that this guy's a "real UI expert" and Jobs was mearly relying on "guruism"...but, in the end, Jobs liked it as a human, did he not? (At least, that's the way the article put it.) And, given that computers and GUI's are operated by humans, shouldn't that account for something?
Most of the GUI process is dominated by the actual look & feel of the interface. Is there a formula for that? Really? I would be interested to see if his formula included stark and boring design elements, or curved and shiny ones. I would then be interested to watch a wealth of people throw their arms up in disagreement. THAT'S the point. That's why customization is important.
Some people love the default look of WinXP. Some people like the "Sliver" color scheme instead of the "Blue" one. Some people hate the skinned look and revert to the "Classic" interface. The list goes on...
Certainly, a little bit of context switching in certain situations is better than having some scientist dictate what the Ultimate GUI (tm) should be? Generally speaking, if the basic layout and whatnot is basically the same, the human brain is pretty quick at readjusting itself. When I switch skins from "Classic" to some crazy skin I grabbed from the web, it doesn't take me THAT long to adjust myself.
I'll take the risk of losing 20 seconds of productivity to staring at an interface I think is absurdly ugly and can't stand to look at for the better part of my waking hours.
-Jayde
What's a sig?
Good point, there. Some of us just can't find the time for fritterware.
When you see someone with their operating system's default setup, you can conclude one of two things:
- they are too dense to know that they can customize it
- they are actually using their computer to compute
It's a damned shame there are so many more of the former than the latter.--
the strongest word is still the word "free"
The quote:
"...the interface that Eazel tried to design was an undisciplined and uninformed hodgepodge (my apologies to my friends Andy and Bud, but there was a severe lack of interface science in their work). "
I have to agree wholeheartedly. When will we all learn that the prettiest, "differentist" UI is not necessarily the best.
A good example would be comparing Vi with MS Word XP. They both edit text, but I'll take ugly 'ol Vi any day over Clippy and his cohorts.
I have no pants and I must scream
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".
Yeah when you are so full of yourself, it gets hard to think about alternate solutions...
I think the adoption of fully skinnable GUI that is built into the OS is the way to go. When that happens, the OS will be able to allow the customizations to follow the USER instead of staying with the MACHINE -- like the limited way login preferences are currenly used.
That way your Britney Spears themed desktop will appear no matter what computer you log into in your workgroup.
Problem solved. Everybody's happy.
after all my life is just... a soul trapped in a little dust...
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.
... before we resort to Taylorism (the guy who invented productivity measurement and thus can be blamed for filling out ridiculous detailed task-sheets), I'd like to point out that experts and amatures have different mechanisms when interacting with computers. Studies from cognitive psychology shows that we are constantly reordering our environment much like a cook (assuming you're not the nukem couch potato) sequences the tools/bowls in sequence before starting work. We put umbrella stands near doors, grocery lists on fridges, and address books on phones. Basically we impose some sort of spatial-temporal orderordering to cut down the
performance of the task to a series of unreflective actions, triggered by environmental cues. Hence the desire for people (especially those visually oriented) to customise their too frequently sub-optimal desktop (hands up those who use multiple deskviews!). On the other hand, experts have been shown to encapsulate knowledge with a much higher set of rules and create powerful keyboard expressions (think regular expressions rather than GUI). The prefer a deep, consistent and even orthogonal ruleset (multiple modes such as shift=extend, alt=reverse, etc) and woe betide anywho dares take away their favorite editor. The problem is that too many applications have evolved independently which causes cognitive friction (the Mac is probably the best at ensuring conformity) and skins are not the real answer. In fact a recent talk a Linux Conference Australia pointed out the absurbity of putting the full-screen next to the close window button. The problem is that we just don't pay enough attention to useability, going for fweeping creaturism (see FLTK) instead, and window dressing will not solve the problem. Computers are complicated and expecting an internet appliance without some significant simplification (aka labotomisation) is like hoping a high-school kid to understand nuclear physics (see chain reaction go boom).
Productivity would be enhanced by letting a free-market reign by individual choice of development tools, independent (no more benchcrafting!) evaluation of maintenance costs (the biggest part of any software budget) and a better understanding of cognitive limitations (including why marketing people should not be given release control of things like clippy).
LL
As a professional in the field, I feel qualified to say that everything that comes from the Mouth of Raskin should be consumed with a liberal serving of NaCl. The field of HCI moves as fast (some would even say faster) than the technology it works with. Anyone whose major claims to fame are the original Macintosh (~18 years old) and the Canon Cat (~15 years old) is only showing his distance from the bleeding edge.
During the summer of 2001 I was unfortunate enough to attend a Humane Interface workshop led by Mr. Raskin. The day long session served only to show how out of touch he truly has become, acting more as a soapbox for venting about his importance to the HCI world than as a platform for teaching. He frequently contradicted his own book (a copy was provided for each table), strayed from his own agenda which he had liberally changed from the published one, and told un-interesting stories about his major innovations from fifteen years ago. The only worthwhile part of the day was to hear the one-mouse-button justification from the mouth of the man who originally created it. A significant number of attendees (myself included) asked for and were granted a full refund of the CDN$150 fee due to the poor quality of the session.
Dancing lessons for bears! inflectionDesign
I know that for different people, different working methods are better. However, I think people tend to be way too defensive about their preferred style of working. Take programming for instance. At my place of work we had to have a long, drawn out discussion about the syntax used in our programs. The largest sticking point was whether opening braces go on the same line as the method declaration, conditional clause, etcetera, or the next line. We had a half hour discussion on it, with people giving reasons back and forth (e.g., "It makes more sense to have the opening brace aligned with the closing brace", "Having the opening brace on the same line is what Sun recommends."). The truth of the matter is - it doesn't really matter. There may not be One True Way, however I sincerely doubt that there is more than a handful and the majority of those are for people with various learning disabilities. Having skins to put the open/close boxes in different parts of the window, or letting you move where a certain menu is, is most likely simply a waste of programming time that could have been spent making the OS or application more effective. People scream "interface Nazi" reading Raskin's opinionated piece - I think you're a little too insecure to realize that you don't have special needs when it comes to your UI. I will never waste my time trying to accommodate Short Attention Span Theater kids who feel they need to tweak an interface when their tweaks have no functional value.
Here's a better one.
"Dell, Compaq and Hewlett-Packard have made a mountain of money from Microsoft, but consider what the worth of their computers might be if they could not run Windows."
Fuck those stupid assholes. If SSSCA doesn't pass, Linux will have a BIG time.
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.
Jef: I remember one client of mine who boasted about his customizable desktop and how he never had to reboot his software. I set the system font to red and the background to red. You couldn't see a thing. He spent a few minutes trying to find and open the now-invisible menus that would let him change one of the colors.
He had to reboot. His system was good in that it automatically saved the user preferences, so it came up red on red. He had not only to reboot, but to reload the software, losing all his demo data.
Ha ha, Jef! That sure was a good trick you played on him! I'll bet he'll never argue with a "GUI expert" again (or even let him into the room). I'll bet you get lots of loyal clients with lessons like that!
I remember one client of mine who boasted that his computer was really fast and reliable. I chucked the whole lot out his fourth story window: cost him $4K to get it replaced. Sure was fast on the way down, but I guess the reliability wasn't quite what he expected when it hit bottom.
Another time, a friend was boasting about how customizable his Linux box was, so I showed him how lame a customizable OS is by removing his kernel and most of his utilities...but that's a story for another day.
Rock on, Jef! And remember, everyone: if you let users set their desktop wallpaper, it will make them spectacularly less productive, and at that point the enemies of freedom have already won. It's user choice vs. freedom, folks, and heaven help us all if you aren't willing to sacrifice your petty preferences in defense of our liberty!
Anyway, that's the first problem with any kind of interface user preference: How many of us really know what works better? Most users and most programmers have at best dim and often incorrect ideas in this regard. I know this from lots of experience.
Either Mr. Raskin is speaking about something different than what I understand, or he is missing a fundamental issue: one size does NOT fit all.
My understanding of "skinning" and "themes" is they are the means by which to modify the UI colors, fonts, background images, etc. to my liking. And certainly some choices are generally more effective -- more usable -- than others. But these decisions are made based on aggregate data; they are geared towards the "standard user." But what if you're not a standard user? Are you forced to use an interface that is not optimal for you? If Raskin has his way, then yes.
In contrast, a most staid of industries, the automobile companies, have been moving away from such an attitude over the years. The seat size, pedal placement, steering-wheel tilt angle, etc. (car UI) are all based on a standard driver; an average human. But recently, they've been providing drivers features to modify these parameters: tilt steering wheel, 8-way adjustable power seats with lumbar support, pedal positions (we can "theme" our car). This is because one size does not fit all. And if I can tweak the interface to best fit myself, my physical function, and my tastes (for comfort), then I will better enjoy the device and better use it.
His claim that changing the desktop will hinder productivity is largely like claiming that our ability to drive different cars is hurt by their various colors. I suppose Raskin and Henry Ford would get along famously, but I want to choose the color of my car.
And somehow, I don't think my mom will have problems driving my green car, even though hers is "skinned" with a white theme.
ShoutingMan.com
The simplest solution is this: allow all preferences to be stored in a single file. Allow that file to be posted on the web. And then allow every OS to have a box somewhere, which itself is not customizable (but which is easily ignored), where you can type the URL of your preferences, and thus put them into effect.
I can create my custom settings at home, put them on my personal webpage, go to work, type the URL into the box, and wham! Just like home.
Somebody else wants to use my computer, hey, they just type their URL into the box, and it looks just like ''they'' are used to.
Everybody else can do this too, and of course the "defaults" would have a URL, too, so if you didn't want to use your own you could just enter that URL and be back to normal.
Then everybody gets what they want, unless what they want is the ability to impose their user interface decisions on other people.
-- Sunlighter
Sunlit World Scheme. Weird and different.
... since they are utterly useless...
Who cares what is actually the most efficient way? Humans dont want to be efficient, we want to enjoy life in all its glory... and yes, that includes skinning and having 20 aplications with 15 different interfaces, driving our fuel inefficient Mustang GT's, eating burgers and getting drunk every friday night.
Humans are not robots, and we should not be treated so...there even this crazy idea new economy that happy workers are productive workers... dont know about you, but being stuck in the cubicle with black & white screen for 10 hours a day does not make me particulary happy
just my 2c
I started out on MS DOS 2.11 and stayed with MS until Mandrake 6 and while my history of UI interaction on PCs doesn't lend me any experties it does lend historical perspective. For the sake of discussion I see VisiCal as the prototypical PC app followed by WordPerfect. VisiCal and WordPerfect are both fundamental data input apps permitting manipulation of the data in terms of formatting or the performing of operations on the data. From this foundation of data processing the windows paradigm has be devised. I'm suggesting a new paradigm is needed that aids creative thinking. Maintaining the data processing paradigm may be supportive for clerical types but the current desk top flies in the face of the recurrent talk of AI and PCs with SuperComputer power fundamentally resturcturing the way we interact with and direct our environments. Polishing the desk top may be a way to parallel the MS Windows market but investing in a dynamic UI empowering creative thought may be the way to ultimately steal Bill Gates' fire.
heuristic algorithm seeks stochastic relationship
I'm appalled that nobody has taken issue with Raskin's stab at Linux yet.
Anyway:
You have to start with a UI and then build a system that supports it. Linux failed in that regard.
He seems to be confusing a UI with a GUI. A command line is a user interface in my humble opinion. You can argue whether it's good or bad, but it is a UI.
Apparently what he means is that Linux hasn't made Windows obsolete on the desktop yet, because of the fact that Linus did not start off with a fully designed GUI. Duh. Just for the sake of argument, let's say I want to use Linux as a server OS. Why would I need a GUI? Server machines often have monochrome screens and no sound, and for a good reason: color and sound only get in the way.
But, I do agree with a lot of the points he makes in the interview. I think inconsistency is annoying, and therefore counterproductive. Ever used one of these eval versions of WinZip where the buttons of the nagscreen on startup switched all the time? That was to make the nagscreen annoying, and boy did it work! The dynamic pull-down menu's on the 2000 versions of MS Office tools are another example: probably well-intended, but generally annoying, therefore bad.
Also, I find that I like a boring monochrome background for my desktop best. Same for websites: the best designs do not use a picture as a background.
This does not mean that tools should not be customizable. Customization is not always a matter of preference. When you learn a tool you often have different UI requirements than when you're an expert. Might like to add shortcuts for often-used commands. Another example: I may want to customize emacs so that the Java syntax highlighting is the same as in JBuilder. This actually adds to the consistency.
My main objection to Raskin's (indeed somewhat arrogant) position is that he seems to think he knows what's best for the rest of the world. My opinion is that users should be able to customize tools to their liking. If I like a different skin better, then that should reduce my annoyance, hence increase my productivity.
Having said all this, I'm currently reading Raskin's "The Humane Interface". I'm about halfway through, and I must say I like it very much. He describes some good and bad UI concepts, and explains clearly why they are good or bad. There's some human psychology concepts as well. I've gained quite some understanding from the book already (probably because I don't know the first thing about UI design), so I still recommend it, despite the somewhat dubious interview.
MSN 8: Now Microsoft even has bugs in their ad campaigns.
Wouldn't the ideal desktop environment simply carry your own settings over to the new machine when you logged in? You should never have to be exposed to someone else's customized environment; you could just carry your own environment around with you...
(maybe Jef spent too much time around single-user Apple machines and not enough time on UNIX, hehe =)
Think about what would happen if you customized the position of pedals in your car. What would happen when you let your friend drive this car? What if you are used to your modified pedals and then rent another car? Could you react fast enough? This is _exactly_ the point the author is making in his article.
Did you know that all icons displayed in your car are standardized?
The article seems to focus on the UI as being entirely a functional issue, and trying to do away with any type of form or artistic side. So maybe studies do show that most people are a little more productive on a UI with a specific background and certain size buttons, etc. Does that mean that people actually want to work on this type of a desktop. We could make everyone work in identical shaped cubicals in an office with a certain colour on the walls, but do people actually want to work in an environment like that. We don't design houses entirely for function or cars or practically anything that we use.
I think the real issue has to do with ease of use. If a UI is so intuitive that things just make sense as you do them, it doesn't really matter if you give your windows a strange border or have a picture of Fiji on your desktop. I don't believe that allowing users to change skins could really put a big damper in the ease of use of the program or the productivity of the user.
Enjoy the reading and e-mail me for more references/discussion if you are interested!
adam
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?
As the parent post says, he has become quite arrogant... and I daresay a bit jealous of Steve Jobs.
Please understand, I've been what is called a "Mac Fanatic" since I first saw one in 1984. I now make my living on NT, AIX, Linux, etc. but still prefer to use Mac OS X at home.
At least once I day a cuss (under my breath, or, often loudly) NT 4 because it is HORRID interface design. And this is AFTER I moved the task bar to the top so the Start "drop-down" menu doesn't "drop-up" (with "Shutdown" being the first thing you encounter on the Start "drop-up" menu!), created "shortcuts" (that don't work completely or intuitively) to all my drives on the desktop, etc. to mimic the interface Jef created in, what, 1980? 1982? (with improvements since, of course)
But he needs to lighten up now. He is, IMHO, the Father of the GUI. (I'm not trolling and don't want to rehash old Xerox PARC arguments. I'm talking about mainstream markets.) But he's been really hard on OS X's Aqua interface and I like it. There are things in Aqua that I, personally, think are a step backward. But I can modify those THROUGH PREFERENCE SETTINGS (or other easy modifications) to be more like what I'm used to, and overall, I think OS X is the coolest OS in the history of personal computing. ("You got your point-and-click ease in my powerful OS!" "You got your powerful OS in my point-and-click ease!") Two great tastes that taste great together.
My message to Jef is: Rest on your laurels or continue to innovate. Either way I can respect you. But don't become the guy that makes his living by criticizing what others are doing. That's John Dvorak's job.
Most user interface customization offers that level of customizability. Gnome and KDE seem to both be pretty good at letting you change the things that would correspond to the mirrors and seats on your car -- you can make the various widgets look a little different to suit your personal taste, and I think that's fine.
I should also note that whenever I install a new system I always spend about 4 minutes configuring gnome the way I like it and about half an hour getting EMACS to the point I'm used to it being. Fourty fiddly E-Lisp thingies...
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
Jakob Nielsen would say the spelling hurts the usability of Jef's name because it goes against our conditioning. Everywhere he goes, he must bear the burden of correcting people who misspell his name.
I'd say Jef is either quite angry with his parents' creative streak or there's something inconsistent about his behavior.
I recently stumbled across a GUI book on-line:0 000000057.html
http://www.joelonsoftware.com/uibook/chapters/fog
I found Raskin's interview to be elitist... So much so that he appears to have no company on his
lofty plane... No current OS has anything admirable... Because you MUST start with the
UI first...
Before the chips are selected,
before the nasty hardware access layer is defined
and the "Operating System" fleshed out
we are not operating...
we are interfacing...
don't you see...
don't cut... splice!
don't sever... graft!
Before Dr. Fronkenstein's
creation bolts up-right and "It's alive!"
No! stop this madness... storm the Dr's Castle...
Where's the UI? Where's your UI spec!?
So, First the UI... (Is it in Jeff's book?)
then... a massive round of VC
funding to build a system to enable the UI...
How much money will it take? How much have you
got? Spend, spend, spend because this UI is:
p-e-r-f-e-c-t-i-o-n
This would be a system you could dance with...
The Ginger to your Fred... moving effortlessly...
backwards in high-heels... tap-4-tap...
[Oh yes, don't forget to test it and change anything that delivers less than the highest potential human productivity.]
Simple really,...
it's amazing to see so many GUI's just slapped
willy-nilly over some hugh wad of software
that crawled out of the swamp of technical
feasibility... without an intended User
Interface... madness...
How can you be a UI expert when you have no system of judging a UI. Any good subjective analysis has have to various gradients of good and bad weights and there must be examples on all ends. You can't call yourself a UI expert just by saying that every UI out there sucks. That's like tryign to become a good movie critic by saying every movie stinks--oh wait, there's a great number of people like that. If you're only justification is that you have to write a book in order to explain UI, then by god, you've definitely missed the User Interface boat as no user would want to interface with something that pointed at pie-in-the-sky nothings. Ponte tu dinero donde tu boca esta. I'd bet on Steve Jobs with his blockbuster Apple sales than on some supposed pundit who spoke with Steve once.
Philosophistry
First of all, is it the interface of MacOS 9.x and older that he is praising? The most hopeless interface that I have been stuck with for any period of time. It breaks a host of interface rules, and clings like fanatics to others. I have no harder reaching my taskbar on the top of my screen (where it belongs very much thank you) than I have reaching my menus in IE.
I agree with him in many ways, but he is way too arrogant, and doesn't have nearly enough behind his so called science. A uniform way of using the OS and applications is very important, especially for the non nerds. If you teach a class, and the computer the person then sits behind at his/her new job looks completely different, everything will be lost.
But to use novice computer users as some form of basis for how an interface should work, it's just plain dumb. I belive one needs to put into science the art of making interfaces, but before that we definitly need to get one thing straight, there are different people using computers, and most of them will spend quite some time infront of the glow of the screen, those should be just as important as the novice. After all, how would cars look if all of them were built for peoples first try? I am sure they wouldn't come to the Viper that way...
Mr. Raskin needs to use a multiuser machine some time. On our machines *I* get *my* setup and skins and nobody else has any buisness logging in as me, or using a computer where I am logged in. And everyone else gets their own customized setup on their accounts.
An intuitive interface that solves that problem is easy to design. A big password box, and a big 'permission denied' and he wont be confused when he tries to use someone elses account.
Raskin makes a good point that it's a pain in the butt when you switch to someone else's computer, and they have skins set up in some weird friggin' way that takes time to adjust to.
Various posters here have made the point that it's human nature to want to customize things. Which is true, but this reply is an inadequate response to Raskin's argument that time is wasted whenever you jump to a computer with a foreign skin.
So here's my humble little proposal, and, to my knowledge an original idea. This problem could largely be solved simply by using the net. Why not have a website or registry somewhere that stores YOUR OWN personalized skin? That way, when I go onto your computer and am horified by your choice of screen colors, button preferences, etc...all's I need to do is visit UIpreferences.com and type "Schlemphfer" and my password. Voila. Now your machine has preferences identical to my personal system.
When I finish using your machine, I'd then hit a button which would restore the machine to the owner's preferences.
Seems to me that prior to the net, when we all used non-networked computers, the case for Raskin's "adhere to One True Way" argument was far stronger.
I'm generally "Interesting," "Insightful," and even "Funny" here. What the hell happens to me at parties?
Perhaps a way to make your system easier to use is not to make the UI more consistent, but rather never to let Jef Raskin near the keyboard!
BH
Fools! They laughed at me at the Sorbonne...!
I got to the part where this man said something to the effect that professional UI designers can do the job better than amateur users and got the willies. That sounded just like a spate of management consultants I know.
With management consultants, if they can get you to believe they're smarter than you and can help you out, they got themselves a job. If you work out for yourself that 'Hey, this is my operation, I'm sure I know it better than a guy who cracks the management textbooks a couple of times a week' then these guys are out of a job.
This is quite the same; if all of a sudden we discovered that users actually knew more about what they want their individual interfaces to look like than the professional designers, guess who would be the first person in line at the employment office? That's right. UI designers such as Mr. Raskin.
This post is not a judgment of whether Mr. Raskin knows better than Joe Sixpack; it is mrerly to point out that this man is not qualified to state his opinion so long as he can't remain objective.
He's fighting for his bread and butter, after all.
Blearf. Blearf, I say.
While everyone else is griping about how arrogant Jef sounds, I'd like to point out that this is very much what the KDE and Gnome projects are doing. Granted, they're not succeeding as completely as Apple is on their own system, but then it's a totall different system of development. That said, you can get a fairly consistent UI using either one of these environments. Stick with Konqueror, Kmail, and Koffice for a consistent KDE interface. Likewise, you can stick with Galeon, Evolution, and Gnome Office for a consistent Gnome interface. Granted, you may not have the best apps available in all cases, but you'll get your consistent UI.
What I think is really interesting about this is the fact that most of us use apps from both environments, mixing and matching per our preferences. I personally use both Konqui and Mozilla, Kmail, xmms, and gaim combined with kyahoo, as well as StarOffice 6 in a KDE desktop, but that's just me.
The fact is, that a consistent UI doesn't wind up being that important when it comes to application functionality. People learn to use their apps. This isn't just the case in the Linux world, people still use the weird Kai interfaces for instance, or the quirky winamp UI. Hell, even the big boys break their own rules with their media players, and no one really cares that much. The fact is, if the program has a good amount of functionality, the user will learn the UI (witness the shitty UI in Napster) to access what the app offers.
But then again... I am posting this fairly drunk at a very weird hour, which is always a bad idea, so take it as you will.
"I may not have morals, but I have standards."
I just *hated* the way he used arguments like "others having to use your computer" or that "in my group I encorage every one to leave things at default".
Isn't this exactly what multi-user environments are about?
It is just plain arrogant to have everyone do things exactly the same way. People have different tastes. Perhaps one person needs larger fonts because they have poor eyesight, but the larger fonts just annoy the hell out of everyone else?
The "right way" is just to set up a multiuser enironment and a networked file-system, so that all machines are equal, but all users can do what they want. Almost all current operating systems are multi-user in some way or another.
Testing the quality of GUIs does not require any magic.
You can f.e. create two alternative GUIs, get some testers that are representative of your target group and give em typical tasks (test cases). You then record how long it takes these two groups to finish the tasks without help.
There exist verified tables of typical times and simple formulae that let you estimate times required for basic actions. (Like switch between mouse and keyboard or 'how long does it take to click a button of this size that is so far away'.) Tools making use of this (like GOMS) can help you quantify interfaces even before testing.
Large parts of GUI design are quantifiable.
The real productivity losses are when the paradigm shifts drastically. I'll be perfectly honest, the first time I sat down at a box running afterstep I didn't know what to do... And I'm sure almost everybody who sat down at my machine (I had the kickass gaming machine the hall in the dorm I lived on) running lightstep was completely lost. Some things about OS9 are annoying to me (namely the lack of a maximise button, and the fact that the mouse sensivitivy seems so low by default), although I haven't really gotten a chance to use OSX yet. But once I'm past that learning curve (may be 30 seconds, may be 5 minutes) as long as everything makes sense, it really doesn't matter after the fact.
Actually I just thought of something. If you're concerned about productivity, instead of making everything look exactly the same, make it behave exactly the same. I'm thinking specifically of shortcut keys. What is copy on a Mac? Apple+C or something? I dunno. Then again most [l]users don't use shortcut keys anyway.
And what's this little quote?
Wow, that makes almost as much sense as, well, something nonsensical. Sure, I can make it really annoying for somebody to use their computers, but that doesn't mean the ability to do that means I automatically will.
Of course people have been telling that to the companies that put out copy protected CDs for quite a while now, but they haven't been listening either... Hmm...
I need to lay off the crack, my inability to form a coherent thought process is impeding my karma whoring!
a load of crap. Thanks for wasting my time. I figured there would be some inciteful argument or something, but it was just another standardization asshole preaching the word. Maybe we should all use Windows(tm0, too.
Damnit, Jim, I'm an anarchist, not a F@#$!^& doctor!
http://www.osopinion.com/perl/printer/16564/
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.
In the real world some of us like to customize our vehicular transportation modules as much as we like custom UI - and in fact, I'm probably a hundred times faster with my customized setup than with the default: there is no illusion of speed here, as all gains have been substantiated by years of building motor memory. Make no mistake about it, there is much value in changing default GUI to match one's personal preferences - unless you trust the dingos who build wishy-washy default configurations for you more than your years of ingrained habits. I'll concede that some people's habits are poor, but then again, mine are extremely good. I'm somewhat of a power user, with shortcut knowledge and I resent anybody's interference with my extremely efficient system.
--you have been trolled--
Every single article that I read by Mr. Raskin has vilified and bitchified Steve Jobs and his user interface. I am starting to wonder if there isn't another reason why he seems to bitch so much. Is this more of a personal assault to Jobs & company or does he truly think that skins and OS X, in particular, are just poop? It makes me wonder if he doth ccomplain too much?
He will continue to complain until he starts to become annoying and then folks will start to dismiss him. It kind of happened to Jakob Nielsen. He complained so much about new web technologies that I started to see people ignore him. These are both super smart people, but I haven't seen a whole lot of cutting edge stuff from them. I am starting to become tired of the hooey from both of them.
Just a thought.
Problem statement: If everyone uses a different skin, then you're lost if you use someone else's computer. Obvious solution: Have multiple accounts on same machine. Or, alternatively, tell idiots to go sit on a flag-pole, or buy own damned machine. Assumption: anybody I let near my machine will be intelligent enough to figure it out. I don't care about their immediate productivity. Same goes for me when I need to use somebody else's box. What the article says: Obvious solution not mentioned. It is assumed that you Raskin is sitting on a flag-pole. Your subconscious chokes. Bad bad bad. UI expert conclusion: erase image from mind. What the article should have pointed out: retards have no place playing with my computer. Period. That said, there's no logical reason left why I shouldn't tweak my GUI as I see fit. Take away my freedom to meddle in my own affairs, and I'll buy beer instead of your product.
--you have been trolled--
I think that if you look carefully, you'll find that free choice makes a muddle out of ANYthing.. just look at those Brits: somebody let 'em drive on the wrong side of the road, and now they're all fucked in the head. Then look at the Yanks: somebody let them carry guns and start wars, and their beer brewing skills went all to hell. I think you'll agree that from a logical perspective that this precocious tenet of basic humanity has no place in our world.
--you have been trolled--
Ash UI durbatuluk, ash UI gimbatul, ash UI thrakatuluk, agh burzum-ishi krimpatul.
--
The Cap is nigh. Time to get a fresh new account.
Another point to raise is that main-stream computers for non-techies is somewhat of a recent phenomenon. It takes time for people to learn. Example? At the moment I'm learning how to drive, and a car is one of the simplest and most staightforward interfaces I could imagine. It makes sense and yet isn't obfusciated. Wheel turns car, pedal 1 brakes, pedal 2 accelerates. But it's still not immediately easy. It takes practice. A computer on the other hand, is much more complex in its uses than a car. I have met people who have trouble wrapping their mind around the concept of a directory heirachy or difficulty understanding that a
But at the same time, a little bit of commonsense should be used. It's probably a bit foolish to make the default keyboard shortcut of Ctrl-C for one app to be "copy" and the other "reboot without confirmation". Make it customizable, for sure, but the people who are actually making the interface should try and consider their target audience. Making a kiddies program to teach maths ? Big buttons with colorful pictures and cutesy pics probably will be popular. Making a CAD program? You probably don't need to care then about whether your users know how to use the second button or whether they can handle multiple views.
But regardless of what you build or design, what we should be encouraging is choice, and understand of different approaches to problems. KDE 2 is in my mind probably the best Linux GUI I can think of. It's clean, well structured, fairly familiar to most people, and consistent. However, even that, I don't use it. I use enlightenment, because I *personally* really don't like "start-button" based interfaces. I like eye-candy, but if it's not functional, then I won't use it.
One last example I'd like to give is that my room is currently a mess. There are papers all over the place, ranging from exams, to newspaper articles to printouts from the onion etc. Every now and then, when my mum decides that she's sick of it, she'll clean it up, and then I won't be able to find anything, and my entire "system" will be destroyed. Something similar are the people who have "clean" desktops with as few as icons on the desktop as possible, to those who have heaps all over the place. Is either of these more efficient? The clean desktop users, will often have their "core" apps on the desktops and maybe keyboard-shortcuts to the others. The messy desktop people will probably get used to just double-clicking or whatever. I doubt there's any real difference in terms of speed/efficiency. Most of the time users are spending time typing, or editing something, not fiddeling around with menu options and the like. You type in your word processor, or rotate that grid for your 3d-comp-generated-movie, but most of the time your working on small goals at a time which you get used to and stop thinking about.
And finally the only thing I would like to see change is the open/save dialog business. Personally I'm tired of having to find the directory to save/open my file. I really would like if when I clicked "open", rather than having to search amongst my gigs of music/movies/pics (that have forced me to use a structure that gets multi-branched...e.g. downloads/ and then downloads/src and then downloads/src/apps, downloads/src/libs etc.) that maybe the computer could figure it out for itself. "Hey its an mp3", shove it here. I'm thinking MIME types + index files, I realize this is probably hard to do, but it's just something I wish could happen.
Font sizes. That's all I have to say. Well, I'll go on. Large fonts mostly drive me batty. But most co-workers I've worked with seem to like larger fonts. They also frequently like fonts I wouldn't be caught dead with. Would Jef make me use 12 pt just because everyone else is? Nuts.
Here's the thought: if you're working in an environment which necessitates the occasional working-on-another's-computer, instead of worrying about creating the perfect customization, why not create a module of the os or an app which allows even MORE powerful customization, a la storing customization profiles on a separate server? You have to work on Alice's box (sorry Alice), you sit down and call up your customizations. If you're polite you call up Alice's again when your finished. If you're not then Alice has to do it for herself. Pretty simple. And avoids all this debate about efficient vs non-efficient.
Let's see, if common UI's are best in order to avoid the mental disconnect that comes from switching between different systems, then since Windows is by far the dominant desktop OS (and hence UI), then it would be better if all other OSes (Linux, MacOS, etc) opted to mimic the Windows UI. That would certainly remove the mental gear-shifting and uncertainty that occurs when switching between different systems.
He also seems to be under the delusion that there IS a composite average person out there, and that we are all just like that fictional entity. No leftys, no color blind, no differences in personal taste.
What a Maroon.
It's easy to get used to something and forget about its minor annoyances. There might be something better around the corner which you don't look for, because you are already using the current UI subconsciously. I might even say that it's a sign of 'good life' to try and do everything consciously: question what you are doing, why are you doing it. Ask yourself honestly if your current UI has shortcomings.
If you've done Monte Carlo analysis you may know that sometimes your algorithm gets stuck in a local minimum. You have to introduce some randomness if you want to find the global minimum. Even if you're comfortable with your current UI it might be worth trying out others, you could find something even better.
This approach works in other field as well, but certainly not everywhere. The problem with the mathematical picture is that you're never sure if you've found the global minimum. So for example, if your girlfriend is not perfect in every respect, it's not always wise to dump her in search for something better... fortunately window managers don't mind :-)
--
The Cap is nigh. Time to get a fresh new account.
Consistency is the king, but personalization is the kingdom. Without the kingdom, the king is nothing.
Being a user, I use to work in a heavily personalized environment: skin, background picts, icons, directory structures, window positions, namely everything customizable. If I'll have to use another machine, I want consistency, and this means not only the way to operate application programs, but the way to deal with *my* environment, not any other people's, not the defaults. "I like it" is the strongest case.
Consistency and Personalization must stay together. There is no reason we have to abandon either one of them just to make other people's live easier.
Here lies the problem of today's desktop PCs. No matter how portable the files and the programs are, the personalization settings (system, applications, interfaces, states, etc.) never migrate out of the individual's system, not to mention the ability to resume the environment on another system.
There are implementations done by then Digital Equipment, IBM and Sun that users could maintain their sessions and states across different terminals. Yes, that was the thin client model, still the concept of being able to save any individual user's environment and states and restore them later on another different machine, in my opinion, is the perfect solution. Sadly, it never make it to the desktops.
J. Tachyon Bates
tachyon_at_m1_dot_ffn_dot_ne.jp
No the user soon gets bored, and then productivity dramatically decreases. If you give the user of the machine the ability to easily change the interface to suite their particuler needs the interface submits to the user instead of vice versa.
What makes you think the developer knows what I need? This article takes the approach that everyone that uses a computer is stupid and needs to have things presented to them in the simplest fashion. If I change my desktop, I won't be able to adapt to a new computer? That is absolute crap!
I agree skins get a lot more attention than they deserve, but it's a bit of an exaggeration to say they are a productivity killer
Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
..."Amen, Brother!" It doesn't seem like they do; he seems to just piss everybody off. I think I've sat in chairs designed by this guy's brother. "Research shows that this seat is perfectly suited to supporting the human ass. If you think it's giving you a back ache, you must not know what your body really needs."
To improve consistency and ensure that all employees are able to share shoes, effective immediately all employees will be required to wear men's size nine wingtips to work. Any deviation will be met with disciplinary action up to and including termination.
--Catbert, HR Director
I took a class in usability back in the Fall, under Rex Hartson, another well-known expert in the field. The points brought up in the article make so much sense to me now. In the course students were teamed randomly, and due to the class mix, about half were human factors types and the other half were CS majors. Through the course of the semester we worked on a project which was the design of an interface for time and project-keeping purposes, and the one thing that came up so often was the human factors engineers' concerns being paid no more than lip service by the CS majors. This was not always the case, but the programmers simply could not be shaken from their desire to create whatever worked, with whatever code, and 'that's the way it should be' without understanding that their job is to code to the needs of the user, which are demonstrated only through a concerted effort to elucidate such needs, through all available techniques if need be (interviews, ethnographic analyses, etc.)--it is only then can we even attempt to approach the 'ubiquitous GUI.' All too often, and certainly not limited to interfaces, user requirements and expectations are not incorporated into the design, or, if they are, it is usually in a 'mopping up' fashion, which will always cost more.
By Jeff's indication we should all be living in houses with the same furniture and decor. I think I will call it ISO 1234567, 'the standardisation of home decor' and then apply the DMCA to prevent people from try to change the implemented decor standard.
Jumpstart the tartan drive.
So, what happens to the "ultimate, perfect interface handed down (in unchangalbe form) from we the enlightened" when it comes out of the lab?
The first thing that happens is that icons and buttons are added for "try AOL", "Real", etc. etc. etc. "No no no. We can't have that button there, that's where we are putting the 'sign up for Passport' button."
The second is that the interface is re-engineered to focus the eye upon the marketing of partner products and the interference with "efficient" use of compeditors and/or discouraged products, formats, etc. (.mp3 on XP, anyone?)
If allowing customization is the only way to get rid of all that, then I'd rather be "less productive" on my co-worker's computer 5-10% of the time than consistently annoyed and harassed (by my own computer) 100% of the time.
If Raskin produces the "one true interface" (OTI), let him cast it adrift and see if becomes dominant in a world of choice. Make skins for not only the OSs, but also the major skinnable apps. Encourage others to adopt his OTI by providing help, a free specification, and free license to use it. The OTI should gather a community of developers who want to bring the OTI to everyone.
If not, well, I guess it wasn't that great after all.
Levity aside, surely on a Linux system you do the simplest thing - log out of that user's system and log in again as yourself. Ta-da! Instant default interface. Or do as we do at work - share home directories, so that wherever you log in, you always get your very own preferences.
And either wait several minutes for your preferences and images to download over a 56K modem, or pay upwards of $200,000 in some areas to move to an area where broadband is available.
Will I retire or break 10K?
I suppose if Nullsoft had done their job and hired interface diety Jef Raskin, people wouldn't feel compelled to skin winamp? That seems to me like saying if Microsoft had done their job properly people wouldn't feel the need to change their desktop background image.
Changing that image to a picture of, say, a landscape or sunset probably makes item selection .0001% less efficient, but surprise surprise, people do it anyhow. And I don't think they'd appreciate being prevented from doing it because Jef Raskin has decided that this particular shade of puce makes them two tenths of a percent more productive.
You know, this has always been one of the reasons that I don't like the Mac approach. Basically they say:
*we* will make all the choices ahead of time and they will be right. you just sit back and accept what's handed down from on high.
I've never liked this approach for the arrogance of it.
Sigh. My id isn't prime. 2 2 2 2 2 3 5 313
Ever notice when you're sitting with a bunch of other people in a group, say at a lunch table in a cafeteria, and someone comes along to sit down with the group - they'll move the chair trivially before sitting down? Maybe an inch to the left or something? Ever notice yourself doing that and wondering why?
It's human nature - the subconsious desire to at least perform some token act of control to demonstrate one's dominance, even if it's trivial in nature. It's just the way humans are. UI Nazis who come up with obtuse rationalizations to justify things like this article are just subconciously trying to enforce their own dominance on the situation, like the priest who chooses those non-movable pews for his church.
Next thing you know, they won't even permit us to even move or resize windows. (like all the various apps that launch and run "full screen" - - - iMovie anyone?).
These are my friends, See how they glisten. See this one shine, how he smiles in the light.
is cli the best way to browse the web? depends. lynx is good for news sites, not so good for porn!
Have you ever read an erotic novel? Or you just point-and-drool at the pictures?
is cli the best way to edit graphics? ummm.. let me know when the cli version of photoshop comes out and ill let you know.
Yes, drawing directly on a layer really needs a coordinate input device such as a mouse or tablet, but imagine the scriptability that could arise from a command-line image manipulator:
Here's a clone of Photoshop Elements with a command-line.
Will I retire or break 10K?
Customize all you like, but the interface should be smart enough to recognize that certain cases is a "no no".
So how should it know that red on a green background is not a crash for a (color-blind) person's eyes? And how should it know that a 9-point default system font isn't a crash for somebody with vision problems? Heck, how does it handle the case of a blind user?
Will I retire or break 10K?
It's clear that most of you have never read his book. In fact, it's clear that most of you have never studied UI design at all.
For those of you on multiple systems that try and achieve consistency between all of them, despite different UI paradigms, you're on the right track.
For those of you that say that Jef Raskin doesn't seem to know that you need different UIs for different applications, he does.
And for those of you that think that a unified UI would be somehow heretical, you're wrong. One of the huge draws of Windows today is everyone knows how to use it. Never mind that it's complete crap, filled with design snafus EVERYWHERE. It's consistent from machine to machine, and that's appreciated. They've done actual *scientific* tests where people have worked with a customized (by them) interface, and a standard interface. Very often, even though people FEEL that they're working faster with their personalized interface, they aren't.
Most of you commenting on this story are working off a gut reaction, and are just mumbling the same things over and over again. "*I* like it, so it must be better." You're probably wrong, and you're probably provably wrong.
There're a lot of elements to interface consistency, too many to explain in this comment, and too many to explain in that dinky little article, so of course Jef comes off as arrogant. But there's a reason that he's widely excepted as a master of Interfaces. He's studied them, and he's done it as a scientist.
Go out, read Jef's book. You'll think him a lot less arrogant, and you'll certainly be able to see all the incorrect interface designs around you. If nothing else, after you read the book, you've got a solid basis for criticizing the Windows UI, other than saying 'it sucks'.
Raskin's major peeve regarding skinnable UIs is that productivity is harmed when a person is forced to use an interface even slightly different than what they are used to. This I can believe, as even small changes can be distracting at times. However, I have to disagree with this claim that the only solution to this is to get *cough* someone *cough* to design the ultimate-God's own-Uber-UI and mandate it for all systems. I, apparently along with a lot of people posting here, doubt that such an interface even exists.
... well, that's life when the user is empowered.
What he should really be arguing for, then, is portable user-associated UI skins; if any computer you sit down at automatically calls up your preferred skin and thus reconfigures its appearance to match your expectations, the productivity loss to distraction disappears. The means to do this aleady exists to an extent, via centrally exported home directories with stored preference files. The pieces still missing are the usual Linux-UI gripes: not all apps listen to global preferences (e.g. non-Gnome apps don't care what the Gnome/GTK theme is), and some apps actively use their own separate theming system (e.g. Mozilla). This is why we heard a few weeks ago that RMS wants Gnome and KDE to work on making their skinning systems interoperable. Of course, even doing that wouldn't help with the army of older apps that use Tck/Tk, Motif, etc., as their interface layer.
Some distros are working on this kind of thing; Redmond Linux (sorry, Lycoris/LX) comes to mind. All they really do is decide on a desktop environment, and then choose only apps that work well with that environment, including only those that listen to the global skinning system.
Anyway, Raskin's bitching about the availability of skinnable UIs isn't going to get him anywhere, since all the interesting Linux interfaces are free software. Suppose for a nanosecond he convinced Gnome, KDE, etc, to remove themability in favor of his pet UberUI. Fine, five minutes later any developers who disagreed have posted their own versions of the software with theming back in. If Raskin really wants the One True UI to dominate, the answer is to get some developers together and write the damn thing himself -- if it's as good as he says it can be, everyone will use it. If not
Quantum mechanics: the dreams that stuff is made of.
It all depends on situation.
:)
Let's take a clothes analogy. If someone came up and said that this type of dress is the most functional for everyone on the planet... that'd be insane. Yet, many people believe that children are more disciplined and focused when they have to wear a school uniform. Similarly a mega factory should have all the computers with the same UI, yet I should be able to make my personal desktop look like whatever I please. Life is full of learning curves and going to a friends house to browse the web in his funky new skinned desktop is a mere nuisance as well as an insight into the users lifestyle as well. Bottom point: Leave things the way they are... it working fine and nobodys breaking down cause their spouses desktop is shaped like a flower...
(But they may soon enough
Eddy.WriteLinux.Com
...in the definitive sense of the word.
I have worked at a company where the user was forced to have certain windows in certain locations, with certain sizes, &c. in order to achieve this vaunted "efficiency by consistency". This was one instance of a mindset that was prevalent in the company. I found the environment opressive, even though as a support analyst I didn't need to conform to these standards. I did get flack over my use of a favorite text editor over the "approved" editor, though.
The company may or may not have had the right to make that imposition, on the understanding that efficiency is the most important thing. That is a separate argument entirely. But to say that UI Experts should dictate what everybody's desktop looks and/or feels like is to remove my ability to use a working environment tailored to my tastes in an environment where I should have total control: on my home computers.
Like Jef, I place a premium on efficiency. It is for exactly this reason that I think the ability to customize my UI is a must. As an example, black on white is fine on a passive piece of paper but I find it blinding on a computer screen. I set up my UI to behave the way I think so that I don't have to think while I'm using it.
To force my computing environment to be the same as someone else's -- the same, in fact, as EVERYONE else's -- well, I don't like that kind of centralization of control. It assumes that all people are interchangeable, uniform, and homogeneous, and it assumes that the people making the decisions are infallible. I have never been in any situation, computer-related or otherwise, where either of these conditions were true.
It's my UI to ruin my own way. It seems obvious to me that Jef would grit his teeth over some of the UI customizations I've made. That only underscores the need to allow me to make them.
Once upon a time, technology would restrict me from working in a computing environment I was truly comfortable in. Today, technology is capable of enabling just that. I think it should be allowed to do so.
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.
WTF this is great, MOD UP
Setting somebody's colors to red on red is just being a jerk. It's like setting somebody's default runlevel to 0 or 6. It's just being an idiot. Nobody would do that on purpose. It just confirms to me that this guy is an idiot because he thinks that doing that to somebody proves his point.
"When you use an interface for a given amount of time, your subconscious mind records its intricate UI elements, allowing you to focus on the task at hand rather than on navigating the interface. "
:
But Jef... Where is the fun? I use skins because
- they are beautiful
- they make me feel like I am in a science fiction world
- i don't care so much about productivity as my boss do.
In some degree, a CLI is a consistent UI that adheres to Jef's principles.
Even if Jef has studied on how to make me 69% more efficient doing my work on a computer, I still don't want a boring UI that is calibrated to squeeze as much juice as they can out of me.
PPA, the girl next door.
-- I feel better now. Thanks for asking.
What have you done for me lately?
The entire discussion neglects this complex issue. At the core of a usable interface is a good organizational structure. The tools used at and the appearance of each level is a secondary concern, worrying about these first is needless micro-management. Good organizational structure comes from a scientific analysis of user goals and a subsequent prioritization of those goals. The problem is that not all users share the same set of goals. If the designer chooses to prioritize with very broad goals(Start->Programs for example) that one level of the hierarchy suffers from clutter. This makes any task difficult. This example also shows the designer chose very broad goals based on a technical user's perspective, the interface structure clearly reflects the organization of the technical system underneath (Windows Update, Programs, Settings, Find, Help, Run, Shutdown). Help and Run near the end of course, they were clearly after-thoughts and placed early in the hierarchy for users who know technically what they are looking for. Overall, these broad categories are poorly chosen and are fit for a very limited user demographic.
Obviously a Network Admin vs. an Office User vs. an Artist have have very different goals. If you were to forget completely what you know about computers, and imagine any one of these people sitting down and turning a machine on, what are 5-8 broad categories of things these people would be trying to accomplish? Let me take a quick shot at it. The artist may want to do some video editing, some web work or maybe some sketches on their tablet. The Office user might need the use an office suite and maybe a few simple games(for lunch break of course). The network admin might want to configure some server applications or manipulate the configuration of the machine. All of these users would likely use email and the web too, these could be in an Internet category.
I propose a system to standardize broad categories of applications. Photoshop for example could register itself under Graphic Arts, Image Editing and Web Design. Image Editing and Web Design would be sub-categories of Graphic Arts. A cd-ripper could register itself under MP3 and Audio, with MP3 as a sub-category of Audio. An office suite would register under Office. A video tool like Final Cut Pro under Graphic Arts and Video, with Video as a subcategory of Graphic Arts.
The interface would sift applications down the hiearchy, only having them appear in the lowest levels. Photoshop for example would appear under Graphic Arts->Image Editing and Graphic Arts->Web Design but not under Graphic Arts. Yes, it would appear twice, but the sensible organization would make it unambiguaous how to find it. One of the few customizations allowed would be a feature to "bump up" an application so it appears in it's parent menu, for commonly used applications and users with few installed applications in a particular category. To allow this option in a usable way, a customization screen listing of all icons within their hierarchy would be suitable, with a trinary radio-like toggle associated with each. The toggle could switch between an up arrow(bump up), X or check-mark. Another customization would be the ability to shut off a sub-category and have it's options selectively bumped up to the parent, using similar trinary toggles associated with each sub-category.
This idea hasn't been developed much and could probably use some revision but I think the basic idea is there and it shouldn't be much more complex than this. Maybe some of those snazzy open-source interfaces like Gnome and KDE could pull this off. At first users would have to manually categorize their applications, but with time applications could register with a standard database. Early on, a small app could even be made to search a file system and generate such a database using known pre-categorized programs. This could also be a good thing for the litestep people to try, to truly give their system a unique and useful interface, without the massive manual config file maintainence that comes with those added strengths.
>> We don't have GOTOs in modern programming languages (should I put one back in so that you can write spaghetti code if you prefer it?).
Every language has a goto that lets you jump to a specific label.
I use goto's all the time to keep from nesting if thens 50 levels deep. Or to jump out of a set of nested loops to a found or not found label. Goto's are one of the best feature of any language, but like any other feature, they are only good if they make the program easier to read and not harder to read.
I think his views on interface design are just as screwy. I hate many of the things on a Mac, but am unable to change them, because the OS won't let me. I prefer to have my min max and close buttons on the right hand side, because they are easier to use that way, when switching between mac, linux and windows. It is annoying to have them on the left hand side. I also hate how applications don't go away when you close their last screen.
--
What I want to be able to do is to make any changes I want to the user interface and save these changes to a web site on the internet. Then I want to goto any computer and set a preference to down load all my settings from the web site. Then I want the OS to always act like I want it to act. No matter what OS it is.
It is time to divorce the behavior of the GUI from the OSes and make that a seperate fully skinnable layer that is fully cross platform and should control even how applications of certain types layout and present their controls. Then I can enjoy the same computing experience that works for me on any system that I am forced to use. Not screwed up system that some self proclaimed GUI designer has decided is best for me.
-- Never make a general statement.
Mistake #1: everyone needs to be able to sit down at any machine within an organization and get right to work.
True in something like a call center or a bank branch, but for the most part in those situations you're interacting with an application on which you've had some level of training, some of which is about how to use the interface in the first place.
But for most of us the set of machines we deal with from day to day doesn't change much, and we're free to set them up the way we like. In situations where more than one person uses a particular machine, I find that people are pretty rational about keeping customization to a minimum.
Mistake #2: the science of user interface design can predict what works for me.
Wrong. It is very useful for predicting interface needs of a group of users about whom your knowledge is incomplete, but it can tell you nothing about the needs of a particular user. You can make assumptions, and if that particular user is similar to your hypothetical user, you may even be right. But you don't know for sure.
The only person who knows for sure if an interface works is the person using it. And that knowledge doesn't apply to anyone else. That's the whole point of customization, or skinning, or whatever--an interface, good or bad, can be made better (or worse) by a user for that user.
Yes.
I don't want knowledge. I want certainty. - Law, David Bowie
"Blah blah blah blah
Even if a user interface sucks, as long as its consistent, its a better interface.
Besides which, you are an amateur, I am a professional, therefore your opinion is pretty much worthless.
And anyway, Microsoft sucks"
I think that sums up Jef's "intelligent" comments.
Raskin is often introduced as a UI guru and the father of the Macintosh. Why?
In every account I have read of the creation of the Mac, Raskin had a product idea an a name. Jobs came in, threw everything out but the name, and created the Macintosh. "Macintosh" is a great name, and Mr. Raskin deserves credit for it, but, to the best of my knowledge, not much else.
AFAIK, Mr. Raskin's only contribution to UI design (aside from a myriad of articles and speeches lableing everyone else's ideas idotic) is the PET, a computer he designed after leaving Apple in a huff. I think he sold about six of them.
When Mr. Raskin creates a UI that people use and is successful, we should listen to him. Until then, I wish he would just pipe down.
When the Mac came out in 1984-ish, a big part of the computer industry (including the press) talked about ideal user interfaces. That was the whole point of the GUI pioneered by PARC.
A lot of the interfaces done up until that time were unique; 1-2-3 had a unique interface, Wordperfect had a unique interface.
What apple did that was revolutionary wasn't the GUI, it was they defined an environment for all programs to use which enforced a common user interface. The net result was groudbreaking because a lot of the basics were the same in each application. Remember, WYSIWYG was the hot buzz word (words?) back then, so the entire frame of reference was different.
So, the evolution so far was:
No interface -> Application Specific Interface -> Common User Interface
What's wrong with this picture? Well back in the day when the CUI was the hot ticket, the idea of desktop "metaphors" was in vogue. The trash can was there because you understood that deleting a document was like throwing away a piece of paper on your desk. So in fact, the interface of the Mac was supposed to be a metaphor for a user's desktop (in fact, its still called a desktop).
But a desktop isn't a good metaphor for recording music, or taking pictures, or doing an unlimited number of tasks.
Why shouldn't the garden hose have the same interface as the typewriter? Because each interface exploits the unique features and attributes of that type of device.
So in 2001, computers are several orders of magnitude more powerful than 1984 and we really have the ability to create much closer metaphors.
In fact, you could say this argues against a common user interface. It says the interface should be the best to do the job at that time.
I fall somewhere in the middle. There is clearly a value of having the idea of a desktop metaphor. There is value in having experience framework for doing computer related tasks. But ideally there is a balance between consistency and the best interface for the task at hand.
So I find Jef's interview interesting, but I think he's still stuck in a 1988 period of UI design. No offense to him, he's a smart guy and I don't discount what he says, but based on the comment's he's made, I think the computer world has passed him by.
You were mistaken. Which is odd, since memory shouldn't be a problem for you
There IS a science of UI design and usability. Jef came to his conclusions through years of development and scientific testing.
Usability tests are psychological tests. Gather a diversified group of test subjects. Split it into your control group and your test group. Set up your application and tasks identically for each group. Carefully choose the independant variable (UI element). Perform the test and then analyze the data. One can then use statistics to determine the relationship between the independant variable and the dependent variable (time to complete the task).
Jef Raskin is a scientist. I highly recommend his book. Yes, he sounded arrogant in the interviews, but the opinions he expressed are grounded in logic and years of study.
-Mike_L
mike@NOSPAM.tamale.net
Jef says:
No, you'd be admitting that your way may not be the best way FOR ME. How arrogant can you be?
Unfortunatly current skinning technology is pretty shoddy IMHO. Every time I try a new skin somewhere, I'm disapointed in some way. Current skinning technology is not customizable! You choose someone elses favorite design and are forced to live with its style. It's like having choose between a sports car in hot pink, and an SUV in green... what if I want a sports car in green?
There needs to be finer grained control. Key bindings and pretty buttons should not be glued together. I may like the default look of a plane MS windows style interface, but I may not like the key bindings and the sounds; those things need to be customizable seperately and easily.
Skinning/theming systems do not *have* to compromise consistency (between user's machines, sure, but I believe that's the users problem), as long as they're done properly. A really good example of how to do it very wrong is Mozilla. Why would I want my web browser to look completely different than any other app on my desktop?? When done this way, it truly does kill consistency.
The basic low level design of X makes it even worse, because skinning does not effectivaly (if at all) cross the desktop boundary; KDE has a theming engine, Gnome has another... For skinning to really be effective, there needs to be a lower level look and feel engine that all desktop systems can be built on top of. This would also greatly reduce the amount of work needed to build a desktop system.
I think it's time in the evolution of UI design to take another big step forward; I think enough is understood about how to build good interfaces that we (the programming community) could design something that is both consistent and customizable, and modular enough to survive the next 20 years of evolution.
I could go on and on of course, about the short-comings of current UI implementations, but I believe that no matter how poorly current systems might be designed, I refuse to agree with anyone who believes that science can negate the need to provide people with customizability, or that customizability is a bad thing. While I agree that many (if not all) current skinning implementations are horrible, I completely disagree with Jef's assertion that skinning/theming is a bad idea in general.
I find myself disagreeing with the consistency issue which the author brings up so often in the article. This article seems to be geared toward the type of developer who will be working on a lot of different machines at varying times throughout the day (or week). For a lot of us, especially telecommuters, our own personal/business machines are the only machines we'll interact with, so this consistency issue seems to be a non-issue. Furthermore, nobody will be using our machines, so such consistency/compatibility is again a non-issue.
My personal belief is that, if a modification increases your own productivity on the machines you use 99% of the time, a need for consistency goes right out the window.
His entire argument for his grand unified UI theory is that consistency increases productivity.
So?
Efficiency should hardly be the only basis for UI design. Imagine if they applied that philosophy to cars or clothing.
-acidboy
So as I understand the article, consistency is desirable for increased productivity because it allows the user to work at different computers without getting confused.
Did an article have to be written to state that? It is just common sense. Consistency is obviously the ideal in multi-computer work environments such as computer labs, POS terminals, etc.
However, it is preposterous to assume that everyone works like that. Many, like I, work on one computer consistently. Let us for the moment forget the "Gee golly, I'm expressing myself" customization arguments. I use my computer at least 90% of the time I use computers. And this is the truth: My customizations are all geared to make me more productive. And this is another truth: They do make me more productive.
For us unicomputartians, eliminating customizations is ridiculous. Should I have to sacrifice my productivity 90% of the time I am working on my computer so that I may be more productive the other 10% of the time I'm working on other computers?
Jef does make a good point:
"Of course, there are no really well-designed interfaces out there good enough to prove the point that you don't need preferences."
Maybe that is why customization is essential to my productivity. Maybe a utopian UI will eliminate the need for customization. I doubt it. Not when you put humans into the UI equation.
This should be obvious. Don't we as people interact differently with different people? Do you interact exactly the same with your mom as you do with your friends? Do you tell your mom dirty jokes while kissing your friends and telling them you love them? Well, you might, but it is still a sure bet that you adapt your interaction to the person you are interacting with. In essence, you CUSTOMIZE your interface for the benefit of yourself and the individuals your are interacting with. Why do you do this? Because each person you interact with is different and your relationship with them is different.
This can be made analogous to our interaction with computers. Each person's aptitudes and behaviors are unique. Also, our relationships with our computers vary: In other words, people do different things with their computers.
Can one UI be developed that can most effectively deal with ALL the differences in people and ALL the different work situations? Without the benefit of customization?
Heck no.
Maybe in some utopian parallel universe, it could be done. I highly doubt it. And it certainly will not be done in the present software development environment of feature bloat and "we'll fix the bugs after release."
Despite Jef's insightfulness, he does make some silly arguments:
"But what if you say to me, "So what, I like it better my way even if it doesn't work as well." Then, if I give you preferences, I am abdicating my role as a responsible designer."
Yes, I've seen people customize their interface in a way that makes them less efficient. But these people are determined to be less efficient no matter what UI you put in front of them. It is well recognized that there are organized, efficient people, and those that are not.
And if a person is stupid enough to change their colors to red on red, then that person is going to have a hard enough time dressing himself, let alone use a computer.
So it remains true: consistency for the group, customization for the individual.
You can't have Yin without your Yang.
Ihuj
ihuj_at_yahoo_dot_com
Skins are nice in theory, but in practice, all the skinning implementations I've seen suck. I can't stand them.
For example, take Windows Media Player 6 versus Windows Media Player 7.
WMP7 is skinnable, but WMP6 is not. WMP7 has a "WMP6" skin to make it look like WMP6. No problem, right? Wrong.
WMP6 was not skinnable, but it gave me a pretty good degree of control over what widgets got displayed. You want a minimalist interface with only the video window and the buttons? Fine, toggle a few checkboxes and half of the clutter is gone.
Unfortunately, in WMP7, the "WMP6" skin mimics the default look of WMP6 and nothing more. It includes all of the widgets and all of the clutter. There are no checkboxes to hide or show certain widgets.
Well, if you don't like it, just edit the skin, right?
You have got to be kidding if you expect a normal user to modify skins or to make his/her own.
Maybe WMP isn't the best example of a skinning implementation. I'm sure that there are some really good implementations out there with easily editable skins. In general, however, most of the skinnable programs I've seen aren't much better, nor are their skins so easy to edit that my mom could do it. What's worse is that a lot of these applications seem to use skins as an excuse to ignore usability in their default skins. Hey, if the community doesn't like it, they can't make their own skin, right?
You want customizability? Fine. How about giving the user a choice between skins or not? Me, I'd rather have no skins, with a faster, well-designed interface that's consistent with the OS.
Skins may have their uses, but they aren't some magic panacea.
You mean like changing the font preferences to chinese for an english user?
This article addresses issues that deal with the every day computer user, not /. readers. Computers have nothing to do with lifestyles or religious beliefs for 99.9999% of individuals that use them. Raskin points out the deficiencies inherent in computing systems for the everyday non-/. user, not Igor the l33t hacker.
-- uh...
Just because somebody might choose stupidly, that's not a reason not to let them choose. Just because I can paint a fake door onto a house and cause confusion, we should forbid paint? That's the point of *customizability* you know, that if I rented a car with the pedals switched I would switch them back goddammit. Everyone is arguing how terrible customization is, just because if you customize it one way I can't customize it back? Sheesh.
Tomorrow will be cancelled due to lack of interest
Good idea, and in fact, that's exactly how it's done. I've worked in several UNIX setups that share home dirs via NFS, and use automounter to locate and mount your home on each login. So, you login to any machine on the net, and voila, you have all your settings /and/ your files. Of course, this requires a fast LAN but is a standard practice...
So I kept wondering what the heck was Jeff blattering about in his article - customisations are a solved problem.
--
I refuse to use
Every first name is a palindrome!
Regarding he is a computer scientist, I doubt that this is an accident. :)
Regards,
Marc