“Apple released Mac OS X last year. Having used computers for almost twenty years, I can say that this new operating system is undoubtedly the best ever. Yet, there is an undercurrent of disappointment. There is something wrong with this OS. It’s not the MACH microkernel, the UNIX core or the fluid motions of the new GUI. Well, actually it is the GUI. It’s not a bad interface, it’s probably the best yet, but isn’t it about time we stopped pointing and clicking?” Read the editorial at MetaMute.
Graphical User Interfaces: It’s Time for an Overhaul
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
yep, and sony is going to try a system w/ a webcam and some balls for the PS2, but maby they will hold back for the PS3 instead…. It is a very good Idea, but really Star Trek like computers would be better for ignorant people because it could tell them what was wrong if something went wrong or find files just by the user saying ” I want so and so file(The paper I was typeing yesterday; The {place band here} songs I riped last night)” Don’t discount A.I.
A great many of the aspects of our so-called “modern” computers are sorely in need of revision. Unices, Linux, Mac OS 9/X, Win XP- they’re all quite firmly based in ideas that were conceived of in the 50s and 60s. No, that’s not bad in and of itself- and if it’s not broke, don’t fix it. However, while the computer market has remained largely stagnant since the 80s, since the days of MS-DOS, SysV and the first Mac OS, computer science has moved on in ways that would benefit user and developer alike. The computer market holds the state of the art back in everything but backwards compatibility.
The GUI is one where we don’t actually have 20 years of research of something better than what we have now.
Don’t discount A.I.
What’s that supposed to mean?
Sure, eventually, we might be able to interface with computers ala Star Trek, but we are very far off. AI is easily discounted now, there are definately non-research applications for many facets of AI now, but not in the way most laymen think of AI and robots.
“Star Trek like computers would be better for ignorant people”
i think that, with better voice recognition and the right setup(yes, some ai will be needed to make this work best…) that a computer system/network/starship could be much more productive… imagine being able to say, sit at your couch and ask the computer to put on some Mozart to relax with. at first attempts at making this system may be crude, but i have faith it will lead to a much improved experience overall.
Why is it people say the desktop is dead?
Or
Why do people think there is going to be something else?
our current desktop GUI are the way they are and not changing because they are of the most logicaland common sense design. Most of the apps work like they would if you were doing it on a peice of paper, it makes sence that way. 3D desktops haven’t happened because that is a step backwards. Desktop means to be like the top of your desk, to be like the real world. And to use it that way. We don’t do paperwork and such in a 3D manner it’s not going to happen in a computer. The fundementals of how the desktop is laid out is not going to change much aside from details since it’s peaker where it needs to be. It’s like your tv, it pretty much peak when we got to color and a remote control. Everything since is refine ment. They don’t change in fundemental ways since there is no place left to go. Unless you want Smell-o-vision, or Taste-o-Vision, or Feel-o-Vision. And there isn’t much point. Ok well the last one could have it’s moments. and still they would just be add ons.
About the only big change you could do is the talking to a computer with no screen, (talking to one with a monitor would just be an addon to current design) the star trek idea would be usefull for things, but not for sitting at your computer editing a photo or browsing the internet. It would be for you house and such. Besides being in a office of people all talking to their computers would be hell.
Part of me thinks the whole “desktop is dead ” idea is around just to give people something to write aricles about, say in forums, and to give them a job in some R&D lab some place.
the desktop idea isnt dead mainly because it works
how is a voice controlled system going to be the least bit time saving for anything other than a home media player?
i do think that a 3d environment is possible as an interface if its a full environment (goggles, gloves, etc..) but not on a 2d monitor display. (i will note that microsoft research did a 3d interface project that truely was promising for such a requirement, but they havent done anything with it obviously)
the author of this article also attacks mice, but gives no reason for his opinion that it is the worst interface ever. this article is just someone speaking about something in which he has no real bearing on.
Anyone know at what university the author is studying for his PhD in “New Media interface design”?
The problem is backwards compatible, legacy cruft in all our operating systems.
And that problem is never going to go away, only change (old cruft, meet the new cruft).
And there are projects working on new ways of interacting with computers like MIT’s Project Oxygen http://oxygen.lcs.mit.edu/Overview.html
Let’s see, we get an article hosted on a completely horrid website that takes 7 paragraphs to say:
The desktop metaphor is dead, give us voice recognition.
Never mind that as much as the complaint is something along the lines of the computer being a ‘glorified type-setting machine’ (what? umm type-setting doesn’t account for the majority of what I do on a computer), the only solution offered as an option is turning the computer into a glorified dictation machine.
As an added bonus, there’s this line:
‘Few users today realise that by clicking on Macintosh HD they are programming the computer’
Sorry, but someone else programmed the computer so that you could view the directory in a little window, You aren’t doing anything of the kind. WTF is a PhD in New Media Interface Design, and how do you get to the point where you can make the claim you’re researching for that degree without having a clue how a computer works?
1) Screw any enhancements that aren’t going to make me more productive or easier on the eyes. I could give a rat’s ass about special effects. Hell, if I want special effects, I’ll go rent a Star Wars movie.
Point and click works, and it isn’t broken. Don’t attempt to offer me a solution if it attempts to solve a problem that doesn’t exist in the first place.
2) I hope 3D GUIs aren’t like 3D platform games. I don’t have a good sense of direction and tend to get lost easily playing those things
3) I personally don’t like the Mac OSX GUI at ALL! Especially the dock .. why is the damn thing so big? It’s completely counter-productive. I like my launch bars to be small so that I can cram as many icons on it as I can. On my Win2k box, at 1024×768, I’ve got the quicklaunch toolbar going all the way across the screen (minus the start button)with icons all along it and then some (50+ total).
People, get real.
The reason why keyboard and mouse have been around for so long without major changes to them is because they are the most efficient ways to interact with a computer. Voice recognition, you say? Just how fast can you talk into a microphone to the computer? How many instructions per minute can you give it? And what’s going to happen to your tounge after you spend two hours working on your comp like that? Imagine an office with a hundred people in it all mumbling something to their computers. Imagine playing Counter-Strike through the voice recognition, LOL.
Even in StarTrek they only use voice interface for a very small fraction mostly still relying on keyboard.
I hope you realize that the dock is fully customizible. You can make it VERY small, though if you use OS X regularly, I think you’ll find yourself making it bigger again and moving it to one side or the other. That seems to be the most productive setup as far as I can tell. Its also quite transparent, and when you have it set to be hidden, it pops up and back again very quickly, unlike the taskbars in most linux distros I’ve tried, which just can’t match it for responsivness and ease of use.
The artical, though a bit on the pointless side, still does get the mind going a little bit, and Im glad I read it. One brainstorm I’ll put forward is that there is a very important thing he is missing. The primary reason we are still on the point-and-click desktop model is that our feedback still comes through a monitor. As long as that monitor is our output device, we’ll keep using a mouse and keep clicking on our desktops, because it just does not work well for all these next generation interfaces everyone keeps dreaming about.
The desktop metaphor we’re talking about here has, fortunately, moved away from the real desktop long ago.
Real desktops don’t have trash bins on them. Most of the windows I’ve ever seen are firmly fixed in walls (the same with wallpaper). Menus and taskbars have no equivalence in desktops, either.
Anyway, it is not because it works now that no one is going to research some different ways of doing things. The abacus has once been a very good calculator, until they invented the computer.
> Point and click works, and it isn’t broken. Don’t
> attempt to offer me a solution if it attempts to solve a
> problem that doesn’t exist in the first place.
> The reason why keyboard and mouse have been around for
> so long without major changes to them is because they
> are the most efficient ways to interact with a computer
We pointed and “clicked” thousands of years ago, when we didn’t have well developed spoken languages. Try driving a car with a mouse and a keyboard. Using a computer is an experience at least (if not more) as rich as driving a car.
A brain-computer direct link would be the ideal interaction device, but it will take some years to get commercial…
what we need is a pad that can follow the hand motion and corelate it to a position on the screen, then all you need to do to open a file or folder is move your finger down, and to move a folder or file from one location to another, you just need to make a fist over the item…..
that is the only improovement I see in the future, and we are still using a gui.
the GUI is at a peek…it peeked in the late 80’s and was popularised with windows 95. everything else has been refinment….
I can see a place for genture point and click however.
#1. “…an Apple Lisa user from 1983 could understand OS X as part of a continuum.”
There were Apple Lisa “users?”
I keep reading articles saying that the current desktop interface is ancienct, decrepit and dead. Has anyone made anything better? Anyone can point and say “That’s old” but not many people have tried to make anything better.
This guy doesn’t even give many examples, he just states a bunch of facts that we all know and anyone can find on a zillion other pages.
This article is a rehash of many, many others.
Actually the guy didn’t just mean Voice Recognition that we really do need to have the most efficient way to deal with our PCs, Buy he means we need diff. ideas than those old dull ones. Think about Integrating Mice with keybords so our hand wont travel much during a PC work, better yet think about a more advanced Shortcut system used on our PCs. Let MS GUI idiots look how the hotbox is used efficiently in Maya 4 By AliasWaveFront. I for example can not switch the WindowsExplorer ExplorerBar form History Bar (Ctrl+H) to a FolderBar with any shortcut because MS didn’t assaign any shortcut for it. Try to find this out go to Start>All programs>Accessories>Windows Explorer, Then Click on the following path of the menu bar : View>ExplorerBar>… Now you can see that MS just assaigned 3 shortcuts to Search, Favorite, History no more, all of this ****ness is happening of what is called the best innovations since the sliced bread “WindowsXP professional”. Can’t I be more clear.
what we need is a pad that can follow the hand motion and corelate it to a position on the screen, then all you need to do to open a file or folder is move your finger down, and to move a folder or file from one location to another, you just need to make a fist over the item…..
I’ve got something like that. Wacom sells em.
Actually, you can assign IDs to the devices you use on a Wacom digitizer (stylus, mouse, etc). Maybe you could fashion a glove with the equivalent to stylus tips implanted strategically on it.
what I WANT is a for someone to make the same interface used in Minority Report into a reality! i would absolutely love it if everything on my computer were “objects” and I could manipulate them with my hands.
it’s really just the old power glove idea taken a step further. people don’t seem to think about how the mouse is really a step between you and the computer… you translate what you want to do into mouse movements, which translate to on-screen movements. removing the mouse and directly controlling what’s happening via your hands is a step “closer” to the computer.
a more clearer example:
i want to throw something in the trash can.
click, drag to trash, release
vs.
grab object, drop in trash
which is more natural to human thinking?
=) now if someone wants to take it a step further, you could implement physics…
grab object, throw into trash! if you didn’t throw it hard enough, it will fall short of the bin. =) how cool would that be!
It’s one thing to say the GUI is dead, but this really needs to be backed up with a viable alternative interface.
Today’s 2D desktops are all pretty much the same whether its WinXP, Mac OSX, AmigaOS, BEOS, KDE, GNOME. If you’ve used one, you can quickly adapt to another. PDA’s use pen based variants but are pretty much the same.
3D interfaces have been suggested in the past but SIMPLY DON’T WORK. They make look very flash, but just aren’t practicle. Computers are generally used to produce output that is inherently 2-Dimensional. How would you edit a Word document in 3D?
So what does that leave? Speech recognition? Thought recognition?
Nah, I think the GUI is safe for some years to come. Simply IT WORKS. Sure there will be improvements, variations, and a lot of flashy and pointless eye-candy effects, but the basics will remain.
A comforting thought for all the Apple Lisa users?
This is getting boring. Every few weeks someone
bashes the desktop metaphor, storing files in
hierarchies or the point and click interface.
I always go over the article and hope that at
least one useable idea is offered by the author.
My hope is always in vain.
Unless you really know what you’re doing (command
line), the computer should let you know the available
options. And what is more natural than to point at
one? Only replacing mousewith touchscreens can make
this more intuitive.
The arbitrary hierarchical storage for your docs?
It works way better than people seem to think. You
can test the alternatives with BeOS live queries,
but there doesn’t seem to be anything better.
Desktop metaphor? I’d like to see something better.
But so far I haven’t heard of any really good ideas.
I mainly want the big curved screen. Though the hand tracking (no glove) might be handy from grabbing things from the far reaches to bring them in for precision manipulation with a mouse. I could use the exercise too.
In the end it is still a “desktop”. Not such a bad thing really.
PS: I was just imagining myself waving in windows and then dismissing them with a flick of the wrist. Its good to be the King
“What is required is an acknowledgement that the entire
desktop metaphor is holding us back from making computers that are more than
glorified typesetting machines.”
Typesetting and graphic design are what computers are best at. To
anyone who has spent hours laying down Letraset, the ease and
convenience of designing a layout on a computer screen is wonderful.
I do plenty of things with the computer, but the typesetting is the
one function I would pay for. Everything else can be done as well or
better by traditional means.
Jermey,
the down side to that is it would be very tire some. Take your arm and hold it up in the air, out away from your body for a while, you and move it a bit but just do it. It gets very tiring very fast. I don’t think it would be liked over a mouse.
Minority Report,
You are going like jermey, but i say basicly the same thing. The place i see this working would be Keyoscs (sp?) in the mall and such. Where you walk up and use it for a bit. Like a mall map or city map for be great. You walk up to it and maybe speak to it about were you want to go and it takes you through but you could also grab hold of the model and move around in it as well.
Actully i think if we could do that i htink having a arrow show up hovering were we are and it points to were to go and it walks with you guiding you would be slick.
hraq
Those kinds of keyboards are out there. I have one from IBM, it has the track point in it like on a latop. It was ment to plug into think pads to get a full size keyboard. But it doesn’t have things like the number pad on the side. It has the keys that get used. Its nice and small. And is really nice since you can hold it with your forarm and use it with the other so you don’t need a desk to your computer. I use it when I lean back in my chair real far.
I really don’t get peoples problems with current ways of filing things. We do it the way that makes sence. I can see people having trouble since many people just put ever thing in the my documents folder with no orginization. But as long as you think with some common sence things work fine. Just orginze things so if you had to find it fast you instints would guide you right to it. It’s fine if it take you ten kicks though folders as long as it only took you 2 seconds to get there. Good rule of thumb, orginize your files like it’s your job and maintaining your job depends on it. You will make the right choices.
“A brain-computer direct link would be the ideal interaction device, but it will take some years to get commercial…”
Yeah, right, just what we need, a way for Microsoft to eavesdrop on even your innermost subconscious thoughts. Welcome to Hell.
Please, what you are doing when you point and click about in your applications is that you more or less tell the computer exactly what to do in every itty-bitty detail.
Why is this? Because the interface is more or less stupid, it’s not an interactive system where you use your brain at what it’s good at and use the computer what it’s good at.
A modern day desktop interface doesn’t assist me in anyway with the tasks I want done. It simplifies organisation of the tasks and specific items within that task but it doesn’t provide any additional value. I am not better at solving equations, developing software or writing an essay just because I do it in front of a computer. I might do some of those faster but not better.
This is what I feel is really lacking in todays computer interfaces and applications. They should be able help me in my tasks, to provide additional value when I do something to complement my thoughts with what a computer is good at instead of just sit there like a dumb box waiting for input.
A brain-computer direct link would be the ideal interaction device, but it will take some years to get commercial…
And how would you distinguish which thoughts are to be recognised by the computer, think about it. There is no real way to tell the computer wiith your mind to recognise some thoughts, but not others. Imagine what it would be like on chat rooms, when the real truth comes out. (Flame wars, would probably become the only thing on the internet) And think about all those teenage boys, handing in assignments that have descriptions of naked females hidden inside the essay. No, thought recognition would not work.
Neither would voice recognition, when a moniter is your primary form of output. Presume you’re running a screen with a resolution of 1024 * 768 (fairly standard). Humans can usually move a mouse to within a 10 px * 10 px square within a second. That’s about 7000 options which can be displayed and selected within a second. Using voice recognition, you would need to be able to get it to distinguish between commands and input for text documents etc. (eg. Making sure that when you type “Close the door, it’s cold” it doesn’t close the program, then input “the door, it’s cold” into the next program. ) So that wouldn’t work with a moniter.
And why move away from visual as your primary output. You can’t quickly scan through audio, but you can look at for the words you’re looking for with visual. This is amazingly helpful on the internet. Imagine having to listen to everyones username, location, post date, email address and signiture, as well as the messege, at every forum you went to. It would be time consuming, repetitive and just generally annoying.
The current system has been thought through almost completely, it’s logical, it’s easy, it’s powerful, it works
>what I WANT is a for someone to make the same interface used in
>Minority Report into a reality! i would absolutely love it if
>everything on my computer were “objects” and I could manipulate
>them with my hands.
He is mate, they guy who designed the 3d ui in minoriy report is named Jared Lanier, he also did the ford concept 3d projector (almost as nifty as te minority report ui)
>it’s really just the old power glove idea taken a step further.
>people don’t seem to think about how the mouse is really a step
>between you and the computer… you translate what you want to do
>into mouse movements, which translate to on-screen
>movements. removing the mouse and directly controlling what’s >
>happening via your hands is a step “closer” to the computer.
it has been done… linuxvr amongst other things.
The reason why keyboard and mouse have been around for so long without major changes to them is because they are the most efficient ways to interact with a computer.
Let me add my two cents : The reason why keyboard and mouse have been around for so long without major changes to them is because they are the most efficient ways we currently know how to interact with a computer
Even in StarTrek they … still relying on keyboard.
I feel this is the best way to zoom around my computer. I only use the mouse when I am forced to. If I want to close a window, or Copy/Paste, I use the shortcut keys over the mouse. I actually dislike the mouse for many different operations, as it takes longer for me.
I don’t think that voice recognition or some hand references (Nintendo Power Glove) is the next big wave. We need to always keep looking though, or else something more effective could go undiscovered.
This article stated many points of opinion without solid facts or sources to back up their claims. That is why I think the article was a waste of time to read. An article where the author rants and raves about their far-fetched ideas with solid backing is a better read than those without.
Thank you for your time;
mouth
Conspiracy theory: all the companies planning to release a “next generation” GUI are “encouraging” the “journalists” to write articles about how old and pointless the desktop/mouse metaphores/methods are so that the mindless computer consumers to believe that buying the “next generation guis” is a “good thing” (thankfully most of you folks here aren’t falling for it).
Probably just a news fad, but conspiracy theories are fun (more fun the more unlikely they get). Especially at 9am, when you’re weary from only 1 hour of sleep.
There is an improved version of WIMP emerging. It’s called Palm OS. The only place in Palm OS you *ever* see icons is in your Applications folder; in every other application, documents are in lists. Some applications have menus, but those menus generally have few more options than Preferences, Cut/Paste, and About – so the average user may never have to encounter a menu. Text can be selected and the text cursor moved in traditional ways, but there is no mouse pointer, in favor of an absolute-positioned touchscreen – in effect, the “OK” and “Cancel” buttons on the screen might as well be physical buttons. Generally, no one will ever need to activate a hidden menu or feature in any app.
Why not extend this metaphor to desktops? Styli are awkward and can get lost (I can attest to that
), but a suitable replacement is available in the form of eye-tracking – for example, if I stare at the “email” field on this form for a second or so, it will bring me there, and the same for the Submit button. Or, you could replace the mouse with a Palm-sized LCD touchscreen that displayed a list of applications one could run, or features in the current application that one could activate. (Think of it as a universal remote control for the PC.) For content manipulation apps (everything from word-processing to Photoshop and beyond), you can replace LCD screens with high-resolution touch-sensitive Bluetooth paper (Bluetooth-enabled superthin screens, perhaps OLED-based). The combination of Bluetooth paper and mini-control LCD together can replace the monitor as it now exists.
Imagine the possibilities with a system like that. Content and control are completely separated, making it painfully easy for any user to figure out what they can do with an application – just look at the remote control, and anything you can’t see, you can’t do. No more mouse pointer to get lost, or to activate different menus depending on where you are (god… menus may be bad, but context menus are even worse). You get the point.
That most of those speaking about keyboard and mouse (slightly off topic, no?) would be happy with this : http://www.fingerworks.com/index.html ?
NOT. Just bashing current UIs without ANY reason just by saying it is bad.
PainKilleR: Let’s see, we get an article hosted on a completely horrid website that takes 7 paragraphs to say:
Well said. But wait, are you the same PK from osOpinion? If yes, hey! This’ remaja here 🙂
Admiral Horror: Even in StarTrek they only use voice interface for a very small fraction mostly still relying on keyboard.
Uhmm, the last I check, they didn’t have keybaords. They had touch screens, running a system with one of the best Hollywood UIs ever, LCARS.
Devon: You can make it VERY small, though if you use OS X regularly, I think you’ll find yourself making it bigger again and moving it to one side or the other.
Making it small makes it counterproductive. Why? Two reasons
1) The icons are photorealitic. They may be nice at 128×128 or even 48×48, but completely useless at 16×16 or 24×24.
2) There’s no label besides the icons, forcing you to go through all the icons before finding the one you want.
Now. if they put the icons from OS 9 (no matter how ugly they are), your suggestion would be feasible.
You can never tell what will happen in computing, but I think the basic desktop metaphor was a huge step that will continue to be around. I see there can be complements to it…voice recognition in certain areas, tablets in certain areas but, in a historical sense, the GUI and desktop metaphor just happened yesterday. What I don’t like is that average users don’t even take advantage of it – they install AOL and never leave it to do anything else. Another reason why so many average users know nothing about computing.
Keyboards work. Mice work. They’re great! I can type much faster than I can write, as most people can. The keyboard is one of the best inventions out there, and the mouse isn’t bad either. The mouse is extremely intuitive! You move your hand, the “hand” on the computer moves. What more do you want?
As for things like file systems, what’s wrong with it? I once saw this interview with a “futurist” who said that heirarchy doesn’t work. He developed a file system where things were just arbitrarily placed and required no title or position. Which is easier to find: your English homework in the My Documents/English folder, or the printout on your wooden desk?
Although it’s not what we’re used to, it works, and we should probably be using it instead of randomly throwing papers places. In fact, it was based off of something that we use! Large metal file cabinets!
I think the problem with the Desktop is that it is basically a model that was developed in a non-connected world, i.e. before the Internet existed. This was a time were people experienced work as a solitary experience working at a desk. This is no longer the model today.
With the Internet people are communicating and collaborating electronically all the time during their work day. The Browser is the tool that has changed all of computing, from information access, to application development. The problem is that the Desktop, and the Internet, are not designed in a compatible way that aids people to do their work.
This problem can be defined with examples of things that we use that really need to be removed. Things that need to change are: the seperation between and application and a document. There is no need for applications as a seperate thing. Windows are no longer needed. We don’t need to package sections of the screen like this anymore. Files, these are no longer needed, as they are a technological idea imposed on users. Users don’t understand, or care about hard drives, we don’t need them.
The real combination of the Internet and the computer will yield a new computing model where there will be no more Browser, but rather the entire computers OS will be dedicated to group communication, collaboration, and cooperation. This will be the next stage of evolution of the Desktop… the Webtop.
There are different sorts of input and output, meant for different enviroments.
Voice recognition is well suited for a home system, to control the lights, turn on and off the tv or radio and stuff. A mouse is best suited as a pointing devices, which it is currently used for. But I see that a mouse is better suited for a 2d environment than a 3d one. Touchscreen is also a pointing device like a mouse, really usefull in a 2d environment. A glove is also some sort of ‘pointing’ device, better suited for 3d than a mouse. The we also have a keyboard, which is used for text input. What about a remote control? Such a device works great to control a tv.
How would you write a letter in a 3d environment? NOT! How would you manage a desktop with voice recognition? NOT! For writing a letter we use a keyboard, and for managing a desktop we use a pointing device.
For a workstation on your office, I can’t see a better sollution than a monitor with a keyboard and a mouse, or some other well suited pointing device. At the mall, library, what about a touchscreen for simple input? an extra simplified keyboard for text input? I think that’s a perfect sollution? voice recognition is good for simple commands in home, but for programming the tv channels I still prefer a remote control.
A good example to see how a mouse could really work well in a 3D environment is the game Black & White, it’s really impressive. Also, round menu’s located at the mouse pointer is very interesting and productive too, iirc there’s a lot of research on this.
You just need to find the best sollution for the best situation. Easy as that!
But to come back at the article, the keyboard with the mouse and a monitor is still a perfect sollution.
Those are my 2 €-cents.
Click and Click.
I don’t do point and click much. I use an A4 Tech 4D mouse. Have been using it since 1999. With the wheel and the button I can open any document or programs regardless of where my mouse is pointing at that time. So this is no need to point somewhere and then click.
Mouse Gestures.
I got introduced to mouse gestures with Opera browsers and loved it ever since. Made a search for a similar utility to be used on a windowing system. Found StrokeIt, a free util. This one can also be gesture trained to launch programs, but being accustomed to the A4 Tech I didn’t needed it. Other gesture commands such as New, Open, Back, Forward and many more you can train is quite useful. I feel that if I am happy with mouse gestures I might as well be happy with hand gestures. Some one please send me a sample J .
Meta Data
With the coming of Yukon, I think there won’t be any more Hierarchy file system view. Devices also could come with firmware having knowledge of where to download relevant drivers (no need for our “point and click” attention here).
Desktop
There are a lot of interesting ideas presented in the screenshots submitted by user on Object Desktops website. I particularly liked the MSN desktop shown there.
http://www.stardock.com/products/desktopx/screens.html
Wide Screens
Personally I think the Monitor is the Desktop. As long as the Monitor is our feed back method we wont’ be seeing a quantum leap in the way we work. It will still be the “Desktop Metaphor”. But a 16:9 screen I think is a very productive environment. And products like Microsoft’s XP MCE, and apples iPhoto seems to be great productive products. In the formers case they are introducing a realistic remote control instead of a Minority Report style hand gestures. With apples iPhoto it is simply great software engineering at work. Both of these would greatly improve if they could relate to our
Motion Sensors
As far fetched as hand gestures may look, the concept of motion detection should greatly improve the way we work. We can start from simple steps, like using motion sensors to toggle with hibernate on Operating Systems. Also the efficiency of sign language is admirable. A very simple hand gesture language should be easy to adept.
All this said, the authors original article reads a little better on his own site here:
http://jasonwalsh.net/april02/killyourmouse.html
>>I personally don’t like the Mac OSX GUI at ALL! Especially the dock .. why is the damn thing so big? It’s completely counter-productive. I like my launch bars to be small so that I can cram as many icons on it as I can.<<
Why don’t you actually use the OS instead of toying with it the Apple store? you can make the dock as small as you want to fit with your “I want to go blind looking for apps” preferences. BTW, the dock even autosizes itself so that if you add more icons than it can hold at its current size it makes itself smaller to fit them all on the screen. How can you not like that? It is even offering to clean up things for you.
User input:
The traditional keyboard and mouse click have been supplemented by mouse gestures (Opera) and mouse chords (Plan 9). These have much of the versatility of the “Minority Report” glove, without needing stereoscopic webcams and gigaflops of signal processing.
Computer output:
The desktop metaphor is deeply flawed. My actual wooden desk can have many documents arranged on it, that I can view almost simultaneously. The computer screen is like a sheet of cardboard with an 8 inch by 6 inch cutout, placed on top of my real desktop, that has to be pushed around to let me see what’s underneath, and won’t even let me see an entire page of a single document. Until we have “Minority Report” screens that are comparable in size to a real desktop, the desktop analogy should not be pushed too far. But it needs real imagination to invent something better – which is why the desktop metaphor has not been displaced, 30 years after its invention.
Data:
The heirarchical file system is a brilliant invention, and it is hard to think of something better for the underlying data structure. But it’s hell when files get disorganised – and many non-techie users aren’t aware of the tools that can be used for finding misplaced files. Such files are permanently lost. Yes, better file metadata would help; so would a database-backed search engine (see OSnews articles on Windows Longhorn). It’s easier to find help files on the internet (using Google) than on my own hard drive.
In this case, a solution to the problem is known, but of course it is easier to describe it than to program it; and, make no mistake, we need such a solution if free software is to compete with Longhorn.
Compared to what you see in Minority Report, it’s in its infancy, but [url=”http://www.squeak.org“]Squeak[/url] is like you describe. Every tiny piece of UI, down to every button, every piece of text is but an object that can be directly manipulated.
This is great for programming/testing/prototyping/debugging “normal” apps as well using it as a playground for the kind of stuff you see in the fictions like Star Trek or Minority Report. It also comes in handy as a user environment.
For example, when I was doing some modifications to the built-in Squeak IRC client, I was often closing part of it and re-opening it [1]. One can start the IRC client via the “World menu” (a menu of various apps and development tools, comes up when you click on the desktop). But getting to the place where the app is listed takes a bunch of clicks, and I wanted to restart quickly. Sure, Squeak has sticky menus, but who wants a whole menu to sit on my desktop? Using the middle mouse button, I navigated down to the menu option that started up the IRC client, selecting layers inside and inside. Once I got to the “open > irc client” option, I grab it by its “handle” and made a copy that I had sit in the lower left-hand of my screen. Deleted it when I was done. After you get used to not just being able to *customize* your UI but manipulate every aspect of it, directly, it’s hard to go back to even OS X.
[1] Actually kind of rare in Smalltalk to have to re-start apps you’re working on- changes go into effect immediately. However, what I was working on caused the connection to become corrupted, so I was restarting the app to clean up the mess I was making.
So your saying you want more of Clippy, All clippy all the time. Throw in wizzards for everything that can’t be turned off?
Yeah, that’s right. Microsoft BOBs are really the end of computer assisted work, they have sure helped me a lot. And if you don’t get it, I am being sarcastic.
If I KNEW what was needed to do this with a computer I would have done it already wouldn’t I? I mean, it would only be the most fundamental breakthrough in human computer interaction ever.
Why do you think I have a solution to the problem I stated? I merely stated what I would like my computing experience to be like not how to do it.
reading a bunch of posts here saying point and click is the most efficient that ever will be to using a computer, I think you can safely say that these clowns arent going to think of the next big thing that moves technology forward.
To tell you the truth, these nitwits would have never thought of the mouse window interface in the first place when there were only command lines.
yep when making a comparison of how things were between 1900 and 2000, you can safely say that it wasnt people like you who made the difference.
> Never mind that as much as the complaint is
> something along the lines of the computer being a
>’glorified type-setting machine’ (what? umm type-
> setting doesn’t account for the majority of what I do on > a computer), the only solution offered as an option is > turning the computer into a glorified dictation
>machine.
Metamute is for graphic design and art proefessionals, and as such the article is aimed a perople very much like myself, for whom, the comput is indeed a glorified typesetting machine.
Had I been arting about computers per se, I may have reached different conclusions, but I wasn’t. All writing is targeted at a specific audience.
> As an added bonus, there’s this line:
> ‘Few users today realise that by clicking on Macintosh > HD they are programming the computer’
> Sorry, but someone else programmed the computer > so that you could view the directory in a little window, > You aren’t doing anything of the kind.
Actually you are – instructing the computer in any fashion is analogus with ‘prograamming’ (which in itself is a redundant term). If shell scripting is programming, so is pointing and clicking.
> WTF is a PhD in New Media Interface Design, and
> how do you get to the point where you can make the
> claim you’re researching for that degree without
> having a clue how a computer works?
Let’s see, a PhD is a doctorate and interface design is… interface desugn. And as for the person who asked where I was studying, I live in Eurpoe, so there’s a hint.
I do have several ‘clues’ as to how a computer works, but I am not some wing-nut programmer. The degree is based on information architecture and graphic design.
The original article
Kill your mouse,
http://jasonwalsh.net/april02/killyourmouse.html
April 2002, Jason Walsh
As far as users are concerned, the Graphical User Interface is the computer. The GUI and its attendant desktop metaphor are what we all interact with, every time we use a computer, and though, in usability terms, it is a vast improvement from the command line, the processing power available on even the most humble PC should allow us to move on.
GUIs allowed new uses for computers, notably desktop publishing, because they promoted interaction with computers in a manner understandable to those of us who weren’t educated in computer science, but an Apple Lisa user from 1983 could understand Mac OS X as part of a continuum.
The paradigm is flawed – it was fine in 1983, but the processing power of the average el-cheapo generic Windows box, or iMac would be better deployed replacing the anti-intuitive desktop metaphor, than rendering old mistakes with special effects. When the iMac debuted, commentators noted that the round mouse was clumsy. Well spotted. Sadly, none of these people went on to note that mice are generally ill-conceived devices.
Primitive 3D GUIs already exist, notably for the various UNIX operating systems. The Silicon Graphics Inc engineers, Joel Tesler and Steve Strasnick, have written a 3D filemanager, FSN. 3Dsia is an open source alternative, this time attempting to be the entire interface rather than just a filemanager, based on descriptions of ‘cyberspace’ in William Gibson’s Neuromancer. It is interesting to ask why Gibson’s fiction should become a model, after all, he wrote it without ever having used a computer?
If major firms like SGI have invested time and money in 3D GUIs, why have yet to see a commercial development? The developers of 3Dsia claim to be creating an intuitive 3D environment. As Gibson was, at the time, not a computer user there is some cogency to this argument, as he was conceptualising something from the perspective of a non-user. However, Gibson’s work is fantastical – he was not intending his computers to seem intuitive, rather it was the obverse which he was interested in. His machines are complex and frightening – extensions of the complex mainframes embedded in our culture at the time.
The developers are wrong. There is an innate problem in displaying 3D on a 2D screen. No matter how well visualised, it remains an optical illusion. Furthermore, 3D will entrench the problems of the desktop. I have difficulty locating files in the mess that is my computer, making me look underneath a virtual-sofa isn’t going to help.
Futurologists have long foretold the prospect of voice recognition, so where is it? Clumsy attempts by IBM and Apple amount to little more than automating desktop events. What is needed is software that knows when we’re present, when we’re addressing it, and has some basic conception of its purpose. This doesn’t require consciousness, or so-called artificial intelligence. There is a game for SGI workstations, Lumbus, which uses webcam motion tracking as the control system. It’s a fairly small step to include face recognition and give the computer an idea as to when we’re addressing it. What is required, is an acknowledgement that the entire desktop metaphor is holding back computer development.
IBM’s BlueEyes project is attempting to remedy this. It allows machines to track our eyes as we work with them. This is they key concept. If Human Computer Interaction is to develop beyond the point-and-click of today, we need to think about working with the machines, not on them.
A combination of speech recognition, motion tracking and biosensing could allow us to change the computer from a mere tool, to a partner in our work, albeit a dumb one. Indeed IBM is currently researching algorithmic constructions which will allow affect detection from our faces. If pervasive computing is to be more than a concept, we need to kill the mouse.
And I guess that grecoman hasn’t thought of the next big thing either. Or if he has, he’s quietly working on it and not telling us. Do let us know when your product is for sale (or better still, available for download). Till then, please refrain from calling us nitwits and clowns.
Mobile phones are the likely cadidate for the next-generation input device (for the home-wide computer/entertainment system). we all carry them around anyway, they have a small keypad microphone, and 3g would bring things like a small colour-screen, camera etc. Even now they have rudimentary “speech recognition” : voice commands that dial certain phone numbers. It wouldn’t be difficult to add commands for using a browser or for activating your favorite programs.
Oh and on the issue of startrek and keyboards, i seem to remember Scotty seeying a 20th century computer and saying : “Oh it uses a keyboard, how quaint” (ST : The voyage home
Why is it that whenever I read these articles about the “future of interfaces” the author always makes the obligatory reference to the Xerox PARC but never suggests going back to find out what they are up to *now*?! The PARC is still alive and at least 5 years ahead of what we are doing right now. Come on, people…This is a no-brainer!
The desktop metaphor was never anything more than that, a metaphor. Some people did try to implement such a thing literally and oh dear… The desktop metaphor itself is however long dead in its original form, most modern OSs use the ‘desktop’ as a storage area for data and program launchers, and ‘launch bars’ are more common providing an effective method to organise and access programs (even the Windows start menu can be used in this way).
I can agree that it’s not perfect, but a hierachical file store makes sense, add a behind the scenes DB and metadata and enable folders to be tied to views and that would be perfect for most data storage requirements.
Now, the WIMP interface, ie the actual GUI… That will be with us for a long time yet. Why? Because it works well. Sure, there will be small improvements here and there, and stuff like voice recognition will eventually be available for certain categories of task, but day to day interfacing will still use the mouse and keyboard or equivalents (eg touch screens).
This gets trawlled out every now and again, but none of these articles actually propose a workable alternative – and that’s because ther currently isn’t one and probably wont be for a few years at least.
> This gets trawlled out every now and again, but none
> of these articles actually propose a workable
> alternative – and that’s because ther currently isn’t one > and probably wont be for a few years at least.
I actually agree with the above statement. Bear in mind the following points.
1. The article was commissioned. Have Mac, will write for food.
2. The original article was more of a critique of the ‘alternatives’, than of the current GUI. Somewhere along the line that got changed. Hence my critique of 3D FMs and so on as ill-conceived.
3. It was intended to be speculative, because though at the time I was involved in interface design, it was geared more towards web interfaces than HCI. Thus, if someone replaces the GUI, it certainly won’t be me.
4. The article was purposefully provocative.
> Why is it that whenever I read these articles about the
> “future of interfaces” the author always makes the
> obligatory reference to the Xerox PARC
In this case it’s because the target audience is not expected to be au fait with the history of HCI and GUI design.
> but never
> suggests going back to find out what they are up to
> *now*?! The PARC is still alive and at least 5 years
> ahead of what we are doing right now. Come on,
> people…
PARC may well be alive and kicking, but when you’re under a VERY strict word count it’s not possible to properly disucss the alternatives. Hence the article remained firmly critical and speculative and could not make full reference to current developments.
> This is a no-brainer!
Is it?
If major firms like SGI have invested time and money in 3D GUIs, why have yet to see a commercial development? The developers of 3Dsia claim to be creating an intuitive 3D environment. As Gibson was, at the time, not a computer user there is some cogency to this argument, as he was conceptualising something from the perspective of a non-user. However, Gibson’s work is fantastical – he was not intending his computers to seem intuitive, rather it was the obverse which he was interested in. His machines are complex and frightening – extensions of the complex mainframes embedded in our culture at the time.
Microsoft Research has also been working on 3d interfaces, both to replace the desktop and for organizing information (I believe it was the favorites menu in IE). Neither of them have gone beyond usability testing as far as I know, but that may change in the version of Windows after next, or later.
The developers are wrong. There is an innate problem in displaying 3D on a 2D screen. No matter how well visualised, it remains an optical illusion. Furthermore, 3D will entrench the problems of the desktop. I have difficulty locating files in the mess that is my computer, making me look underneath a virtual-sofa isn’t going to help.
I think you’ll find that there are more people working on changing the methods of interacting with the system’s files than on the desktop metaphor. The only unfortunate part is that much of this work is being done assuming the current environment rather than some future 3d interface. What I’m thinking is that, given MS Research’s 3d interface work, the filesystem changes in Windows Longhorn may be helpful in moving the interface away from the desktop metaphor as well (though the desktop becomes less and less the focus of the interface over time).
The combination of the file manager (My Computer or Windows Explorer in Windows, though the only people I know that still use the latter are people that like the Win3.x way of doing things) and the graphics shell are the things that most people associate with their computer. However, if only one changes and not the other, then we simply end up with an extension of the same problems. I think that, perhaps, the file management interface is the best place to start with the changes if one must be done first, but we may get a chance to see for ourselves soon enough.