Top 8 Reasons HCI is in its Stone Age 547
UltimaGuy writes "This Editorial describes 8 reasons why HCI (Human Computer Interaction) is in its stone age. It laments about screen corners, filesystem, GUI Design and also 'spatialness'. "
This discussion has been archived.
No new comments can be posted.
Top 8 Reasons HCI is in its Stone Age
Related Links Top of the: day, week, month.
"And remember: Evil will always prevail, because Good is dumb." -- Spaceballs
And here's the answer of an amarok developer (Score:4, Interesting)
Interesting read, I think.
Not so great? But what about focus-stealing. (Score:4, Interesting)
"Browse the internet by hitting the screen corner? Check mail in the screen corner? Get Info in the screen corner? System preferences in the screen corner? Switching applications in the screen corner?"
The first and most obvious problem with this concept is that the user must know what each corner does. You should not expect the user to remember this by heart. Therefore, you have to either allocate screen real-estate to show it (doh!), or pop up the information about what happens when you move or click here (doh!). If you allocate screen real-estate, then that should be clickable as well. Doesn't sound like such a great idea anymore, does it? If you pop up information, then you just made your interface more annoying because the mouse sometimes tends to end up in the corners by mistake.
"Ray Charles figured that out. Stevie Wonder figured that out. And they would probably make a better design team than any money-driven market thugs."
Gee, which market thug are you thinking of?
I wish Microsoft would fix their most fundamental user interface problem: Never, ever, ever, ever, ever steal my input directed to one window and start providing it to another. I don't care if the applications are not playing ball properly. Don't allow it. How many times have I hit "enter" while typing, say, in a word processor, but just before I hit "enter" a message box pops up and my enter key is swallowed by it, taking the default action, and I don't even know what happened because I never got the chance to see the question. Or my password being entered into one window's field but ending up in another. Bad.
Re:computers: still not for lay people (Score:3, Interesting)
I really fucking hate how every program you install nowadays has some kind of agent running in the background on startup. What's worse is that a lot of new programs make it impossible to disable them.
You know what? I'll decide when I want a certain program running on my computer, thank you very much!
GUI's suck at iteration (Score:2, Interesting)
for file in `find . -name \*.[ch] -print` ; do mv $file
I have yet to see a GUI that allows me to select files in this manner, and perform the same operation on all of them. A large collection of archive files that need to be unpacked is usually quite difficult to do in a timely manner on Windows, or in any KDE or Gnome desktop. Oh, you can use the GUI filemanager in Windows or Unix to find files whose names match a pattern, but how do you apply the same operation to each one? In Windows, you will get a new Winzip window opening for each archive, and you will have to operate the controls for each file: extract, close window; extract, close window; extract, close window - over and over and over. What makes it even worse is all the repetitive mouse movement required to hit all the buttons. The extract dialog pops up over here, the close button is over there... you end up moving your mouse around excessively just to land on the controls. Click click click click click click click. I get sick of it.
Suggesting that the corners of the screen ought to do something is right out of the stone age, too. Stop making me move my mouse cursor all over the place. I hate that.
Um, this guy is off... (Score:5, Interesting)
Examine his (central) point about corners, for example. Yes, corners *can* be hit easily with the mouse. Isn't that a long way to travel to achieve ones goals? His point about scrolling with the spacebar press is on target (and a feature I appreciated), but then he goes on a tangent about the biggest key on the keyboard producing "nothingness". Considering that each and every word must be separated from each and every other word with "nothingness", I fail to see where its place of honor is diminished by the lack of pixels being illuminated by its use.
Crying shame too: usability *is* important and should be a central consideration. Sadly, I don't think this guy is the one to much of that consideration. Maybe once he grasps the utilization of natural language a bit more, I would consider his ideas on more natural interfaces.
Theory, meet reality (Score:1, Interesting)
Look, here's the things with the four corners and the like - they're great but humans need *identifiable* concepts and workable UIs. That is, there are uses for the four corners (and BOTH MS and Apple uses them to some extent) but an argument of "do everything in the four corners" is silly without concrete, "here is how to do it" suggestions that can be refuted or supported. This is just yet another person telling us - hey, usability could be better. I read this theory in this book, and even today's GUIs don't base *their entire GUI* around it. What horrible interfaces!! Please. So how would our blog author design their OS? I'd really like to know. All I could tell from the blog is that it would have a spacial file manager and probably not allow you to do many of the things GUIs today do for you because it would break a theory.
It's all great to wave your theories around, but OS 9 supported many of these things better than many modern OSes do, and that went the way of the dodo, with a few die-hard usability people screaming bloody murder and the rest of us saying "gee, I'm glad I don't have to have 20 windows popped open to get to a folder." OS 9 was great - in theory. In reality, I wouldn't touch it after the move to OS X.
Re:Pet peeves... (Score:3, Interesting)
My biggest complaint about caps lock is that it's very rarely used but is layed out on most keyboards opposite the enter key. Shouldn't we be able to shove caps lock into a deep dark hole on the keyboard and use that space for a key that's used a bit more often (like control)?
Re:computers: still not for lay people (Score:4, Interesting)
I also agree with the blog, too many preferences and too many flashing notification everywhere are very distracting. GNOME I think is on the right track with this, especially in the Ubuntu distro version. Applications are simple and streamlined.
Today most computer users are all tainted by MS Windowz interface, that is what they know and they won't learn anything different even if it means improved usability and efficiency in the future. Therefore there are two philosophies for designing new interfaces:
1) Design what is familiar to the users even if it considered "bad design" according to standards and HCI research
2) Design what is believed to be correct according to HCI research, even at the expense of confusing the Windowz crowd.
It seems that KDE has mostly addopted the first approach and GNOME the second.
An interesting point, in one of the HCI classes I took, we read a paper that compared the command line to the graphical point-n-click interface. It turns out users are slower to learn the commands but once they do they remember them longer. For example it might take a while for my grandpa to learn that 'ls' means 'list the files in the directory' as opposed to just double-clicking the folder. But once he will learn it he will know it for a longer time, as opposed to asking him to open a folder in windowz a week later -- he might try to click on it once, click with a wrong button or try a mouse gesture.
Re:1. Screen Corners (Score:3, Interesting)
My intent was to produce some stats on the very basics of user interfaces so that they could be used to evaluate more complex interfaces. The first test was designed to look at how long it took people to click on something.
I started out fairly basic, just a box with that appeared randomly on the screen, and then moved up to having boxes that appeared in ordered patterns and at given locations on the screen (including points in the corners), the given location tests where mixed up with random locations to make it a little more realistic.
I was looking to measure a learning curve as the user 'learnt' the location of the fixed boxes but didn't get enough data for proper analysis.
For all tests I recorded the time of the mouse click and the location of the mouse click.
I looked at things like the change in time over time, and looked to any patterns that related to the ordering of the boxes etc...
In the end, after a little practice all the results were showing a straight line (least squares fit) [nctm.org] with reasonable t test results for the correlation between the line and the data. Removing points that had exceptionally large times (where the operator had paused) gave an even better fit.
More detailed analysis of the corners of the screen showed that they were no better than anywhere else (for a small box)
So, unless your UI is only made up of points (I never got around to looking at critical sizes of elements) the corners will probably be faster, but if it's made up of anything else then there's no difference.
The tests could have been a bit more scientific, but I was running the project at home with the help of a few friends before looking to take it to a wider audience and expand it further.
Re:computers: still not for lay people (Score:3, Interesting)
For a large percentage of people, that's fine. But for me, I have my task bar at the top. (used macs years ago, switched and that's where I put it. Your program menu bar is up there, why shouldn't the task bar be?)
Anyways since some programs open up windows at the TOP they get covered by the task bar, and I cannot see the top so that I can move, close or mimize them. I am forced to change the size of the task bar to nothing, then make it a single line big again, and then the pesky new window automatically gets moved down!
Why doesnt' that happen in the first place!
Re:computers: still not for lay people (Score:3, Interesting)
What you want is the Canon Cat [wikipedia.org], designed by Jef Raskin.
People, get a grip! (Score:3, Interesting)
From the Stone Age blog post:
``After more than 20 years of research, development and competition in the field of HCI, not one single leading operating system developing company has come up with an OS that utilizes the four corners of the screen.''
That doesn't mean that HCI is in the stone age. It just means the leading OSes have it wrong. The GNOME version I am running uses all 4 corners. I don't use any of the functions from the corners on a regular basis, but that's a different story; they are used, and it's obviously because the GNOME team realized their power.
The Fitt's Corners article writes about this:
``why don't any major Desktop Environments exploit the screen corners?
I have a good reason: it's because they are the easiest spots to hit with the mouse.
Setup your OSX box to trigger Expose when you move the mouse to a corner. Now count how many times during the day you nudge the mouse into the corner and trigger Expose by accident.''
This has nothing to do with screen corners, and everything with mouse gestures. It's the fact that just moving the mouse (without any indication that some action is intended) triggers actions that causes these accidents. This is why I always disable mouse gestures in apps that support them.
From the Stone Age:
``2. OS GUI's are Designed for Beginners.
Ooooh. there's nothing wrong with that, as long as you can grow with your user interface.''
Yes, GUIs are designed to make computers easy for beginners to use. For those who want flexibility, there is the command line, or, if you don't want to leave the GUI world, scripting (think DCOP, AppleScript), augmented with macro recording (think Automator).
What's _really_ wrong with respect to GUIs being for beginners, is that many aren't actually easy for beginners to use. What idiot came up with double-click? Do you have any idea how much trouble this is causing?!
From the Stone Age:
``You have to actually drop focus on what you're looking at and move your eyesight in order to find that tiny little resize button of the window.''
What would you rather have, genius? A 1x1 inch resize widget cluttering up the screen? At least with people I know, resizing isnt a very common operation. If you want to temporary get the current window out of the way and look at another one, just throw the mouse to the dock or taskbar (yep, they're at the edge of the screen in all current GUIs) and click the widget for the window you want to look at.
Perhaps it would be useful to be able to resize a window by holding some key and dragging a corner of it (where the "corner" could be up to 1/4 of the total window size - after all, you need to hold the magic key to activate this mode), but then, holding a key and dragging is something very advanced for many users I know.
Or you could do like a number of advanced GUI users I know, and just partition the screen into non-overlapping frames, put your windows inside these frames, and never have the problem of overlapping windows in the first place!
More insights from the Stone Age:
``Situations like these make me feel sorry for the spacebar. So big and strong... He totally rules over the other keys, and yet all he produces is... nothingness.''
Maybe, just maybe, it's because inserting a space is a very common operation? How usable do you think a keyboard would be if the space bar were as difficult to hit as the 'Q' on a Dvorak keyboard (it's where the 'X' is on QWERTY)? For the same reason, the return key and the backspace are (hopefully) larger than regular keys, but smaller than the space bar.
The Stone Age guy also complains about modern GUIs offer
Re:computers: still not for lay people (Score:2, Interesting)
My point is that I should'nt have to go through that kind of hassle because properly developed applications shouldn't be using agents in the first place.
Try telling Granny what tfswctrl.exe is and what happens if it's disabled.
Stop telling people how they should work (Score:3, Interesting)
People should stop assuming that real life metaphors are a better solution for everyone.
His arguments in favor of the spatial model are fine as long as you assume that everyone is more used to manipulate real life objects in closets, drawers and boxes than they are to manipulate stuff on a computer.
Because of my job (and centers of interest), I spend most of my time manipulating stuff on a computer. As such, I'd rather have my closet present its contents in a list tree than have my computer files presented as a real life metaphor.
Of course, I don't pretend to know what's best for everyone. That's why suggesting that preferences are unnecessary is idiotic.
The only solution that would be acceptable as far as I'm concerned would be "reasonable defaults" that people more familiar with physical objects than stuff on a computer would be able to deal with more easily, preferences out of the way by default, but existing, and let people switch back to the current way of working if they want to do so.
Also, his article is very critical of the way things are done currently, but don't provide much practical solutions, except get rid of preferences, put stuff in the corner, and a couple random specific use cases, so it's essentially pointless.
Re:Disabled users aren't normal users (Score:2, Interesting)
What I was trying to stress on, was that if we think really hard about the needs of the disabled, we find that other problems are relatively easy to solve. We don't seriously design a UI based on what can be good for users, and instead we simply do what we assume will work for everyone.
Like you said, blind users might prefer text only. You are obviously not blind (or you'd have said that already), and you are making assumptions about what blind users will prefer.
Re:computers: still not for lay people (Score:3, Interesting)
Its adding another function to the application that masks the problem and allowing multiple instances that fixes the problem.
If I want to do two separate tasks which, while similar enough to be accomplished by the same program, are otherwise completely different: What sense would it make for me to have to use the same application? The tasks are completely different. Arguably the settings I set for one might not be optimal for the other (and if you have one for each instance, then you might as well have a separate application running). Why should they be forced to share threads and stack and heap space? Why should the crashing of one take down the other?
Modern operating systems already share (readonly) memory between separate processes, so there are no resource savings in creating such a monstrosity.
Re:Not so great? But what about focus-stealing. (Score:3, Interesting)
Re:computers: still not for lay people (Score:3, Interesting)
The original article is yet another whinge without any realistic solutions. There's a great series of demonstrations by 37 Signals where they put their balls on the line by showing how they would make real improvements to an existing scenario. e.g.:
http://37signals.com/better_fedex.php [37signals.com]
They took the Fedex shipping manager screen/process, and redesigned it to make more sense and increase usability.
Be sure to note their lack of weak jokes about aliens or Russians being able to design better GUIs, or the absence of Stevie Wonder mentions.
The parent post at least adds some realistic suggestions or obvious problems. My pet peeve is another window stealing focus when I'm typing elsewhere -- very, very annoying. Google is a bitch like this; if the page is still loading while you're entering a modified/new search, it will overwrite what you've entered with the old query when the page load completes. Ridiculous! How about having some JavaScript that detects existing focus on the field and cancels the other script if a user has already started typing?