Slashdot Mirror
Usability and Open Source Software
Martin Soto writes "This article by two user interaction researchers, discusses many of the usability problems in current open source projects. The nice part is that, unlike many /. readers, it doesn't stop there, but goes into suggesting novel (at least for the OSS community) approaches to cope with those problems in an open source compatible way. Worth a read to those that, like me, still think that OSS should find its way to every desktop computer."
Actually, that would be http://www.whitehouse.gov/, NOT whitehouse.org. Did you do that on purpose?
The roots of education are bitter, but the fruit is sweet.
--Aristotle
Yeah. You didn't even read the article. They defined usability right at the beginning of the document: "Usability is typically described in terms of five characteristics: ease of learning, efficiency of use, memorability, error frequency and severity, and subjective satisfaction (Nielsen, 1993)."
They wrote a well-researched, seriously informative document with tons of examples, quotes from experts, and more. I find their comments far more compelling and substantial than yours.
My Greasemonkey scripts for Digg &
I suspect this is a troll but i'll bite...
Useability is something you add on the backend of a product to market it. What they really mean by useability is a nice GUI where you can get a mouse trail going.
No. You obviously know nothing about the subject. Usability is not something you hack onto the backend of a product after all is said and done. If you're serious about designing an application to be highly usable, then the interface becomes a central development point just as important as the actual functionality (perhaps more important because if you can't figure out how to use it and use it well, what good is it?). And to say that usability means just putting together a whizbang fancy GUI is a massive understatement. There is years of research on the human brain and how to best present information to it on which the principles of good GUI design are built. Sure, it's obvious to you now because you've had lots of experience with applications from which you can borrow concepts from, but at some point in the past someone had to ask the question "what is the best way to do this?"
What most people don't realize is that it's not just the "big" things like provide menus at the top of the screen, etc, but also little things that are taken for granted. You probably don't notice them but when they're not there you miss them. Ever used an application that just didn't "feel" right but you couldn't really explain why?
Useability is great in most OSS work, extremely efficient and powerful...it just has a higher learning curve to the uninitiated user
Here you contradict yourself. How can useability be great if the learning curve for a new user is high? Usability is about reducing that learning curve by making the interface intuitive (among other things).
Usability Testing of Athena User Interface
Voices from the Open Source Revolution
KDE Usability - First Steps
A few of these books grace our desktops here at work.
Next time don't go the rpm route, use Apache Toolbox from http://www.apachetoolbox.com any job is easer when you use the right tools, for compiling Apache "the way you want it (tm)" Apache Toolbox is the perfect tool. A fine example of some one in the OSS community taking a difficult job and making it simpler. Yes some projects are there to make things easer for the less experenced.
Furthermore, most OSS work has outstanding useability, but lacks the bells and whistles of non-OSS work.
Without a doubt. One of the most amazing GUI based programs is The Gimp - with an extraordinarily powerful graphical front end, infinitely customisable and high-powered scripting behind it, it's about the strongest interfaced large OSS app. It's widely criticised by those not in-the-know however, just because it doesn't "work like photoshop". If people would take the time to think a little differently and actually learn the program, they would find it very intuitive.
In my biased opinion, the audio editor Audacity is a success story in OSS usability.
I've been working on this project for almost two years now, and the experience has completely shifted my priorities and my perspective in software development. Before I started working on Audacity, I had the mindset that I think many OSS programmers have of only caring about the capability and raw power of a program. I never really considered the non-programmer users a significant concern.
Audacity's project lead is Dominic Mazzoni, who is uniquely excellent at both programming and user interface design. He comes from a Mac background, a world where interfaces generally don't suck. From day one he was writing for maximum usability and maximum use. Doing simple things with Audacity is child's play. Dialogs and messages are written to be easy to understand. Audacity is portable to Windows/UNIX/MacOS9/MacOSX, so right off the bat the potential audience is much larger than an application written for only one platform.
There is an audacity-help list that is advertised in big letters on the web page. This is an open invitation to ask questions that most would see as newbie questions not worth their time. This gives us a chance to see what users are having a hard time understanding. Most of these questions are answered in a timely fashion, which means these users don't abandon Audacity.
Documentation is another area where Audacity shines. Tony Oetzmann has been writing some really excellent, concise, useful documentation.
As a result this focus on usability, a lot of people use Audacity. We're pretty consistently in the top 20 downloads on sourceforge. People write often to ask if they can incorporate Audacity on CD compilations. We've been reviewed in the Washington Post.
I've really come around on this in the last two years. Usability is worth it. Anyone can appreciate software that is usable, even programmers. This doesn't mean dumbing things down -- right now a feature is in the works that will allow a project to have a speed envelope, that will allow you to have the speed continuously vary (with appropriate resampling). This is a pretty advanced feature that most users would never have a use for. But a lot of thought is going into how to integrate it into the GUI in the best way possible. It's not going to just get bolted on.
You're comparing apples to oranges.
h tml i nal/macos-x-1.html
"The Rhapsody system [precursor to OS X] has been in development since Apple bought Next Computer in December 1996" [1]
"So I started the GNOME project at that point in August 1997" [Miguel De Icaza, 2]
So not only was OS X started well before GNOME, it was based on NextStep, arguable a more solid foundation for a desktop than Linux and X. Of course, Mac OS X was released on March 24th, 2001 [3], nearly 5 years after it was allegedly begun, not "A couple of years."
I don't see why you're so quick to write off OSS.
Erik
[1] http://www.aessf.org/newsletters/may98.pdf
[2] http://primates.ximian.com/~miguel/gnome-history.
[3] http://www.arstechnica.com/reviews/01q2/macos-x-f
Progeny is a small company, debian is a huge organization. The progeny installer might work for some people in some common situations but it doesn't work well for you, so don't persist using it. Debian is easy to install you just need to read the manual. The major difference between Windows and a debian system install is that you need to tell it what you have (hardware) and what you want to install (software). If you don't want to do this try KNOPPIX http://www.knopper.net/ it is the same debian on a bootable CD and it will probably help you get a idea about what you are doing, in a fairly safe environment.
Nooface is where I end up going for UI stuff. They cover anything, but love to dote on 3D UI and alternate input devices. OTOH, they've also given decent pointers to CLI stuff and the down-to-earth stuff from Jef Raskin.
Don't get me wrong GUI's can be useful but I suspect the problems you are having are not because of the X GUI but because you don't understand how to use your system.
The major merit of GUI's is that they allow you to learn to use the system by playing with it. Pull down the menus, push the buttons and see what is there.
The problems arise when you have to do something that the designers never envisaged. This happens after you have learned the basics and actually want to get work done.
The second problem is that the more menu's/buttons you add the less effective this become, to the point where you have to read the help to find out what you want to do.
At this point your GUI has become less useful then a shell tool.
If you can compare DOS to a Bash shells, you should learn a lot more about shell tools, there really is no comparison. Learn the following and many of your shell problems will be resolved.
--help -h will give you a summary of available commands
man will give you a more detailed guide.
info will give you an interlinked guide (you might find this easier then man if you just want to find one thing)
There are also ways to get a list of available man pages but I don't know of the top of my head. I generally use tkman & tkinfo, which gives a window GUI, and some cross-referencing of man pages.
I think the major reason that OS X has managed to come up with a simple GUI quickly is because that is what they were trying to do.
Most Linux users have real work to do, for us functionality and access to information is the most important and deciding factor.
I use Windows at work and have installed cygwin because my time is limited.