Top White Papers
Editor's Note: This Whole Usability ThingApr 30, 2004, 23:30 (3 Talkback[s])
(Other stories by Brian Proffitt)
By Brian Proffitt
There's been a lot of usability, GUI, how-to-make-Linux better articles out of late, most of which have presented--at the very least--interesting ideas about what's "missing" in Linux.
Since I use Linux for 99% of my daily computer activities, I have to admit that I don't find much missing from it, but I also know that I am a first-class geek who thinks nothing of tweaking and twiddling my computer to get what I want.
One of the problems I have had this week is with spamassassin. Earlier this month, to jump away from Red Hat 9, I ended up settling on Fedora Core 1. With that yum-based upgrade came a new version of spamassassin and there's where the trouble began. The new version still worked, but I could no longer train it to recognize what was spam and what was ham. So many pieces of spam went uncaught and started trashing my inbox.
Finally, I located the solution ("sa-learn --import"), which upgraded the database spamassassin uses to store all of this nifty spam/ham data, which in turn allowed the sa-learn function to work again. Of course, as soon as that happened, suddenly all my mail was spam, which means I have spent the remaining four days of the week endlessly re-training spamassassin.
Now, at this point, I could possibly start talking about this as an example of what's wrong with Linux as an operating system, that its too user-unfriendly, blah, blah, yadda, yadda. Many articles have taken this approach, including some I have written.
Today, I say forget it--I like training my spam software. It gives me an enormous sense of control on what is and what isn't coming onto my machine (I know, it's on my machine, but I am a big believer in out-of-sight, out-of-my-reality. Rob Enderle? Never met him, must not exist. It's a fun mindset, you should try it sometime.)
Is spamassassin the end-all be-all of anti-spam software? No, of course not. I would have liked that database to have been auto-updated when I upgraded the software. But I am very much in favor of people taking real proactive action every once in a while to manage their systems. Microsoft Windows users, as a rule, don't do this, and they end up paying the price. Microsoft has led them to believe that all is well, everything's under control, while spyware and trojans march merrily through the front door of Outlook.
Linux users, by our very nature, tend to be far more territorial about our machines, so we tend to watch the doors more closely.
Many people ask me what kind of software I use at home and at work. "Linux or whatever I need," I typically reply. While I really like Linux, the truth is, I see it and all the other operating systems as tools. They help me get my job done. So, while my left eye may start twitching when I sit down in front of a Microsoft box, I'll still do it if that's what's available to get a job done.
I think this tool mentality is a good one, because it brings to mind something a lot of computer users (maybe even some Linux users, too) can forget: like any tool, you have to take care of it.
My father-in-law is an Ohio farmer who has a vast collection of old woodworking tools that my wife hopes she will someday be given when her dad is too old or too busy in retirement to use them anymore. She loves to do woodworking, and I have visions of our garage being slowly converted to a New Hoosier Workshop in years to come.
These tools are in great shape even today, because my father-in-law takes the time to sharpen them and oil them and keep them in good condition. I believe that we should all be taking the time to do similar things with our computers. Keep things patched. Tweak firewall configuration settings. Lock down unused services and ports.
The more you maintain a tool, the more you will know how to use it.
So yes, let's get that Grand Usability Theory going--but don't make us too complacent.
0 Talkback[s] (click to add your comment)