.comment: What Are We Shooting For, Anyway? - page 3
The Surprise of Linux
While we're all of us happy that Linux does its miracles on the S/390, and while we have all delighted in statistics that tell us that anybody with more brains than an annelid runs his website on Linux to the dismay of the annelids courted by Microsoft, those of us who do not have S/390s or run websites as our chief computing activity are missing the point when we're cheerleading.
The point is that Linux is moving away from the desktop, and unless a lot of us raise holy hell bigtime right now, the advances of Linux mean nothing to desktop users.
I have a friend in Australia known as Skippy, a Linux user, and a fine fellow he is. He bought a USB digital camera not long ago, and made some images he hoped to publish to the Web. He has spent a lot of time trying to make it work. He's not especially interested in devoting drive space and money to making a Windows partition so he can use the provided software to download his pictures.
I monitor mailing lists and one of the continuing concerns is the ability to import and export documents in Microsoft Word format. Microsoft makes filters difficult, in no small measure because they keep moving the target. This is not because, we in our snottiness believe, they wish to prevent our development of filters. It's because if they don't change file formats, they can't sell upgrades -- there's not much you can do to upgrade a word processor past a certain point that Word achieved at about 2.0. The issue here is that there's nothing to stop them.
I remember the OS/2 experience, when it was assumed that hardware providers would sooner or later cough up drivers. They mostly didn't.
Discussions of Linux, now that it has achieved status such that it is discussed in the broader computer world, has often involved the phrase ``critical mass.'' This means several things. The first is that it is a noticeable player. The second is that it has made inroads somewhere.
The third, and the one that we have not achieved, is that anyone who does not open the specs so that Linux support is possible will lose money as a result. Critical mass has been reached in some places -- the S/390 -- but not in others -- Skippy's camera. And the fact that the S/390 got covered before a USB camera did tells us a lot. Chiefly that desktop users have not achieved critical mass, even if Web servers and S/390s have. And the direction of distributors tells us that there will be little help from that quarter. The money, ther probably rightly conclude, is in stuff that users don't directly see, such as servers, or the guts of PDAs, or toasters and washing machines, whose manufacturers can save millions of dollars in the Microsoft tax by using Linux instead. You'll be able to tell if you get a floppy full of source code with your new waffle iron.
Desktop users are pretty much on their own. Our arrogance can easily come back to haunt us, unless we cook up a way to make it apparent to hardware makers that we exist. Unless we impress upon Microsoft that it doesn't any longer dictate standards.
How do we do this? I don't know, but it seems as if we can't count on distributors to make it happen.
We can do the usual stuff, letting hardware vendors know that if they don't make their specs available for development of supporting software in Linux, they'll not get our custom. We can also throw installfests and the like. We can work like crazy on great apps that will lure users to Linux. We can certainly support people who produce Linux desktop apps. We cannot afford to turn up our noses at software that fails to achieve some political seal of approval, because that seal of approval doesn't care if Linux runs on one machine or a million or none. And until software vendors start to think there's some point in developing desktop applications for Linux, critical mass on the desktop will not have been achieved.
Linux is already a huge success story and promises to become more of one. But that doesn't mean that it will ever be a successful desktop operating system. There's still lots more pushing to be done before that goal is achieved. And it's a goal that the distributions are rapidly abandoning.