![]() |
![]() |
| Linux news | Newbie's Linux manual | Linux links | Link us | ||
| The Linux columns | Book reviews | ||
| DistroWatch + TuxReports | October 22, 2002 | |
Why Linux needs to get "back to it's roots"
by , 5 November, 2001
I have been involved with computers in one form or another since I started attending San Diego City College in the Fall of 1980. I have seen many changes since then, but one thing seems to hold constant. The undeniable arrogance of people that make operating systems that think that they know what's best for us. I have never liked that.
Back in the early 80's, PC's were still a new thing, and MS-DOS with 640K of memory was standard on almost all PC's. Hard drives stored little data, and were horribly expensive. As a result, people learned how to make do. And that is a lesson I have never forgotten.
Software makers wrote tight code. They had to. My first PC was an XT-clone with the awesome speed of 10MHz in turbo mode. It had dual 360K 5 ¼" floppies. No hard drive, and an amber monitor and a Hercules graphics card that, back then, was very nice.
One of the primary programs I used was a word processer called WordStar. WordStar was at the time, king of the hill. The program came on a single 360K floppy that also had the OS installed. MS-DOS so you could boot it, the WordStar program itself, and 100K of space left over to store data if you wished.
WordStar did what a lot of software companies at the time did. They got arrogant. They thought that they were too big a company to have to worry about things so they didn't work hard enough on improving their program. WordPerfect came along, and WordStar woke-up too late. And now they are in the history books.
Lotus 1-2-3 was the spreadsheet of choice for MS-DOS; and dBASE (by Ashton-Tate) was the database of choice. Originally, both dBASE and Lotus were copy-protected programs. Lotus woke-up to the reality that consumers simply did not want that. The idea of paying good money for a program, only to have to use a key-disk to start it up, even after you had installed it on your hard drive, just didn't sit well with the customer. Lotus got rid of it. Lotus survived. dBASE 3+ did not get rid of the copy-protection till much later. By the time they did, it was too little, too late. People had already found other databases to do the work. Programs like dBXL (a 3+ clone from WordTech) and Foxpro were not copy-protected, were compatiable with the dBASE 3+ files, and cheaper to boot. Ashton-Tate went out of business long ago.
Microsoft has a long history of arrogance to draw on as well. Back in the days of MS-DOS, it was said that the new version, "Wasn't done till Lotus won't run."
Microsoft came out with Windows. It was pretty bad. But one thing Microsoft has always had in abudance was a truly excellent marketing campaign. They managed to convince people that they needed Windows, and convince software makers that they needed to write applications for Windows.
I ran OS/2 for a while. I had a copy of Windows 3.1, and OS/2 ran the Windows code in a "box" that would not allow the Windows apps to bring my system crashing to it's knees. I thought I'd died and gone to heaven.
Microsoft, being Microsoft, came out with Windows 3.11, supposedly to add some networking improvements, but in reality what they did was find a way to break OS/2.
Microsoft of course, denied that claim. But history, and the United States Dept. of Justice, proved that Microsoft did this kind of stuff all the time, and to many companies.
Now they are telling people that they are no longer going to support the older versions of their operating system. I run Windows 95b on my laptop. For me, it meets my needs and I see no reason to put money into Microsoft's pocket for an upgrade.
There are many other people out there in a similar situation. Individuals, businesses, all of them deciding that they are content with what they have. Or that they cannot afford the price of the upgrade, when the upgrade to Windows XP requires 128MB of RAM, 1.5GB of hard drive space, and a 300MHz processor.
I worked for a mid-sized company here in Southern Indiana. Kind of typical I guess. About 1,300 employees, scattered around the country in branch offices ranging from Kansas City, to Florida, to California, and a few other places I can't recall off-hand. Corporate offices were in Louisville, Kentucky.
They ran NT 4.x as their network operating system. A workstation had a 166MHz Pentium, a 200MHz if you were lucky, and 32MB of RAM, running Windows 95.
Now, let's see. If you take the 1,000+ PC's on that network, add up the costs to upgrade them to the minimum requirements of Windows XP, and throw into the mix that Microsoft is no longer going to offer support for either their workstation OS (Windows 95) or for their network OS (NT), why in God's name should they spend that kind of money to run a Microsoft operating system anymore?
Why indeed. I know for a fact, they already have at least one Linux server. Won't be much longer till that Linux server has company I'm thinking.
The Gartner Group, not known for their even-handedness in reporting, has issued an advisory not to run Microsoft's IIS Web server. Too many security breaches apparently. I've not seen the report myself, but the Gartner Group is well-respected by many people in the business world and things they say carry a lot of weight.
So how does Linux stack up then? Well, on the one hand, it's Open Source, and that is a very good thing. Proprietary operating systems are not. On the other hand, Linux still has some work to be done before it's ready for the desktop.
Part of the problem is that Linux, in general terms, has gotten the "Microsoft disease" and become bloatware to the max. How many of the major Linux distributions--Red Hat, SuSE, Slackware, Debian, et.al.--can be installed and run on systems like described above? None really.
A typical default install for a Linux distro includes such things as Apache, an FTP server, God only knows how many text editors, both the KDE and GNOME environments, plus a whole gang of other Window Managers into the bargain. Even if you know what you're doing and go in to try and customize it, all too often the install routine won't let you. It's either 2GB of things you don't even know are installed, or a bare-bones install that doesn't have what you want and need. There has to be a better way.
Many Linux advocates suffer every bit as much from OS arrogance as the people at Microsoft. "Linux is the only way, and you're a fool to think otherwise," is not exactly conducive to getting people to take Linux seriously, is it?
Instead, what we, the Linux Community, should be doing, is finding ways to include people that run other operating systems. We should be showing them how to install Linux, in the least painful way possible, and showing them the ways that they don't have to give up all they've learned. We need to shorten the learning curve for people to productively use Linux, and that is something that needs a lot of work, believe me.
I run Windows 95 and Vector Linux 1.8 dual-boot on my laptop right now. Vector Linux is not a major Linux distribution, but they have done a lot of things right and I think it's worthy of consideration. I'll be doing a full review in another column, with just a few highlights to get your interest for now...
Install routines need some work but I think Vector Linux is on the right track, and definitely worthy of consideration. "Default install" of less than 300MB. Not 500MB or 700MB. 300MB. The CD includes both GNOME and KDE, plus the usual assortment of goodies such as office applications and so forth.
I've already been promised a review copy of the new version when it's ready. You can be sure you'll read about it here.
| About us | |
| Latest stable kernel: 2.4.19 | Latest development kernel: 2.5.44 Copyright © 1998-2002 Linuxdot.org. Linux ® is a registered trademark of Linus Torvalds. |
|