The Death of Xenix
As of January 1, 1996 the Santa Cruz Operation (SCO) streamlined its product offerings by dropping a number of older releases from its lineup. Until last December 31, you could still buy OpenServer 3.0 even as version 5.0.2 neared release; not any more.
While few will mourn the passing of most of SCO's 1997 New Year's obsolescence list, I know of at least a few old comrades who will light a virtual candle or two for venerable SCO Xenix, that throwback to the time when Microsoft was a Unix OEM.
Ah, Xenix... My first taste of it was in mid-1985, when it was just about the only thing with genuine Unix code you could get to run on the 286s and 386s of the day. I wasn't at a college where I could play with BSD-filled VAXes, I couldn't afford the 68000-based offerings of the day from Sun and HP, and I didn't feel comfortable with the wide assortment of available Unix clones such as Coherent or QNX or OS9.
In the years that have passed, much has changed; Unix has changed hands many times, DEC and IBM discovered Unix (thus giving us the OSF and the Unix Wars), CPUs have become both more complex and more simple (i.e., RISC), and hardware such as accelerated video and CD-ROMs are commonplace. Until December, though, the one constant was Xenix.
Now before you check the cover of this magazine to wonder why you're reading so much about SCO stuff in a Linux magazine, please understand that the death of Xenix has great significance to the Linux community. It represents a major opportunity for Linux supporters to make inroads into the world of business computing. It presents us with opportunities to not only make money, but also to build the reputation of Linux as a stable platform more than capable of running a business.
While Xenix may be going away, the reason it existed for so long won't. Xenix serves as a workhorse for small businesses—places where you might find a dozen or so dumb terminals and a couple of printers plugged into a system running FoxBase or other small database system. It's simple, small and stable, maintaining such anachronisms in the current Unix market as unlimited-user licenses and command-line administration.
Despite the appeal of GUIs, the simplicity of a command-line interface still holds an attraction for many people running small businesses. While metaphorical desktops are being moved to PCs as the dumb terminals break down, basic database tasks are still handled more than adequately by Xenix running on minimal hardware. Who needs the bloat of the X Window System on a database server anyway?
Most Xenix users will probably continue to use their systems as-is until they break, exemplifying the “if it ain't broke don't fix it” school of computer administration. But when these systems do break, Xenix will no longer be suitable.
For one thing, Xenix doesn't take to new hardware easily. CD-ROMs, PCI, and even mice are difficult or impossible to implement in Xenix; it just barely supports SCSI disks and tapes. Xenix's networking support, while never very functional, no longer exists as an upgrade in any case—someone wanting to add even a single Xenix box to telnet or share services is out of luck.
This is where Linux comes in. It makes a perfect drop-in replacement for Xenix; it also runs lean and acceptably fast even on slow “legacy” hardware, and it supports inexpensive peripherals very well. Furthermore, Xenix adds several new features small businesses now expect from their systems.
Most businesses have probably grown out of the kind of word processors that had been available on Xenix (ASCII-based WordPefect5; Lyrix; even Uniplex—argh!). At the very least, they've become jealous of what they've seen on PCs; even if they love the way their Unix databases are working, most want to move their word processing and spreadsheet applications onto GUI systems.
If you can talk users into considering Linux, running Caldera's Internet Office Suite or Red Hat's Applixware or even StarOffice, so much the better! But even if not, you'll know their new Linux server will be able to run more than just databases. They can use Samba to enable file and printer sharing, and all the tools are there if they want to set up an Intranet or Internet access.
SCO's answer to stranded Xenix users is to offer one of a number of upgrade schemes to OpenServer Release 5:
Xenix to 16-user Host system (no networking): $250
Xenix to 32-user Enterprise system (with networking): $750
Many Xenix users may be uncomfortable upgrading their unlimited-users license to a limited-user OpenServer system, or paying about $4,000 to upgrade to an unlimited-user OpenServer. Anyone wanting to upgrade a Xenix development system will pay an additional $300. Also, OpenServer systems need more RAM and better video than Xenix does to do the same job.
Compare this to the cost of the most expensive Linux distribution. To be sure, there are a few technical quirks going from Xenix to Linux (Linux doesn't have SCO's “custom” installation software, for instance) but these aren't a big problem for Linux installers. Running Xenix binaries has not been a problem for me, at least.
This means that Xenix can be the second big opportunity for Linux systems to make their way into business computing. If the Internet got Linux a foot in the door of corporate computing, upgrading Xenix installations would be like getting in an entire leg and maybe an arm as well.
Of course, all the cost savings in the world won't help if the customer doesn't believe Linux is reliable enough to be as low-maintenance as the Xenix it replaces. There's a common perception within the SCO community (at least, based on the newsgroups) that Linux still isn't robust enough for commercial use. Part of this perception is pure FUD, some is based on experience from people who tried Unix in the BC era (before Caldera), when Linux really wasn't being aimed at the commercial world.
Now that Linux is starting to approach the turf of commercial Unix vendors (not just SCO), the road is going to get rougher. Here's where we leave the realm of computer religion and enter the free market. If something you do or say is going to lose someone a sale and associated commission and profit, they're not going to put up with this quietly. Newsgroup flames are nothing compared to the vulgarities of salesmanship; it's amazing what some people will do to reach their sales quotas.
Until now, putting Linux into a company has rarely been at the direct expense of some other vendor; it's come in as an auxiliary system to do routing, run the Internet connection, or maybe serve as a power user's desktop system. Now with the release of industrial-strength software for Linux such as the ADABAS database, Linux has a legitimate shot at becoming the core for central servers in business. As it succeeds, and as other Unix vendors start to (publicly) take Linux seriously as a competitor, a lot of mud (and FUD) will be flung our way.
The answer to flung mud is not Amiga-style evangelism, but cold hard fact—case studies, comparative analyses and companies such as Caldera and Red Hat and Crynwr prepared to support freeware-based products. The increasing volume of good coverage Linux is getting in the business computing press helps, too.
To be sure, more is needed. The Linux community must agree upon a single software installation and management scheme, just as it standardized file system layouts. As Linux further finds its way into commercial settings, it'll need appropriate features (provided as freeware or commercial software) such as on-line data managers and better RAID facilities. More ISVs need to be encouraged to support Linux; and as they do, they, in turn, need to be supported.
Is the community up to the challenge? I'm betting on it. In the meantime, keep an eye on all those Xenix systems out there. The time will come.
Evan Leibovitch is Senior Analyst for Sound Software of Brampton, Ontario, Canada. He's installed almost every kind of Unix available for Intel systems over the past dozen years, and this year his company became Canada's first Caldera Channel Partner. He can be reached at email@example.com.
|Non-Linux FOSS: libnotify, OS X Style||Jun 18, 2013|
|Containers—Not Virtual Machines—Are the Future Cloud||Jun 17, 2013|
|Lock-Free Multi-Producer Multi-Consumer Queue on Ring Buffer||Jun 12, 2013|
|Weechat, Irssi's Little Brother||Jun 11, 2013|
|One Tail Just Isn't Enough||Jun 07, 2013|
|Introduction to MapReduce with Hadoop on Linux||Jun 05, 2013|
- Containers—Not Virtual Machines—Are the Future Cloud
- Non-Linux FOSS: libnotify, OS X Style
- Linux Systems Administrator
- Lock-Free Multi-Producer Multi-Consumer Queue on Ring Buffer
- Validate an E-Mail Address with PHP, the Right Way
- Technical Support Rep
- Senior Perl Developer
- UX Designer
- Introduction to MapReduce with Hadoop on Linux
Free Webinar: Hadoop
How to Build an Optimal Hadoop Cluster to Store and Maintain Unlimited Amounts of Data Using Microservers
Realizing the promise of Apache® Hadoop® requires the effective deployment of compute, memory, storage and networking to achieve optimal results. With its flexibility and multitude of options, it is easy to over or under provision the server infrastructure, resulting in poor performance and high TCO. Join us for an in depth, technical discussion with industry experts from leading Hadoop and server companies who will provide insights into the key considerations for designing and deploying an optimal Hadoop cluster.
Some of key questions to be discussed are:
- What is the “typical” Hadoop cluster and what should be installed on the different machine types?
- Why should you consider the typical workload patterns when making your hardware decisions?
- Are all microservers created equal for Hadoop deployments?
- How do I plan for expansion if I require more compute, memory, storage or networking?