So you've done the maths and decided there may be a good business case for Linux after all. You certainly aren't alone in taking what, just a few years ago, was seen as the risk management equivalent of Bungy jumping. Just make sure you don't dive into the world of open source without fastening the rope securely to the bridge.
 |
||||
|
|
|
||
|
Contents 
Introduction
From zero to support team
Closing the skills gap
The personal approach
And then there's the desktop
A strategy for open source |
|
||
|
|
||||
|
|
||||
With constant cost pressures on the lower end of the market, it's not surprising Mackenzie explored potential uses for Linux. Years ago, the team built a Linux-based mail server it used at some client sites for several years, until driver incompatibilities and complex support pushed it back towards a Microsoft Windows Server and Exchange messaging solution. "The maintenance [effort] was very high," says Mackenzie. "We needed to have a specialist come make any changes for us. He was onsite, but not always available."
Years later, Mackenzie came across a local startup company selling packaged Linux servers that bundled remote Windows desktop support, intrusion detection, and a host of other open source-based features into a remote office server package appealing enough for EW to agree to install these systems at several client sites.
Despite the bells and whistles it offered, stubborn incompatibilities remained. For example, the servers' virtual private network (VPN) client refused to connect with VPN servers in other offices to carry critical application traffic. And once this issue was resolved, a new problem with the Linux-based intrusion detection system arose. At another customer site, a fresh Linux installation in thin client configuration was producing continual application freezes.
While he was originally optimistic about Linux's possibilities, the problems that have plagued Mackenzie's clients have driven him away from the operating system and its related applications. "I originally advised clients not to put all their eggs in one basket [by installing an all-Microsoft solution]," he recalls. "But we've had nothing but trouble with the machines."
"It really comes down to support. Our provider was trying to help us, but eventually gave up and didn't respond to our enquiries anymore. All of our people are certified in Microsoft and Cisco products and have first and second degrees in IT -- they're no schmucks. But we've hit this brick wall in terms of Linux, and we'll never sell another one of these boxes again."
Linux, along with other open source applications such as OpenOffice, Plone, Mono, and the ubiquitous MySQL, Computer Associates (CA) Ingres and IBM Cloudscape databases, has become an increasingly common fixture within the corporate environment. Yet even the strongest technical pedigree is no guarantee of market success. No matter how many companies are happily using Linux, the persistence of horror stories such as Mackenzie's reflects both the extent of the image war Linux is still fighting, and the critical importance of providing accessible, effective support where there has often been little.
I think this was a pretty negatively-focused article (e.g., starting with a case study of one person's bad experience). It discusses the issue of what are the pitfalls are of linux again and again yet provides little clue as to why people are interested in linux, i.e., the positives in using it. Successes are barely mentioned. What about the horror stories using windows that cause people to leave windows for linux? That is just as relevant to the author's story. And I have heard many horror stories of Microsoft's (and other propietary, non-open source vendor) support. Apache (open-source) is the most widely used web server software out there.