Plenty of CIO readers responded to site editor Karen Guglielmo's Editor's Note regarding Linux. Their responses provided gave a real insight into the promise -- and challenge -- of Linux. Some readers are doing some pretty amazing things with Linux. I've taken a look at some of the more prevalent themes here, and highlighted the chief Linux concerns -- and successes -- belonging to SearchCIO.com readers.
Linux: No longer new
Several responses were from readers who have been using Linux for a while -- in one case, since 1995(!). The reason for moving to Linux was consistent among readers. They all wanted a reliable, stable, inexpensive platform.
One reader noted that his company replaced Microsoft NT 4.0 and Exchange Server because it required a reboot at least once a week and caused downtime in a critical business application. The company now runs Linux on a PIII 500 MHz machine, which has not been rebooted in three years!
To be fair, NT's reliability has come a long way since 4.0. XP is pretty robust these days, although I think it needs much more than a PIII to run effectively.
Another respondent was leaning toward Linux for budgetary reasons. Currently running a proprietary OS, his group's marching orders are to cut costs ASAP -- so it is looking at moving to Linux on x86 architecture.
So, for this camp, Linux is not so much a hot new technology but rather a fundamental part of the infrastructure. However, not
Linux: Where are the apps?Several of the people who responded to the SearchCIO.com Editor's Note were running mixed Windows and Linux environments. Their primary reason was that necessary business applications are either not available on Linux, or their availability trails the Windows release by many months.
In either case, this forces their organizations to maintain a mixed OS infrastructure, despite higher administrative costs. The common strategy in this situation is to migrate all infrastructure apps to Linux and host only those apps on Windows that are not available on Linux.
This strategy takes advantage of Linux's lower costs and greater reliability, while ensuring that critical business processes are still available for use. None of the respondents, however, were particularly happy with this situation and looked forward to a time when a single OS infrastructure is possible.
I think these responses highlight a real problem for Linux adoption. It's tempting to think that because the big enterprise applications (Oracle and SAP, for example) have been ported to Linux, every company can shift to an all-Linux infrastructure. Unfortunately, that's not the case. There are thousands of applications -- particularly verticals -- that are not yet available on Linux.
Companies that rely on these applications will be forced to maintain a mixed environment for the foreseeable future. Be sure to factor this into your IT planning.
Linux: Where are the people?
Another thing to factor into your planning is making sure you have Linux-trained people. If you move to a new OS environment, you need to be sure your staff is ready to implement and maintain Linux.
To some, the decision to move to Linux is a risk because there are no funds to pay for training. Without a training budget, employees may not be ready to use and support the new OS. In this view, formal training is a prerequisite for successful implementation. Without training, the entire initiative is at risk.
Others feel that informal training sources are sufficient for the staff members they employ. The availability of mailing lists, forums, Web-based tutorials and downloadable scripts makes it easy for untrained staff to get up to speed on Linux. This contingent counsels full speed ahead, believing there's nothing like real-world experience to implant learning.
I'm not sure where I stand on this issue. I've seen major IT initiatives fail because inexperienced employees attempt to take on tasks they're not ready for. On the other hand, waiting for formal training may consign the organization to never moving forward with a Linux strategy.
Before you decide to put your strategy on hold, here are two suggestions. First, canvass your employees as to their Linux experience. I think you'll be surprised how many folks have Linux up and running at home, and are very familiar with it. Second, use a low-risk entry strategy. Pilot one or two applications that are not mission critical. Learn from them before tackling a really important system. This approach will make your Linux migration much smoother.
So, is Linux hot or not? Overall, the responses to the Editor's Note demonstrate that, hot or not, Linux is here to stay. Challenges remain, but failing to plan for a Linux-inclusive future is doing your organization a disservice.
Bernard Golden is CEO of Navica Inc., a systems integrator based in San Carlos, Calif. He is the author of Succeeding with Open Source and the creator of the Open Source Maturity Model, a formalized method of locating, assessing and implementing open source software.