Introduction
Get the facts. That has been the catchphrase employed by Microsoft in their 18-month campaign aimed at encouraging companies to deploy Microsoft Server rather than Linux or *NIX offerings. Using a mix of white papers, studies, reports, and online marketing, the Redmond giant has flooded the new world with reasons why Windows Server is a better proposition for SME and Enterprise companies than Linux.
According to Microsoft spokespeople the campaign has worked: “More and more independent analysts and leading companies find that Windows Server System outperforms Linux on TCO, reliability, security, and indemnification.” However the latest figures from Netcraft indicate that since the campaign started in January 2004, Microsoft has not gained any significant increase in market share – and compared to the same report in January 2002, you can argue that it has actually lost ground.
http://news.netcraft.com/archives/2005/08/01/web_server_survey_turns_10_finds_70_million_sites.html
So, who is right here? Has Microsoft’s campaign misfired? Do ‘the facts’ correspond with reality? Well, as usual in such matters, who can we trust to tell us the truth?
Where did Microsoft ‘Get The Facts’?
Let’s start by looking at where Microsoft’s facts come from. For this campaign, Microsoft wanted to avoid producing their own documentation for fear of being too closely associated with the findings, and also to prevent them from being greeted as press releases from Microsoft usually are. Instead they commissioned research companies to perform the studies for them. Whilst this sounds like a perfectly valid way of going about the job, unfortunately the studies were written so that they conformed to guidelines set by Microsoft. Clearly this undermines their validity somewhat.
Notwithstanding this, you can read ‘solid’ research that supports Microsoft’s arguments written by The Yankee Group, Forrester and Gartner - all organisations that seek to deliver authoritative, innovative and flexible research products and consulting services. The most prominent amongst them is the Yankee Group report. This is mainly due to the fact it was written by one Laura DiDio, a now well-known columnist and industry commentator.
Wait a minute, I’ve heard of her…
Most people interested in this argument have. Laura, after all was of the few industry analysts that SCO showed the alleged infringing Linux kernel code to. Her comments in this seminal case were as follows: "…some lines of code in Linux are the same as those in Unix, which SCO controls, even down to the wording in explanatory comments made by the programmers…” These words provoked a major backlash against her in the open source community forums, and greater concentration on her output. Consequently, since the Yankee Group started working for Microsoft’s PR machine, DiDio’s reports and studies have been increasingly at the centre of the MS vs Linux debate.
What did The Yankee Group report say?
The initial report by the Yankee Group concentrated on key sections such as Security, Vulnerability, Patch Management, Availability and Distribution of Security Information, Performance and Reliability, Interoperability, Channel and Partner Opportunities, Distribution Fragmentation, Patents and Indemnification and Total Cost of Ownership. A fairly exhaustive list you might imagine. Further reports by Forrester, Gartner and Veritest have also concentrated on TCO, Security and Reliability.
In every one of these reports, the ‘Get The Facts’ campaign shows that Windows Server is a better choice for IT professionals than Linux.
Windows is more secure than Linux?
Yes, evidently this particular chestnut is one of the key reasons for staying with a Microsoft distribution. The Yankee report claims that although Windows currently has more security issues, Linux is catching up, and by 2006 there will be a greater array of worms, viruses and vulnerabilities affecting the Linux platform compared with Windows. Others also argue that because Windows is by far the more common of the two platforms in the business OS sector, its flaws are bound to get more coverage. Forrester in its turn, contends that Microsoft flaws are less severe, less numerous and fixed faster.
From the other side of the fence, the Linux community, proud of the security record held by their favourite operating system, has fought back. Nicholas Petreley, in particular, wrote an exposé on the security differences between Windows Server 2003 and RH Linux Enterprise, and posted it on Register.com. This is well worth a read through. The gist of Petreley’s argument is that although Linux may now have as many (or more) security alerts than Windows Server, Windows as an OS incorporates many more critical problems that still stem from it desktop origins.
http://www.theregister.co.uk/security/security_report_windows_vs_linux/
Even if, for the sake of argument, you concede that Windows and Linux are now roughly at parity when it comes to the integrity of their security features (a debate which one imagines will rage on and on), the real hot potato from which the Microsoft ‘Get The Facts’ campaign has drawn much of its ammunition is the FUD factor (Fear, Uncertainty and Doubt) revolving around TCO (Total Cost of Ownership).
Total Cost of Ownership
For many businesses, regardless of size or traditional OS preference, TCO is the crucial, single most influential factor that determines which platform is adopted (how many times for instance have you had a project or proposal turned down because of cost or budget issues?). Microsoft sells proprietary software requiring regular upgrades, and for this reason most people think of it as a costly (but necessary) option. But does this automatically mean a clear victory for Linux in the TCO stakes?
Well, as with the security debate, this one has legs too. For her part, analyst Laura DiDio has fuelled proceedings with a study entitled ‘Large Enterprises: Switching from Windows to Linux’ in which she describes such a move as "prohibitively expensive, extremely complex…(and providing) no tangible business gains."
In her study, she lays out the principal reasons for this conclusion:
• Fewer off-the-shelf applications under Linux
• Paucity of skilled administrators under Linux
• A 20 to 30 percent salary premium for skilled administrators
• Open source software increases liability and exposure in data sensitive networks
• Limited and conditional product warranties and indemnification
Now, anyone with a background in Linux will happily tear holes in each of these arguments. And they have. Incensed by what they see as problematic methodology in the writing of the study, Novell created their own website to deal with the issues raised by the Microsoft campaign. On the site, Novell explain exactly why the reports and assumptions are flawed and how the TCO of Linux in the short-, mid- and long-term is far less than running a Windows Server system.
http://www.novell.com/linux/truth/better_choice.html?tab=tco
TechNet also recently posted an article on how TCO can be lower for many businesses by switching to Linux. “Companies with at least 2,000 employees can reduce their total cost of ownership (TCO) by as much as 26 percent over three years by using Linux servers over Windows” contend TechNet.
Industry analyst Steve Hamm went further, claiming “when I confronted DiDio with [my] conclusions…and told her I was sceptical of her results, she explained that the survey really wasn’t as she had described at all. It was done by a third-party survey outfit who solicited responses from a pool of qualified business computer users.”
So where does this leave us in the Linux vs Windows Server debate? Or, put another way, who are we supposed to believe?
Conclusion
On the one hand it is clear Microsoft are not helping themselves by commissioning studies that are ultimately neither unbiased nor independent. This must be particularly irksome for those at the company who have worked so hard to improve Windows’ credibility as a secure environment in which businesses can function efficiently. For many businesses it remains the most expedient choice without a doubt.
No-one disputes that Microsoft retains the right, as all companies do, to promote areas of their product that they feel are stronger than the competition. In the future, it might be better practice for them to include more genuinely independent analysis in their studies. The strong Linux community will at least be able to maintain that they had no part in the survey.
One thing the debate certainly highlights is the difficulty Linux will have as an Operating System in achieving take-up on the scale Microsoft currently enjoy. Without co-ordinated efforts, they may perpetually struggle against the marketing budget of their chief rival, and dispelling the ‘myths’ produced takes time and resources.
Maybe there is a case for an independent not-for-profit organisation to develop and produce standardised benchmarks that fairly review comparative hardware and software issues. But can any organisation be truly independent? And would Microsoft allow their products to be tested?
At present it seems that the truth may be out there, but you just need to be careful who you ask.
_________________________________________
|
Title
Click here to print our PDF pricelist.