A new study Windows vs Linux in a business environment

[Asher]

First, let's get this out of the way: Yes, it was funded by Microsoft, but that doesn't necessarily invalidate what it has to say. In fact, the 47 page PDF was written by Herbert H. Thompson, PhD of the independent firm "SecurityInnovation", which has no ties to Microsoft aside from Microsoft contracting them to perform the study.

But much of the PDF contains some facts that highlight why some businesses may be hesitant to switch to Linux. The vast majority of the Linux cheerleaders, both on this site and in general, don't have the insight to see past the trivial "Linux is free, Windows is not" mantra; nevermind the fact that in most cases corporations do purchase Linux much like Windows contracts, and in this case it compares SuSE Linux Enterprise (SLES) 8 and 9 to Windows Server 2003.

The study was done by comparing the two Operating Systems in an e-commerce environment over the period of one year.

Some of the more interesting findings are Linux administrators took 68 percent longer to implement new business requirements than their Windows counterparts, and the "Novell SLES solution experienced 14 critical breakages while the Windows Server solution experienced none." SLES 8/9 also required 4.79 times the number of patches as Windows Server 2003, resulting in more work needed to be done by the Linux administrators.

It's a good read. I'm not expecting any detractors to have anything of substance to reply other than knee-jerk "if MS funded it, it's false" bull.

[Asher]

The conclusions, for the lazy:

Businesses and customers care about solutions being available to them when needed.
They care about reliability in broader terms than kernels and dlls; they care about the pain
caused when a solution fails. In this paper we present a new model for measuring and
evaluating solution reliability, one that looks at a system in deployed environments as
opposed to components in isolation. Any true measure or predictor of downtime must
consider the system as it changes over time which includes updates to maintain security
and functionality, but also the evolving requirements of the enterprise. When considering
the results here, it is important to understand their scope and applicability. This study
pertains to enterprise solutions that have patching needs and changing demands. It is not
appropriate to draw any conclusions for embedded devices or systems that are completely
static and rarely, if ever, require system change.

In the “Experiment” section of this study we considered one evolution scenario: an
ecommerce company that must move its site from basic purchasing to a personalized,
history-driven portal. These requirements were created with the help and input of some of
the world’s largest e-retailers and the experiment follows the life of two systems, one
Windows based and one SuSE Linux-based over a simulated one year period. While the
sample size of administrators was too small to provide conclusive statistical comparisons,
the results highlight some fundamental differences in the Linux and Windows models.

System administrators on Windows followed a fairly predictable and consistent
installation path for both patches and add-on components while Linux administrators all
followed vastly different paths to both add functionality and resolve conflicts. One of the
most striking outcomes of the experiment was the diversity of solutions under Linux.
Multiple pathways can be both an asset and a weakness; it can allow highly skilled
administrators to solve problems using greatly varied approaches but it also leads to the
“personalization” of systems which can make issues like administrator substitution problematic. The Linux solutions also quickly went out of support from the both the
distribution vendor and 3rd party solution vendors as individual components (such as
MySQL) were upgraded to meet 3rd party solution needs.

For the administrator trials, our choice of requirements was based on a common
evolutionary scenario, one vetted and confirmed by analysts and corporations in the
ecommerce space. Our choice of 3rd party components was based on market leadership in
the enterprise markets. The experiment follows only one scenario with a small set of
administrators, but the results highlight some key model differences in Windows and
Linux. We strongly encourage others to repeat this analysis under other requirement sets
and using a variety of 3rd party products. The industry needs metrics on reliability that are
true predictors of organizational pain. Our hope is that the methodology presented here
will serve as a base that others can expand on and adapt to their own business needs.

[Fake Boris]

Errr... I actually remember you saying that we couldn't take into account the opinion of a LinuxWorld editor because he was, well, a Linux guy.

[Asher]

Do you understand the difference been a systematic study and an opinion piece?

[Fake Boris]

IIRC, it was a rebuttal to a "study".

Do you understand the difference between dogma and discussion?

[Asher]

It was an op-ed piece full of rhetoric by the editor in chief of Linux World.

That is in no way comparable to an independent study conducted over a year by a PhD and his team.

[Fake Boris]

Haven't you ever heard of Ph. Ds working for pharmaceuticals, demonstrating beyond any doubt the efficiency of their medication over the others?

[Kuciwalker]

I'd trust those over op-ed pieces too.

[Fake Boris]

Then I guess you have to trust that every medication on the market is superior to the competition. Well done, boy.

[Asher]

Originally posted by Oncle Boris
Haven't you ever heard of Ph. Ds working for pharmaceuticals, demonstrating beyond any doubt the efficiency of their medication over the others?

47 pages of documentation and details back up and detail the process.

It's hugely different than a 1-page op-ed on LinuxWorld.

[Fake Boris]

The guy actually read the study and criticized its methodology. I guess this is something illegal in the Asherverse.

[Asher]

Originally posted by Oncle Boris
The guy actually read the study and criticized its methodology. I guess this is something illegal in the Asherverse.

I think you're confused, this study was released yesterday...

[Fake Boris]

Anyway, I've got no time for this tonight. Back to my filosofy of art essay.

[Asher]

French people have less intelligence - FACT.

[Kuciwalker]

Originally posted by Oncle Boris
Then I guess you have to trust that every medication on the market is superior to the competition. Well done, boy.

No, because there are generally other [b]scientific studies[b] that provide conflicting claims.