In an economic slowdown that’s getting
worse, open source matters, now more than ever. Only it’s not
in the way you think. Downloading “free” software has
never been all that great a path to budget savings for most
companies, given all the hidden costs that came with it.
What open source has long promised, and is only starting to deliver
consistently to business, is an alternative both cheaper and more
effective than proprietary code. Open source code still can’t
touch the scope of proprietary suites, but it’s closing the
technology gap. And being lightweight can be its own advantage,
rather than throwing ever-more software and servers at a problem.
With its transparent and standards-based development, open source
code can cut the complexity and risk of custom coding for
integration or niche needs.
And those hidden costs? The time it takes to vet a new piece of
open source code, to nurture in-house experts who can test and
integrate it, to negotiate solid technical support? It’s
getting easier for companies to get through all that, as open
source code is no longer relegated to the fringe of the data
center, running a few stray Web applications. Over just the last
two years, open source has gone from illegal alien to full-blooded
IT citizen. Think of the barriers to open source use that have
faded in just the past year or two.
In intellectual property, all the noise about the illegitimacy of
open source, how Microsoft found its code in Linux and other open
source projects, has withered. Microsoft partnered with Novell, and
skilled programmers within Microsoft’s own ranks pointed out
how much open source code is developed on Windows, and how much
open source code interoperates with Windows. And everyone saw how,
if Windows Server 2008 is to have a place in the data center, it
will have to work with open source code. Proprietary software
companies from Citrix to IBM to Sun proved their faith in
community-built code by buying open source vendors.
In the courts, open source licenses got a boost this year when the
U.S. Court of Appeals overturned a District Court ruling that had
cast doubt on how enforceable open source licenses are. In Jacobsen
v. Katzer, the court showed an understanding of open source code as
a new form of intellectual property, one deserving of protection
through the force of copyright.
In security, open source has been dogged by the notion that it
didn’t pay attention to exposures and vulnerabilities.
Through a Homeland Security grant, Coverity, which sells software
for spotting security flaws in code, tracked 55 million lines of
code in 250 popular open source offerings over three years for
security exposures. Apache, Linux, Samba, and many other of the
best-known projects showed their defect counts declining rapidly
and security standing improving. The Coverity counts finally let
open source quantify its security compared with commercial
code.
These developments make open source code a safer choice at a time
when IT managers are looking for safe harbors.
Perhaps most important of all, it’s not just the old standbys
of open source driving this change. New open source products are
competing directly with commercial code for such deep
infrastructure roles as message routing and management
(MuleSource’s Mule and Apache’s ServiceMix enterprise
service buses), data integration (Jitterbit), application
integration (Talend), and middleware (JBoss, Geronimo, GlassFish,
SpringSource dmServer). There are even new models emerging, such as
“aggregators” that combine the code from more than one
open source project—such as content management, portal, and
business intelligence—to create a new product.
As these risks and barriers fade, instead of being on the margin of
the data center, open source is assuming tasks close to the heart
of production systems. To see how open source is weaving its way
into business infrastructure, look at Continental Airlines.
Next Page :
l Page 1 l l
l