More on open source software
Linux server operating system vendors like Red Hat and even Novell could be heading for bumper sales over the next 12 months. Almost 40 percent of organizations are planning on migrating mission-critical workloads to open source software in that time frame.
That's Accenture's line anyway, and perhaps the mega-consulting firm has good reason for saying so: It asked 300 private and public sector organizations with annual revenues in excess of $500 million about their plans for open source software, and announced the results earlier this month. "What we are seeing is a coming of age of open source, " said Paul Daugherty, Accenture's chief technology architect chappie.
What's most interesting is the motivation that these organizations have for moving to open source software. Microsoft in particular has been involved in a long-term campaign to put out the message that proprietary software, although appearing to cost more than free open source software, actually costs less. Something to do with its "comprehensive, integrated, easy-to-use stack of technologies," making the total cost of ownership lower when retraining, administration skills and so-on are all taken into account.
Microsoft might have a point. (Although equally it might not -- one must take many of Microsoft's "studies" with a large pinch of salt.) But actually it turns out that the TCO argument is largely irrelevant, or at least not as relevant as other factors. "Through both our research and our work with clients, we are seeing an increase in demand for open source based on quality, reliability and speed, not just cost savings," said Daugherty. "This is a significant change from just two years ago when uptake was driven mainly by cost savings."
Quality, reliability and speed. Not just cost savings. Now that's really one in the eye for Microsoft and its ilk. My chum who works at Microsoft still believes open source software will never really take off because a bunch of disparate maverick individuals can't possibly develop decent reliable software and fix bugs in a timely fashion. Good software can be developed only by a close-knit and coordinated team of engineers, as far as he is concerned. I suppose if you come from a proprietary software factory, it must be almost impossible to imagine software being built any other way.
Apparently, the overwhelming majority of those responsible for choosing software in large organizations don't agree with him. (And that's quite aside from the fact that open source software like Red Hat Enterprise Linux isn't really developed by disparate mavericks.) Accenture's research found 76 percent of respondents in the United States and United Kingdom cited quality as a key benefit of open source software, while 71 percent cited reliability. And 70 percent said that better security and bug fixing than proprietary software was also an important feature of software produced using the open source model.
As for the TCO argument, the jury, it seems, is still out. Half of Accenture's respondents reckon the open source software has a lower TCO than Microsoft-style proprietary software, and half don't. Fair enough.
So the future looks bright for Linux server OSes, and open source software as well. The only cloud on the horizon is the lack of senior management support for using open source software among the organizations that have looked at it but ultimately chosen not to use it.
That might be because they are crusty old IT codgers who still don't get the open source software model -- much like my friend from Microsoft. Of course there's another possibility that Accenture and open source zealots don't mention: Open source software might not always be the answer. Sometimes good old fashioned proprietary software of the sort peddled by Microsoft, Oracle and other server operating system and application vendors is the best solution for a particular organization's needs. Open source may be great, but let's not forget proprietary software has its merits, too.
Paul Rubens is a journalist based in Marlow on Thames, England. He has been programming, tinkering and generally sitting in front of computer screens since his first encounter with a DEC PDP-11 in 1979.