I picked this story up via Matt Asay and it pretty much summed up the frustrations I’ve had in the last 10 years when talking to certain people about open source - particularly when I was involved in business software. Peter Gyorgy, CIO of GE made this comment in a recent panel discussion:
“I think open source is great for own internal playground type of things but if it’s running vital mission critical applications - networks running on open source for example - then that is a huge, huge risk to the organisation,”
This would be incredibly funny if it wasn’t so damn indicative of so many CIOs, managers and other closed-minded, overly conservative IT people who have long since given up on trying to stay reliably informed and just believe what their vendors tell them. It’s especially amusing given that GE’s healthcare division runs its mission critical software on Linux, which their CIO seems oblivious of. And I would expect the New York Stock Exchange would be considered ‘mission critical’, and it runs on an open source platform (and interestingly the LSE is switching to Linux too) - so clearly not everyone thinks like this.
The one place where he does have a potential point, albeit skewed beyond all recognition is when he says:
“We are not here to be an IT shop, we are here to be the partner of a business and we shouldn’t put businesses operations into risk by running very low cost solutions,”
That’s a very valid point. However, it’s got nothing whatsoever to do with the choice between open source and anything else! This is such a common misconception. Open source has matured - if you need enterprise-level open source there are companies that are quite happy to take your money to remove the hassle and worry of system stability. They’re really no different from the Microsofts and Oracles of this world, except that the software they’re running your system for you on is open source rather than closed. That gives you an additional bit of leverage because if they suck, or if they try to pull a fast one on prices, you can actually get that enterprise management from someone else without having to change your software too. Try doing that when you switch from Oracle to Microsoft or vice versa for services.
You also have the advantage of not having to wait for a central vendor to hear your pleas for feature A or B, or a bugfix that might be low priority for most people but is absolutely critical for you. Instead of hacking workarounds and seething in the wings while you wait for your vendor to get around to addressing something they think isn’t a priority because it’s not affecting that many people, you can pay someone to fix it for you and submit it upstream, where it will undoubtedly get accepted far quicker than it would have got fixed if only a central vendor was looking after it.
You don’t have to be running an IT shop - although if you do, you have the option of trading your own time for monetary savings and greater agility - your support options are just different. Sure, they can be slightly more complicated if you let them be - particularly if you’re looking to save money or drive things in an optimal direction for your company such as tailoring the software - but if you want to have a simple 1-vendor setup using only standard versions you can do that with open source too. Delegating all support to a third party will cost you more but the option is there. It’s all about choice, flexibility, and empowerment - all things a CIO should welcome, not be afraid of (otherwise he/she’s probably in the wrong job).
I think too many IT managers / CIOs have a mental block which prevents them from being really committed to optimising their IT delivery, in terms of both spend, alignment with the business and agility for the future, because they’re locked inside a box of their own making.