I just read an interesting if short blog from Ann All about Linux in the SMB market
. Basically, the question is: How far has Linux penetrated into small and medium sized businesses?
That’s a great question.
I’m a big fan that, in most situations, the best product is the most supportable product. Now, with a platform choice like the one people make between Linux and Windows, there are a lot of variables that go into “supportable”, including:
* Knowledge of the OS.
Easy enough. Do you have the knowledge, or can you find someone that has the knowledge, to manage the servers. Far too often I see people with Windows and Linux servers that are horribly configured and frighteningly insecure.
* Vendor support.
Does the vendor actively support the product at a reasonable cost? This one bullet item could start a flame war, but I have to say that I don’t think either side is better than the other on this. At the end of the day, most SMB-level organizations have to pay a vendor for post-installation support.
* Community support.
Again, another flame war possibility here, but in my opinion both the Windows and Linux camps do well here. If nothing else, both camps have some very smart people in forums and newsgroups that can help.
* Third-party support.
Okay, here is where Windows has a lead. Let’s be honest, there are a lot of really cool commercial applications for Windows and not so many for Linux.
All that said, Linux really packs a punch when it comes to upfront costs. Most people use free versions of Linux (e.g., Debian
), the servers run powerful and free software (e.g., Apache, PHP
), and it just works.
So why isn’t Linux in more than “a fourth” of SMBs?
P.S. And of course, there is PHP on Windows
I've worked for dozens of small and medium size businesses as a software contractor. It used to be said that no one ever was fired for buying IBM. The same applies now with Microsoft. Simply put, the great majority of decision makers haven't the knowledge to make any critical decisions regarding anything to do with their computer hardware and software choices. --
Simple...see bullet point one.
You might just find a few random people walking the streets in spring of '08 that *say* they know linux. Ask them about slapd.conf or named and admire the blank stare.
For an SMB who might just barely be able to afford one I.T staffer, it is make or break to be able to replace that one person real quick when they leave. The smaller ones can't even support a regular part timer and just try finding competent Linux support people from your joe average consulting firm that targets SMBs.
Joel, I agree. One of the big problems is that SMBs are concerned, and rightly so, that if they put in non-Windows products, that 1: they won't be able to support it once installed and, eve more importantly, that 2: they will absolutely not know what to do when they lose the IT tech guy that installed it to begin with!
The "legions of MCSEs" is a powerful card for Microsoft.
So how can that be changed?
Why isn't Linux more than fourth? Comfort zone. It is the same philosophy that was prevalent with IBM in the 1970s and 80s - "No-one ever got fired for buying IBM"
Windows is seen as the safe option and hey - "EVERYBODY uses Windows don't they?" (Perhaps I ought to TM that!) That coupled with Microsoft's agressive marketing strategy gives the perception that it is the safe option.
Will that perception change? I don't know, although if you are at the top of the heap. there is only one way to go, again look at IBM who dominated the corporate server and software market, while Digital had the niche mini computer sector with the PDP series, and HP dominated the programmable calculator market. Where are they all now? IBM are still a strong force, but a very different business model, Digital, Compaq and HP have all merged.
Now although much of that was driven by the evolving computer technology, perhaps the tide will turn in favour of Linux and the open source model of software design at some point in the future. It is a question of critical mass, when Linux is seen as a mainstream desktop OS.