Slashdot Mirror
"Get the Facts" Campaign Working
brontus3927 writes "According to a Reseller Advocate Magazine write-up, Microsoft seems to be winning its war against Linux. Info-Tech Research Group recently ran a survey that is now being used on Microsoft's Get The Facts campaign. In it were some surprising results. 'After polling 1,400 IT managers and CIOs in SMB corporations, his group found that 48% were not interested in Linux, 15% were not sure about Linux, and only 10% plan to evaluate Linux." Despite this, two-thirds of all webservers run Linux. The disparity in these numbers comes from the fact that most smaller companies' websites are hosted by service providers running Linux servers even if the company itself isn't."
Unfortunately Microsoft may be winning the war. And more scary in my opinion is Microsoft has shifted to more subtle means. What could be less intimidating than a web site dedicated to gently walking managers through the maze of technical issues ostensibly improving their (the managers) effectiveness?
For me, all I need to do to consider which platform costs less to manage is look back over the span of my career... I've managed Windows and Unix systems for over twenty years (which means I've managed Windows systems for "x" years -- you pick when you first think that might be -- I know it hasn't been twenty years). And when I weigh how much I invest to keep systems running, Unix (linux included) always wins, easily.
Of course, I found it unusual for management to ask me or any of my technical peers for recommendations, they typically get/got most of their advice gladhanding on the golf course, or from nice glossy brochures, and now, from slick benevolent web sites.
Microsoft is one of the best at PR, and their "Get the Facts" campaign may be one of their most impressive successes (oh that Microsoft would be so successful developing and creating safe and secure software). But, Microsoft knows perception is 90% of reality. What they say only has to feel true and assuage the fears of managers justifying manager's choices to stay with Microsoft. Unfortunately it has become a Nobody ever got fired for choosing Microsoft world (remember when it was IBM?), and with Microsoft's huge lead and head start in controlling the marketplace I don't see this changing any time soon.
What bugs me is when it bleeds into my area (I prefer doing my work in the Unix world...). For example, the time our team got a new member -- a new sysadmin who previously had been working and support Windows machines at our company. Our main server was a workhorse Sun Server and I had with reverence watched it chug away doing good work with an up time that had finally exceeded 550 days (not a huge record in the Unix world, but it was fun to see it go...). The Monday of week two of our new admin I was dismayed to see that our trusty Sun server now only had an uptime of less than two days. Sigh. Wasn't sure why, but reboots/crashes happen. Before I could do any more checking, "Bob" (not his real name) dropped by positively beaming and let me know he had noticed that luna (the server) had not been rebooted for a long time so over the weekend he had rebooted it for us! Universes collide! Sigh, again.
I'd love to see good technology prevail -- unfortunately today the combination of effective PR and FUD campaigns combined with Microsoft's products turns out to be good cough enough.
what are the percentages of IT managers and CIO who were not interested in Linux, were not sure about Linux, and planned to evaluate Linux before the Get the Facts campaign started?
Singularity: a belief in the "God" idea with the "demiurge" relation inverted.
That survey only shows what people are thinking of Linux now. Suppose that before the campaign, only 5% were thinking of using linux and 90% were dead-set against it? Then the campaign would be backfiring. Without any data points on what people thought before, or what a control group of people who haven't been exposed to the campaign think, we can't tell how effective or inneffective "Get the Facts" has been. -Drachasor
Those Apache servers are not all Linux. A lot of Apache servers are BSD or other *IX systems. A few are even Windows boxes. There's probably a MachTen box or two in the mix.
I think Linux is the cat's pajamas, the bee's knees; it does not need to steal credit from BSD and other projects in order to deserve praise.
Back when I was a youngster, IBM held dominance in the marketplace. Every CIO (they didn't call themselves that back then) that had a data center ran IBM. They seemed unbeatable. But then the PC came along and beat the pants off the old mainframe systems. This happened because users were demanding more capability than your average mainframe could deliver. It wasn't a matter of computing capacity, but rather the MIS department's ability to deliver applications in a timely manner. I worked on a project where we dumped a $50,000 app that we had written with a much more capable system on a PC built on Excel. Our customer (in this case another group in our organization) was very happy and we saved ourselves a lot of cash in the process. This new way of thinking wasn't driven by the CIO, but rather by the technologists who knew how to put this stuff together. It was collaborative and creative.
Fast forward to today. Corporations don't really drive the marketplace. Sure they have influence, but to think that by taking care of a very limited group of CIOs that somehow you're going to dominate the marketplace is a ridiculous idea. There are literally millions of small businesses that drive the economy and they don't consult their CIO when making a buying decision. They'll usually talk to another small business owner or their geeky nephew or some other "lowly" technologist. The CIO is nowhere to be seen.
Personally I don't know why Microsoft or any other company chases after large corporations like they do, other than that they're a large corporation themselves and know how to service that marketplace. Sure Microsoft has made billions on this market, but the question is whether or not it's sustainable. Once things become commodities (as software is fast becoming), large customers become very price conscious and beat you up for the last $. So unless they're a prestige account or you get some economy of scale, they're pretty much useless from a profit perspective. You're much better off servicing small to medium sized customers who either don't have the leverage or aren't as price sensitive.
If I'm looking to the real future of computing, I'd rather know what a bunch of geeks in high school think about technology than some random group of CIOs. They'll have the greatest degree of influence over it in the long haul.
If you don't want crime to pay, let the government run it.
48% are not interested in Linux.
52% are somewhat interested in Linux.
15% are not sure about Linux.
Which leaves 37% who have deployed Linux or are testing Linux for deployment.
The company I work for sounds similar to your's. We have LOTS of server apps that will only run on Windows. Except we have more servers.
Possibly. But "now" isn't "tomorrow".
The key issue is whether you're talking about an EXISTING installation or a NEW installation.
Because you have an existing installation, your company has already spent the money to evaluate and deploy that system.
But, at some time in the future, there will be an upgrade. And you will have to spend more money on your system.
There are 3 items to consider when evaluating a system.
#1. Total Cost of Ownership (TCO) - how much does it cost to run this system day after day.
#2. Return on Investment (ROI) - How much revenue with this system allow us to generate?
#3. Migration cost - How much will it cost to deploy this system.
Now, even though Linux may have a far lower TCO and a far higher ROI, the migration cost can be too high to make a business case for it.
But when it is time to pay for the next upgrade, the migration costs need to be compared. So if it would cost $1 million to migrate today, but it would take the company 10 years to make that money back, no one would migrate.
But then you have to pay $500,000 for the next upgrade. Suddenly, the Linux system doesn't look so bad. Particularly if you're looking at ANOTHER upgrade within the next 5 years.
So you (being the pro-active guy you are) get in touch with the people working on the Linux systems. You have the time and they have the incentive. Can they cut the migration costs to $250,000 within the next 5 years (estimated time to your next upgrade)?
After all, it's just 0's and 1's.
If they can do that, then the next upgrade will cost MORE than the migration.
It's called a "migration plan". Only idiots or people with an agenda try to migrate ALL of their systems at once.
Start by learning Linux and seeing where it can be deployed, reasonably, in your existing network. We're running it for DNS/DHCP/backup/webpages/etc. I also have it protecting an old GroupWise system. I'm also trying to establish OpenLDAP as our standard directory service.
The longer you wait to start, the more proprietary infra-structure you'll have to migrate.
Your IT department needs a plan. Otherwise, you'll be driven by the vendors. And the vendors are only interested in getting more of your money into their pockets.
And "staying with Microsoft" is not a plan.
Sure, NT 3.5, 4.0, 2000, XP, 2003 are all different, and so the skills needed to administer and use each is slightly different. Which means that your statement of `exact same skill set' is technically accurate, though extremely misleading.
In reality, people install NT 3.5, then upgraded to NT 4.0, and updated their skill sets somewhat to add any needed NT 4.0 knowledge. Then Windows 2000 comes out, they upgrade, and upgrade their skill sets. The incremental knowledge upgrades are relatively minor. And while somebody who knew everything there was to know about NT 3.5 would be somewhat lost with Windows 2003, he'd pick it up quickly enough. (And while I'm mostly a *nix guy, I know enough about Windows here to know what I'm talking about, even going back to NT 3.5 and even earlier.)
The same is true with Linux, or any other OS. If somebody who was familiar with Redhat 1.0 suddenly was confronted with Fedora Core 3, they'd be lost ... for a little while. Then they'd be OK as things started falling into place. (And remember, NT 3.5 came out slightly before Redhat 1.0 (both in 1994.))