I wholeheartedly agree. I actively use OSS software, and try as much to use it where appropriate. My web browser is Firefox, my text editor is open source, I've played around with linux installs for hobby and some server usage. But I also develop on Windows using Visual Studio 2005 as well.

My feeling is that where OSS really shines is when it offers a *better* product than the competition. It shouldn't matter whether the code is open or closed, as long as there is enough competition for the best product to come to the top. I think this is the problem with a lot of the OSS evangelists. Too often I think OSS/Linux is touted purely because it is "free" or NOT Company X.

Look at the successful open source projects. I truly think Firefox is a better browser than all other offerings, including IE, and it's success shows that. There are projects out there that actually offer a better product, and they are rewarded for it.

But on the other side, I see a lot of "it's like Program Y, except it doesn't do A, B, or C. But it's free and open source!". Sure, technically, if you want a feature you can program it yourself. But everyone is not a programmer, and the vast majority of the projects I've followed rarely follow user input; instead the developers make software for themselves.

Everyone complains about Microsoft software, just because they're the giant. But where is the *real* competition? Where's the OSS offering for a true Exchange/Outlook replacement? IMAP and Thunderbird or Evolution don't cut it. There's just nothing out there. Where's the easy to use yet powerful alternative to VS2005/8? It's not out there. Sure, they'll say this or that does what those do, and better...but who are we kidding? They don't. If they were really better, people would be using them.

Too much is blamed on marketing. Large IT people may have to bend to management, but there's a whole lot more IT people out there that actively look for the best solution, and have the ability to use it. The truth is, the best out there really is Microsoft in some cases. Deal with it?

Some months back, I needed software to record off the Net. I was going to be interviewed by phone on radio the next day.

One free package that I downloaded just did not work. There were no error messages and no results.

Another free package, on first run, stated that I needed something else -- a codec? I do not remember -- but did not give any details of exactly what or where to find it.

The third was a commercial product (Sound Tap by NCH Swift Sound). It had a crippled demo mode that inserted a message at intervals in recordings. I tried it and established that it could record from the Net. On the strength of that, I registered. I quickly got my registration, entered it, and tried it again. Now, it was fully working. The next morning, I was able to record my interview.

I did not have time to hassle with the vagaries of packages that made it difficult to do what I wanted to do. OSS has a long way to go in this regard with a few exceptions. I quite happily use Firefox.

In another case, I was looking for software to do PDF rendering of printer output. I had a similar hassle. Again, I went with a commercial product that worked.

Call'em like you sees 'em. There's no point in massaging egos or imaginations. How do they say, "WYSIWYG"?
1.) We're talking tools here, not philosophies of life. Use what works to get the job done, don't waste your time on chasing elegant or "free" solutions.
2.) Most of us don't do software as a living, it's only an incidental tool to keep the real work going, and the market obviously shows the path of easiest implementation, and the cost is a constant variable in the equation.
3.) One needs to accept the concept of lateral thinking to be a problem solver; beating one's head against a problem with the same non-working or less useful "solution" isn't efficient.
Whatever works with the least amount of wasted time and money is always the best solution. Sometimes it's Bill's world, sometimes not, but ultimately, we choose every day to get the job done. Or get in the unemployment line.

while your figures might support your position, the figures (and your position) don't tell a complete story.

I mean, if Linux is 1/10th the price of Windows, why compare revenue figures dollar-to-dollar?

the point of Linux is to lower the cost of the server OS, so it will inevitably always pull in less revenue than its proprietary counterparts, right?

Luke, I don't think that Linux is 1/10 the price of Windows. RHEL AS is $2499 for 1 year ( see https://www.redhat.com/apps/store/server/ )

You do make a valid point. But, if we can't look at revenue, what can we look at? Considering that Linux is cheaper than Windows, shouldn't we expect customers to increasingly shift to Linux from something more expensive (like Windows)? And at some point, the revenue from a larger customer base paying less by using Linux should outweigh the revenue from a smaller customer base paying more for Windows, right?

But if you can think of a better metric than revenue, I'd definitely consider it.

This is an interesting note:

The Linux OS market growth rate is slowing from ~22% in 2007 to ~16% in 2011.

This is a projection of IDC. But then further down I read that the Linux market will be .967B in 20011.

You further comment that the growth is Linux is based on a loss of market share by Unix systems.

But are we talking dollars or are we talking system installs?

If we are talking dollars, then I would suppose that the drop in system numbers for Unix would be greatly outweighed by the increased number of Linux systems. You don't need multi-seat license purchases to support Linux systems the way you do for Unix and Windows systems.

You say: "If the Windows OS market grew to 2011 and then remained flat (i.e. 0% growth), it would take Linux until 2029 to grow to the size of the Windows market. Lots of things can happen in the next 20+ years."

Again, I get the impression you are talking dollars, not installed systems.

If you do not use a consistent unit of measure when proposing your projections, then the conclusions are less than clear.

If all you are concerned about is revenue, then yes, it will take a very long time for Linux revenue to catch up to Windows revenue. But that is simply the reality of the reduced value of operating systems as they become a commodity. You don't have to keep re-writing the code to manage a disk file system. Once the code is written, unless you are doing something else besides managing a disk file system, you are not adding any value, regardless of the number of times you re-write it. The same is true of the thousands of other tasks performed by the operating system. The only real value Microsoft has added in the last few years is to close some vulnerabilities they had, and add some automatic anti-piracy code, WGA and DRM. The value of DRM is non-existent if providers of content decide not to use it. (Notice that you can now buy iTunes with no DRM.) The value of WGA is only of value to a system where you are afraid someone will make an illegal copy of your operating system. Something neither Apple, because if you own a Mac, you have a license for the installed OS, or Linux because of the distribution model used, are concerned about.

Not having looked at what IDC bases it's projections on, I have no way of knowing what impact the Vista mistake will have on those Windows projections. Presumably if new machines come out with proper video hardware, and sufficient memory, Vista will run in what people consider a reasonable fashion.

I do know that I recently cut a live CD of Ubuntu 7.10 for a friend, and I tried it on my little laptop to make sure it would work. I have to admit, even running off a live CD which is several times less responsive than a disk install, that I was impressed. I may have to give up my Ubuntu 6.06 and do an upgrade.

I once read a poster. It takes a lot of work to gain a customer. It only takes an instant to lose one. My suspicion is that the fallout from Vista will take some time to recover from.