Decisions you make now about application development will still be with you in ten years, so consider this: if you are still writing applications to a specific platform (VB comes to mind), then you are effectively locking your organization out of any potential savings that could be gained by migrating to Linux down the road.  I've written about this before:

"... it's not too late to keep your options open for the future.  You can maintain your future flexibility by making it a policy that all new applications be platform neutral and employ open standards whenever possible.

OK, I was completely wrong, and I want to set the record straight.

It turns out that the radio show "Open Source" really is what I suspected it wasn't - open source applied to radio! I'm going to find out more and report...

Yesterday I blogged about the erosion of the term "open source". My premise was that the meaning of open source is being confused in the industry by certain proprietary vendors, and in the general public by a nonsensical reference to the term by a new radio talk show.

Now this morning, while looking for a podcast in the iTunes store to help ease a long travel day tomorrow, I find this: a podcast called "open source sex" by sex educator Violet Blue (no, I was not searching for anything in particular). What could this possibly mean? It's the most ridiculous thing I've seen all week.

Not to beat a dead horse, but maybe there's life there still (as evidenced by a reader comment on a post from yesterday about the erosion of the meaning of the term "open source").
 
I have a pet peeve - the uses of, and the differences between the terms "free software" and "open source software". Keep in mind that in other languages (like French), the word for "free" (as in no cost) and "free" (as in freedom) are different. Not so in English, and this has caused more confusion than we really need.

Last week I blogged about Bob Waldie and his work on bringing the open source model to hardware. A reader left an interesting comment exploring the difference between a public communication protocol and open source.  Bob replied with an important point. He acknowledges that open standards are critical, but in the end don't provide the same base for innovation that you get with the open source model. As Bob puts is, "we often forget that the key to such open source projects being of value is not that the source information is openly available, nor that the code is licensed for public use. These projects actually succeeded, because they built a community of developers. They removed the dependency on individual developers. And they built a community of users, allowing the project to evolve to deliver robust useful open source tools.  So really notions like GPL only take on substance when a project succeeds in creating a "public"- and most open source projects fail in this regard."

I fear that the meaning of the phrase "open source" is already beginning to degrade.

First, Microsoft introduces Shared Source, a program that lets some customers view (not modify) some code. And they start calling site licensing "open license". Does it sound confusingly like "open source"? I think that's the point.

Now the first mass-market abuse of the phrase is coming from public talk radio. An article today in the NY Times (requires free registration) describes a new show on NPR called Open Source. What does this mean? Will the content be publicly available to take and modify to create innovative new shows? I don't think so.

The practical manager will seek to do more with less, not just cut costs. In some cases, open source software allows you to accomplish more simply because the commericial alternatives were too expensive. For example, sophisticated network monitoring has been typically been out of reach for mid-sized organizations, but there is no longer any need to go without - check out OpenNMS. High availability computing is another area that is now accessible to smaller budgets through open source. Heartbeat and DRBD can be used to set up redundant services and data.

In this interview with Martin Taylor of Microsoft (chief spokesman for the "Get the Facts" campaign), he implies that the Shared Source program is almost the same as open source because some customers can have a look at a portion of the code. Sorry, but this is NOT open source. The point of guarranteed source code availability is that users are not beholden to the vendor for bug fixes, new features, and other modifications.

Martin MC Brown had an interesting post today about applying the open source model to aspects of life other than software. He's having trouble with a replacement for his car, and made the point that only the manufacturer has the power to solve his problem. Such is the nature of proprietary systems, where a single entity controls the system/data/product/process. But if another entity was able to take over portions of the system, he might get a solution faster.

Why is this? Because the car manufacturer apparently has better things to do. Two months have passed, and still no solution has been forthcoming. It's clearly not a priority for them. But if the information necessary to fix the problem was available generally to the public, a smaller entity might find it worthwhile to step in and serve the niche market vexed by the problem.

I know that technology executives tend to be a practical group, so it's no surprise that Linux adoption has been a steady movement starting with lower-risk areas of the enterprise. You probably already have at least a few Linux servers running somewhere. So what's the next practical step?

Take an informal assessment of your server landscape, and determine those areas that may be a candidate for migration based on: ease of migration, risk of end user disruption, expected hardware retirement date, software licensing and support expiration, and staffing skill set. The goal is to choose a migration path that will provide the most benefit with the least risk and disruption. You will find that there is a continuum of migration potential, from relatively easy to relatively difficult, depending on your exact mix of IT landscape features. From this, you can begin to base a migration plan.

I was glad to see some reasoned thinking from Frank Hayes in Our Linux Values. We've all heard so much about about open source as a communal effort, but many people have missed one important point -- open source is creating a freer market for software, and speading our capitalist values in the process.

Open source software is currently in the process of breaking a series of virtual monopolies (I'm sure you can think of at least two or three), remaking the market in the process. With freer competition among software providers, prices will generally be driven lower. Some product categories, such as word processors, are on the path to achieving commodity status, with pricing to match. On a wide scale, the open source model is good for the technology marketplace because it spreads the cost of software development and maintenance.