Re:Why the cost? (Score 5, Insightful)

As more and more vendors begin to port their applications to Linux, there's more interest from the shirt and tie folks in our companies that say, "Well, maybe we can give Linux a try. But, how do I know it will work?"

The way the vendors of these large products (let's use Oracle as an example) cover their asses in "product reliability" is to "certify" specific versions of Linux that they've tested and feel pretty confident about saying, "Yes, this will run stable and we believe we can recommend it."

So we end up with products like RHAS (and now RHES) and SLES 7/8 from the major Linux players. This helps make sure that -everyone- has covered their asses.

See, you and I know that we can take a copy of Oracle 9i and put it on a copy of RedHat 7.3 and expect it to perform rather reliably, right? We've worked with both the products enough to understand their quirks and how to support them. But if something does go wrong along the way your boss wants to hear something more than, "I've done this before and it worked! I have faith in the setup I've recommended." Your boss wants someone he can point a finger at and expect that they're going to provide a solution.

I've had a very hard time in the past few months bowing down to the concept of paying $799 for a copy of something I can pretty much download and patch-up myself. But the guys with the shirts and ties still think it's cheaper than a Sun/Veritas licensing solution, and it definately is cheaper than a Win32 alternative, so they're willing to shell out the cash. Why should I argue with them when after it's all paid for, it's still Linux. ;)

Bottom line? All of these certifications and extra costs for support we'll probably never use is a way to generate revenue while everyone is covering their ass and their product.

My $.02 anyway...