Interesting article. It high lights some problems, in my eyes.
1. Numbering convensions...people place too much emphasis on what version number they have, then complain because it is not what they felt it should be. It's stupid. Why do people feel that they need 3.x.x? Do the 2.6.x thing. Then the 2.8.x. Then, when we have reached that point, we can hop up to 3.x.x. People want a standard method of naming the kernel version. They understand that x.oddnumber.x means that it is a development version. They understand that x.evennumber.x is a (generally) stable version. What the average pc user doesn't understand is when a software company or project jumps from, say, 2.2 to say, 2.6 without there being a 2.4 (or whatever, you get the idea). It's not intuitive enough. That can, strangely enough, turn people off. We don't want that.
2. Due to the increased focus on Linux in the media, we are seeing a lot of less than stellar press. People don't need to know that there is a debate amongst the programmers regarding something so silly as versioning schemes. It's rediculous. This debate, minor as it is, can and probably will have a negative effect. Joe Bag-O-Donuts wants to know that everything is peachy-keen. If Sun is having money problems, Joe won't buy a server since he will fear that Sun can't support him in the future. Same thing here. If there is negative press, then it sullies the reputation of Linux in general.
I hate seeing such silliness. It makes me mad. I guess I am just jealous about Linux. I want Linux to be the true golden child of the computing world. I want it to be the great messiah. I fear that such petty differences will turn others off to my little "religious" experience. Nuff said.