Jon Udell has an interesting post about groups resisting change much the same way that antibodies take on an infection. He addresses some of the rather acerbic comments to my post about his Monad/MSH columns.
In the case of Monad/MSH, I am applauding two powerful ideas. First, the notion of a pipeline made of self-describing objects. Second, the long-term effort to expose system services as instances of such objects.
When I don my critical hat, I can see several ways for Microsoft to screw this up. Inventing a new scripting language seems problematic: I'd rather see IronPython, or other .NET implementations of dynamic languages, supply the glue. It's also not clear to me that the universe of system-services-exposed-as-objects is yet complete enough to sustain the magical illusion that Monad wants to create. Finally, there's the question of follow-through. It remains to be seen whether this idea has a champion who can push it across the finish line.
Still, Monad represents an interesting -- and potentially transformative -- idea. Resistance to the idea fascinates me. Of course, we see resistance to ideas on both sides of the fence.
Indeed, there does seem to be resistance to even the idea or suggestion that Bash should (or could) improve and add interesting new features to make Linux more attractive, or that Microsoft might actually have a good idea for once. That kind of thinking is counter-productive, and isn't any benefit to promoting Linux or open source. I'm reminded of the ever-present naysayers that plague every organization and champion existing policies and procedures because "that's the way it's always been done." That attitude is unhealthy in any profession, particularly IT.
GNU Bash is, in my opinion, the best shell available right now. (Sorry to fans of other nix shells...) But that doesn't mean it can't improve, or add new features that might make it more attractive to the crowd that finds Microsoft's products so wonderful. Why does that matter? Right now, the pro-Microsoft camp is the majority. If you prefer working with Linux, which I do, it's in your best interests to try to make Linux more appealing to the folks who are deciding whether the servers and desktops are going to run Windows or Linux.
There were also several comments that were more constructive, and have inspired me to look into Monad's promised feature-set a little more closely, and to see how Bash and existing tools can provide some of the features that look interesting in Monad. Stay tuned for that.