fallacies in article
This article has several fallacies.
1. The assumption of what I will call "conservation of effort": that if coders weren't working on their pet project, somehow the effort will be magically redirected to a more "mainstream" project, assuming this can be identified in the first place. It doesn't work that way. Instead of seeing all these efforts as failed mainstream efforts, they are really experiments in slightly different ways of doing things. Because they are FLOSS, the ideas can be absorbed elsewhere.
2. The assumption that Windoze is somehow a unified platform. In fact, what you get with Windoze is very little, it would only be the equivalent of half a Linux CD. Look, if you want something better than notepad you have to download something like vim or pay for a full WP. Look at the number of aftermarket addons for Windoze. On Linux you get gcc for free; you have to pay for a compiler on Windoze (unless you get gcc). Someone has remarked that Linux distros are not good at playing the government tender game. Sadly this is true. If they wised up and marketed a very small distro as the basic offering, then had various products called the compiler suite (gcc under another name), the web browsing CD (mozilla), you get the idea. There are signs that this is starting to happen, look at things like the SuSE Enterprise Server distro.
So what has to be done?
Firstly, user friendly distros should take the lead in emphasising the best of the breed offerings in each category. The bottom 95% by count in each category is only noise statistically but they have followings among power users, who will always be able to get hold of what they want anyway. You see this happening already in distros like Lindows, the choice is severely restricted.
Secondly, this is not a dictatorship, so it is not possible to impose that one offering is THE ONE. But there should be efforts to promote reuse, sharing and merging. Such people should have the ability to persuade besides good technical judgement.
Finally I would not like to live in a non-software world that resembles the software world that M$ represents. Imagine that you could only have one kind of coffee, one kind of pastry, etc. The real world is not like that. Let's look upon M$ as an anomaly, not a choices model to be followed. Long live diversity.
is the software mirrored anywhere else?
datagrid.com can't be resolved to an IP address so I can't reach the server.