Hence, we shouldn't just go and post the top stories of the day; most people will have read about them before. The argument is that if we wanted to find links to news stories, we might as well just go to CNN or BBC. Of course, it's difficult to strike a balance. How do you know whether many people have read the news story you've linked to? And when does a simple NewsFilter post become truly breaking news? And how can you change a simple news post into something more, by using an interesting spin on the issue, perhaps by linking to other unique opinions?
It's easy to understand why people post news links to MetaFilter; for the most part, the news is interesting and it does stimulate discussion. It's also far easier to find an interesting news story than something else. However, the problem is that if it's easy for one person to find a news story, it's also easy for everyone else, and this goes against the purpose of MetaFilter (posting links to interesting pages that people have not seen before, or otherwise would not have found independently).
This topic is also continued in part in WhatIsAGoodPost.
It's perfectly fine to make a post on MetaFilter about major world events. A prime example is the Space Shuttle Columbia disaster - while the disaster was on the front pages of all web news sources and TV channels, the [thread on MetaFilter] served as a focal point for up-to-the-minute links. When the attention of millions is captured by a world event (another obvious example would be [the 9-11 thread]) it is almost impossible for MetaFilter members to talk about anything else.
MattHaughey will usually only keep the earliest or most informative post on a major world event if, as often happens, several posts are made within the space of a few minutes. This is simply to avoid duplication and keep all comments in one thread.