Nor is the problem limited to groups of 2 founders. You can have groups of 2 that work together well and groups of 3 that bicker. I'm guessing the author just had a limited sample size.
There's an endless variety of human experience and we only get to live one fragment of it ourselves. That's the point of swapping stories.
The post in question is a good example. It's much more valuable when combined with the comments on HN, many of them by people with first-hand experience of startups, than it would have been as an isolated article in the print media.
Merely enabling us to collect and collate information we already "knew" is surprisingly valuable. The clearest example may be open source software, but I think we will see similar gains in almost all fields.