Software quality is a notoriously difficult thing to measure. Most studies are conducted over large codebases over several years (IBM & Microsoft have spent a

Software quality is a notoriously difficult thing to measure. Most studies are conducted over large codebases over several years (IBM & Microsoft have spent a lot of time on this). So, I can't actually prove it. I have spent a lot of time dealing with it and I know what I'm most comfortable with. And there are always exceptions. I've worked with some really talented remote people -- oddly, they usually end up relocating eventually anyway because they want to. I've also worked with some remote people that really think they're much better than they actually are. Professionally, the least satisfying jobs for me are the ones where people clock in and clock out. They may be very efficient, but I usually don't learn much from them and there's zero contact after we invariably go onto our next thing. That's a long-winded way of saying you might be right and objectively team cohesion is overrated -- I really don't know. Subjectively, I found I'm happier that way and I do my best work when satisfied. I do believe it's resulted in higher quality products usually, too. It's often the stuff you don't think about. Like overhearing a couple colleagues discussing something and being able to interject and save them a couple days of frustration. That stuff usually doesn't land in IRC, Basecamp, Skype, or whatever. But if you've managed to make it work, awesome. It's not my intention to combat the practice if it's working. I just don't like the attitude that anyone opposed to it is some sort of Luddite that sucks at business.