Everyone not in the Yahoo bay area offices is "remote" to them. Talking to engineers in New York is talking to "remote" engineers from the perspective of someone in the bay area. Does it matter that much if they're in an office in new york vs a home in nebraska? Maybe sometimes in some situations, but should companies all locate into one single building?

If you take this to its logical conclusion, every company should build a massive campus and locate everyone there, in the same time zone, even the same building.

I don't buy that Meyer is "giving up everything". She just had a nursery built next to her office (at her expense) so her infant can come to work with her. Is Yahoo going to give space for all the previously telecommuting young moms with infants to have a little nursery too?

Some companies may "move more effectively" if everyone is at the office, but I'm aware of several that have fully distributed teams, or significant numbers of employees who don't come into the office who are moving pretty fast- Basho and Automattic are two. (I think Couchbase does this too.)

I'd rather have someone living in one of the "flyover states" working for me from their home than the overhead of an office... offices are a cost and the assumed benefit may not exceed the increased cost. Certainly it can't be said that it always does.

The assumption that telecommuters are "less dedicated" is a prejudice. I've seen some evidence that they are more dedicated and put in more time. But further, it's assuming that you don't want to hire people who put in less time (for less pay, say a part time telecommuter.) Which seems silly. There are some people who you'd like to hire, but which you're going to be forced to pass on if you've got this "butts in the seats" policy, some of whome are quite good but due to family situations or simply optimizing their own lives, aren't interested in relocating to whatever big city you located your offices in.

Imagine DHH decided to go freelance. As the creator of rails it's safe to say he's the foremost expert, but this example works for anyone with significant expertise that would be valuable to lots of companies. Would DHH's value be maximized by working for one company 40 hours a week, and having all the work that doesn't require his particular expertise along with the really key stuff...or would it be better for him to work 10 hours a week for four different companies, maximizing his efforts on the areas where his expertise is completely unique? I know this is an even more radical idea, but I think it's the future of work.