I used to work for a defense company that eventually got bought by a mega defense company. It was ~3000 people in various parts of the country and it was started by a man who deeply believed in the power of the entrepreneurial spirit. It had that free for all structure that PG says he'd never seen in a tech company. It was the wild west. We had different groups competing for the same government contracts. Managers and hackers alike got whopping bonuses for beating out other groups and they got to decide which contracts they wanted to bag. Entire groups were fired if they didn't bring in revenue. Fist fights, rancor and IP theft between teams were commonplace. But with all that they created some truly mind expanding tech for their time. They owned every angle of a highly lucrative market and showed no signs of slowing down... Until they got bought. The good news is, the fist fights, duplicate (and sometimes triplicate) efforts were stopped and everyone is one big happy family that hasn't done anything new in 7 years. Their market share is in free fall but they say it's fine b/c they are moving away from products in to large scale integration which is too boring to even type about. All the cowboys have gone to other places and I went to a startup. It was fun while it lasted.

"Mediocre hires hurt you twice: they get less done, but they also make you big, because you need more of them to solve a given problem."

I would say Mediocre hires hurt you THRICE. Twice for the reasons you mention, and the third one is all the good folks leaving since they can't stand mediocre people.

One big company that seems to do well, at least from the outside is Semco in Brazil. Watch Ricardo Semler's "Leading by Omission": http://mitworld.mit.edu/video/308/ and read his book "Maverick": http://www.amazon.com/dp/0446670553. This is of course an anomaly since most companies don't work this way.

> producing fresh vegetables doesn't [scale]

The reason starchy and hydrogenated foods dominate is agricultural subsidies, not so much natural economies. If you killed corn and soy subsidies a whole lot of junk food would disappear. Fritos are only cheaper than sauerkraut because you paid part of the frito bill when you filed taxes. Healthy fresh vegetables are most definitely mass produced.

Running with the metaphor any way, government has a major role in perpetuating huge corporations and preventing the emergence of smaller ones. SarbOx is an obvious example. OSHA makes it very hard for a small group of guys to set up a manufacturing shop, even if they know how to make it safe. Read this guy on how it's basically illegal for him to run a small, traditional farm: http://www.mindfully.org/Farm/2003/Everything-Is-Illegal1esp...

I've noticed this too. I think the decline has been going on longer than 30 years. It's because stuff isn't valuable anymore.

Was that meant to be funny? All but one of the points essentially say it's good to start a company in a recession because it's really really hard, and if you somehow manage to survive, you'll end up tougher, leaner, and meaner. Here's a tip, guys: start your company at a time and place where there are no constraints and even the biggest idiot can be successful.

Although I've run apps over Wine for awhile now...I still find myself thinking "ooh, ahh" when I launch a win32 app in Wine and it just works.

Cheers to the Wine dev team for all their hard work.

What people like about those places is the idea of living in them.

Unlike physical stuff, ideas can be arbitrarily expensive, so they comprise a lot of what gets sold to the very rich. I suspect practically everyone who owns a Picasso, for example, likes the idea of owning a Picasso more than they like the actual painting.

And, of course, this is old news, and, of course, it only works with MD5 (and --- we expect, but do not yet know --- SHA1, but not SHA256).

Your takeaway from this article is, "don't ever use MD5 again". It has no value. Use CRC's for error detection, and use a SHA hash for security.