The question is, if you can build something cool (and importantly get press), why just settle for just a job?
One of the projects that I spend my time working on was originally a way for me to teach myself about mysql, the next one was to learn about ajax, the next about CSS etc. This is all fine, but I'm doing this all on my own time, and I don't have the marketing budget to ever see if it can stand up to getting loads and loads of traffic.
Now lets say that I get hired by reddit, or twitter, or posterous, or couchsurfing, or whoever else. The types of technologies that these people use, while similar to what I'm doing with my personal projects, are on an entirely different scale. Getting to work on a database as large as reddit's would be absolutely invaluable to somebody that has an interest in large-scale databases. This is the type of thing that people typically pay thousands upon thousands upon thousands of dollars to go to school for. Getting the same (I would say better, actually) experience and having them pay you for it is awesome.
I guess my advice is: disregard money, acquire knowledge. If you do this then, hopefully, the money will come naturally.
Their database infrastructure is a major source of their pains and why they have to be creative with the caching side.
It doesn't help that the content is low latency and write heavy, more so than Facebook streams would be, I imagine since people aren't screaming into the void, but actually carrying on a conversation.
It's a damn good thing Reddit has a proper queue setup.
Seems to me that if you build a startup that's acquired by Google (say for the relatively small sum of $5m after a couple of years) that's still a whole lot more than you would have earned as a salary - even if you're a cofounder with just 50%. And on top of that you end up getting another good salary from Google as well once they acquire you.
In short, most people are making a salary, and of those that try to found their own company, most will end up failing and making less than they would have if they took a salaried gig. That is the risk of founding a company.
But every year you spend working on a failed startup, you might only earn $50,000 (let's assume you can draw a small salary off of investors or revenue). And let's also say there's a 10% chance of success. The EV for spending a year on startup(s) wound be $95,000, which is probably within wiggle room of what you could earn with a salary.
However, that's average. If you're clever and determined, your chance won't be 10%. At 25%, the EV goes up to $262,500 per year. That's a hard salary to get. But few people are sufficiently awesome to have a 25% shot, I suspect. (It's hard to say because there aren't many truly large sample sizes out there yet.)
Your average net income over the period of time it takes you to get a $5M exit will probably exceed your earnings from that exit, which will likely be less than $1M.
The point is that startups are not an easy or guaranteed way to make killer money. That's gold rush thinking. Think about how many startups (thousands) are in Silicon Valley right now and how many have exited in the last 5 years (dozens).
Don't get me wrong. Live the dream. I'm working at my 6th startup, and I just started my first one. But don't do it for the money. Do it because you're creating something you believe should exist in the world.
Because 99% of people make more money collecting a salary than they do being bought by Google.
He directly compared collecting a salary to being bought by Google, he did not compare earning a salary to starting your own company. The latter would be a fair statement, but if you are going to get bought by Google, chances are you will make more money than by earning a salary (unless you're an investment banker).
No more (and likely SIGNIFICANTLY less) than 375,000 people have been acquired by Google,
There are 154.9 million people in the American workforce (collecting salaries).
The actual number (based on the above information) leads us to believe that 99.998% of people make more money collecting a salary than they do being bought by Google.
It's worth noting, perhaps, that I might be counting the wealthiest 2% of Americans within the 154.9 million census number (or that there are dramatically fewer employees in the workforce since the bubble burst), but those wealthy folks were almost certainly not acquired by Google (although, if they were, it doesn't skew the numbers that significantly.)
Working for The Man also helps you do things you could never do on your own, because the company can fund things you could never hope to afford to tinker with.