Al Pacino’s screen test for The Godfather

Interesting story.

I had two calls with CEOs working on big deals. Company-changing deals. In both cases, they are going big or going home. Pushing the terms, pushing the pace and telling these much bigger companies that they are dealing with how it’s gonna be done!

Late in the afternoon, I met with two entrepreneurs who are building their next business while still working at their current one. They get up at 4:30am every day to code before work. And when their “job” is done, they switch hats and clock in for the nightshift on Newco. The four team members have promised to do this until they launch and/ or secure seed funding so they can quit their jobs and do this full time.

This reminds me of someone I know. Someone I know very well.

The best method I’ve found for hiring is to avoid fancy interview-type questions or put people under a lot of pressure, instead I just to get to know them and see if they are someone I would want to be in the trenches with. I check out their work to see if it is the level I am looking for and if it is we do a trial project (paid, of course) where I can actually see what it’s like to work with them over the course of a few weeks. There is no better barometer of what someone is actually like, how they communicate, collaborate, and manage their time.

That's how I will do it too.

Daily Lessons

When your first child arrives, there’s a perfectly valid argument that you should hunker down in your job and emphasize stability. But I started to see it the other way: I started imagining what it would be like to stay in my job for years while also contending with all of my frustrated ambitions. And I realized that I’d be coming home at the end of every workday still bearing those frustrations as they slowly chipped away at my sense of self-worth and my happiness.

Were I to do that, I realized what a terrible example I’d be setting for my child. Plenty of parents make heroic sacrifices for their children, staying in whatever imperfect jobs are available to them so their children can lead better lives. But to stay in a job simply for stability when I knew I had the skill and more importantly the opportunity to try something different seemed like cowardice. I just couldn’t square the idea of the uninhibited woman that I wanted Thuy to grow up to be with the daily lesson I’d be giving her in suppressing one’s dreams. And I just didn’t think I’d be able to hide any of those feelings from her, no matter how brave a face I could put on.

What’s more, I realized that I have a relatively short window of time in which I could try anything new. Our current situation is this: my girlfriend is gainfully employed, we have a decent stockpile of savings, no mortgage on our backs, relatively paltry debts of other kinds — and our health too, thank goodness. It’s expensive to raise a child any way you look at it, but it’s a lot cheaper to pay for the things Thuy needs today than it will be to pay for the things Thuy will need in say four years — and it will only get more and more expensive beyond that. In short, my tolerance for risk is on a downward trajectory. If I was ever going to try this, the time was at hand.

And that sold for $3.5 billion. Were you instantly rich?
Before the sale, we were completely broke. We had to move out of  New York City because we just couldn’t afford to live in the city. We had three kids, and I was barely scraping by to pay the mortgage.

So you were essentially broke?
Yes, completely. Right before the Geocities sale went through, my wife went to the cash machine to buy groceries for the week and there wasn’t any money in our bank account. I told her to put groceries on the credit card because I knew we were going to sell Geocities the following week. But before that happened, we were living hand to mouth.

That's pretty crazy.

For a game you've got to be better than realtime. Now, I do know a thing or two about writing games professionally, though not really high end stuff. If I've got a physics engine pegging the CPU to 100% then I can't do anything useful with the game. Even if I can dedicate one core to physics it's still going to cost me in terms of RAM and disk usage and software complexity, and even then I've got to make sure the physics completes fast enough that the rest of the game isn't waiting on it. Cores aren't completely free. I've yet to even see a physics engine for a game that can handle a large body of water (tens or hundreds of millions of particles) with lots of movement and reacts to external forces such as character movement that can run silently and quickly on a single core. Good physics is expensive, which is why you don't see it in games on a large scale such as for water.

Now, we also want to do things like run AI on a separate CPU core if possible, so it can be more complex than if run on the primary core. Do we want really awesome water and sand or better AI? AI wins out every time. Physics like in the demo video are also at the tail end of what people care about. It's more important to handle large objects properly, so if a jeep runs over a bomb it's going to react appropriately. No one cares if the dirt on the road is a pre-programmed animation or if it doesn't do anything at all because it's just a texture. Those small details will be great eventually because they'll add to the game world's realism, but they just aren't worth their cost on current CPUs

Context for the previous posts about the physics engine.

The Lagoa Multiphysics 1.0 engine has been developed by Thiago Costa, who currently works as Lead Technical Director at Ubisoft Digital Arts in Montreal. He’s also worked on a smoke simulator which can be seen in the video below:

About

Stashboard is a status dashboard for APIs and software services. It's similar to the Amazon AWS Status Page or the Google Apps Status Page. Stashboard was originally written by Twilio to provide status information on its Voice and SMS APIs. Stashboard is designed to provide a generic status dashboard for any hosted service or API. The code can be downloaded, customized, and run on any Google App Engine account.

It's really well put together. The site is so beautiful, you'd think it's a ruby project.