3 ms·

Reminds me of what I read from the memoir on Skunkwork. They specifically used off the shell components and other strategies to reduce engineering time and spee

by kiba 6y ago

Reminds me of what I read from the memoir on Skunkwork. They specifically used off the shell components and other strategies to reduce engineering time and speed up prototyping. Not knowing anything about aircraft development, they are probably much less ambitious about design specs while at the same time iterated much faster.

ISL 6y ago
There are many lessons to be found within Skunk Works by Rich. Among them is the statement that the SR-71 was not at the edge of the design envelope. If I recall correctly, there were substantial safety margins in most areas, "like a Chevy truck" in order to ensure the speed of development and improve the likelihood that the entire project would succeed. The HN productivity/self-help crowd will find a lot more in there. This passage has had the highest personal impact per word as any text I've read: You don't need Harvard to teach you that it's more important to listen than to talk. You can get straight As from all your Harvard profs, but you'll never make the grade unless you're decisive: even a timely wrong decision is better than no decision. The final thing you need to know is don't half-heartedly wound problems - kill them dead. That's all there is to it. Now you can run this goddamn place. Also: https://www.lockheedmartin.com/en-us/who-we-are/business-areas/aeronautics/skunkworks/kelly-14-rules.html https://www.lockheedmartin.com/en-us/who-we-are/business-are...
- ylee 6y ago
  >You can get straight As from all your Harvard profs, but you'll never make the grade unless you're decisive: even a timely wrong decision is better than no decision What I wrote in my journal after joining an investment bank's equity research team: "Good dinner with two traders, x and y; they have both been in the business for more than a decade and shared many interesting experiences and insights into the business. I did not realize that our traders very much need information flow, preferably proprietary, in order to entice their buyside counterparts to trade through them. It almost doesn't matter whether the information is accurate or not; what counts is being distinctive". I well remember one of the traders slamming his fist on the table complaining about how a former analyst was very bad at returning phone calls, and saying that being 52% right (meaning 48% wrong) is better than a higher correct percentage but not being available.
throwaway0a5e 6y ago
SpaceX also did the COTS thing to great effect.

by kiba 6y ago

ISL 6y ago

There are many lessons to be found within Skunk Works by Rich. Among them is the statement that the SR-71 was not at the edge of the design envelope. If I recall correctly, there were substantial safety margins in most areas, "like a Chevy truck" in order to ensure the speed of development and improve the likelihood that the entire project would succeed. The HN productivity/self-help crowd will find a lot more in there. This passage has had the highest personal impact per word as any text I've read: You don't need Harvard to teach you that it's more important to listen than to talk. You can get straight As from all your Harvard profs, but you'll never make the grade unless you're decisive: even a timely wrong decision is better than no decision. The final thing you need to know is don't half-heartedly wound problems - kill them dead. That's all there is to it. Now you can run this goddamn place. Also: https://www.lockheedmartin.com/en-us/who-we-are/business-areas/aeronautics/skunkworks/kelly-14-rules.html https://www.lockheedmartin.com/en-us/who-we-are/business-are...

ylee 6y ago
>You can get straight As from all your Harvard profs, but you'll never make the grade unless you're decisive: even a timely wrong decision is better than no decision What I wrote in my journal after joining an investment bank's equity research team: "Good dinner with two traders, x and y; they have both been in the business for more than a decade and shared many interesting experiences and insights into the business. I did not realize that our traders very much need information flow, preferably proprietary, in order to entice their buyside counterparts to trade through them. It almost doesn't matter whether the information is accurate or not; what counts is being distinctive". I well remember one of the traders slamming his fist on the table complaining about how a former analyst was very bad at returning phone calls, and saying that being 52% right (meaning 48% wrong) is better than a higher correct percentage but not being available.

throwaway0a5e 6y ago

SpaceX also did the COTS thing to great effect.