5 ms·

> you're able to figure out solutions on your own, and even more so identify problems on your own, and then work out a solution. How do I obtain these skills?

> you're able to figure out solutions on your own, and even more so identify problems on your own, and then work out a solution. How do I obtain these skills? As I told, I am stuck in this loop, tell me what to do, I will research and do it, but when it comes to architectural designs, I cannot select them, I know little about their trade-offs, can research more, but damn it, I cannot decide final solution to select (probably because of lack of enough experience)

itamarst 9y ago
This is a better articulation of where you feel stuck, so that's helpful. A lot of design decisions come down to pattern matching: you learn how to recognize certain patterns and that helps you decide (I recommend reading some of Gary Klein's books, e.g. "Power of Intuition"). So you need to expand patterns you can match. Some ideas: 1. Read case studies, e.g. http://www.aosabook.org/en/index.html http://www.aosabook.org/en/index.html. But ideally case studies tied to specific problems you're facing at work, rather than generically. A case study is specifically "here's what we built and why", which is more useful for pattern matching than "here's a tool you can do some stuff with". 2. Get your experienced colleagues to explain their thinking. Since they're mostly pattern matching, they can't necessarily offhand explain why they chose something, so maybe try to force them to think more explicitly. E.g. ask questions like "here's a hypothetical different approach, why wouldn't you go with that" and see if they can articulate that beyond "it's wrong" - how is it wrong? 3. Switch to job where you can learn more. 4. Build prototypes, see how they fail. 5. Get your colleagues to tell stories about failures and successes.
- godelmachine 9y ago
  The Architecture of Open Source Applications is a very good book series! If ever I get bored of reading the chapters, I immediately skip to "Lessons Learnt" section which usually comes at the end.

itamarst 9y ago

This is a better articulation of where you feel stuck, so that's helpful. A lot of design decisions come down to pattern matching: you learn how to recognize certain patterns and that helps you decide (I recommend reading some of Gary Klein's books, e.g. "Power of Intuition"). So you need to expand patterns you can match. Some ideas: 1. Read case studies, e.g. http://www.aosabook.org/en/index.html http://www.aosabook.org/en/index.html. But ideally case studies tied to specific problems you're facing at work, rather than generically. A case study is specifically "here's what we built and why", which is more useful for pattern matching than "here's a tool you can do some stuff with". 2. Get your experienced colleagues to explain their thinking. Since they're mostly pattern matching, they can't necessarily offhand explain why they chose something, so maybe try to force them to think more explicitly. E.g. ask questions like "here's a hypothetical different approach, why wouldn't you go with that" and see if they can articulate that beyond "it's wrong" - how is it wrong? 3. Switch to job where you can learn more. 4. Build prototypes, see how they fail. 5. Get your colleagues to tell stories about failures and successes.

godelmachine 9y ago
The Architecture of Open Source Applications is a very good book series! If ever I get bored of reading the chapters, I immediately skip to "Lessons Learnt" section which usually comes at the end.