4 ms·
There is no list. It's not possible to say, "I know these 5 things, and they'll make me more productive." Feynman solved all his physics problems with Calculus
by kemiller2002 5y ago
There is no list. It's not possible to say, "I know these 5 things, and they'll make me more productive." Feynman solved all his physics problems with Calculus. That's it. One tool. Part of is is knowing your tools. The other part is being able to see a problem, break it down, and understand how to manipulate it into something that's attainable. That's it.
I worked with this guy Pat several years ago, and he was on a different level than the rest of us. PhD Mathematics etc. I remember the day that I understood what made him different than me. We were standing in the hallway to our suite in the office building, and the first time visiting he says "This will be easy to remember. Lots of 3's." I looked at him for a moment puzzling his response. He says, "2745, lots of 3s." Pat was better than me at programming, because he saw things in a fundamentally different way than I did. He did things like that all the time. He wasn't better because he knew C++ or Java more than I did. He was better, because he would approach a problem and find a shortcut, or an optimization that the rest of us just didn't see. If something just seemed off even if he didn't know the technology, he could go figure it out, because his grasp of what to look for was much greater. That is what makes someone 10x better.
- warrenm 5y ago>being able to see a problem, break it down, and understand how to manipulate it into something that's attainable Wasn't this Alexander the Great's approach: divide and conquer? Every problem gets solvable if you break it into enough pieces
- copperx 5y ago> Every problem gets solvable if you break it into enough pieces Someone needs to just sit down and break P vs NP into enough pieces.
- warrenm 5y agoTraveling Salesman is "solved" basically this way with modern map apps: it may not be "optimal", but it's close enough And close enough, as the adage goes, is almost always "close enough"
- stnmtn 5y ago> This will be easy to remember. Lots of 3's." I looked at him for a moment puzzling his response. He says, "2745, lots of 3s." I cannot for the life of me figure out what he means by this
- aghillo 5y ago(3^3)3.3.5 ? Maybe…
- vqrs 5y agoThe number is divisible by three I guess.
- rockemsockem 5y agoBest guess I have is that 27 = 9 * 3 = 3 * 3 * 3 and 45 = 9 * 5 = 3 * 3 * 5 No clue if that's it though. Wonder if this "Pat" explained it to OP and OP is just keeping us hanging....