A few more tips from Oleh today:
1. Don’t suffer from problem admiration. Once you identify the problem, devise a solution to it and plow through the work.
2. Not all of your time is spent doing science. in fact, you may only do science 2x/week or so and the rest of the time is spent cranking through a plan and checking up on it.
3. Decide what you need to be convinced. Do you need to see 10 data points within 10% error? Or 3? Or 100%? Just decide, make that deal with yourself and stick to it, even if it is a lot of work.
4. Identify the lowest-level fundamental property to monitor your process. That sample doesn’t need to work for the final goal, but you have to be able to see what changed when you changed the process.
4. Try to turn innovation and technological challenges into execution challenges. This is different than proving a concept because, while you do ‘prove a concept’ you design and layout experiments that, when complete, will get you to the final goal only through incremental improvement. This thought requires some development.