HACKER Q&A
📣 launchiterate

Do you iterate or do you plan?


Do you iterate or do you plan?


  👤 nanis Accepted Answer ✓
You can't iterate without a plan. You need to know where to begin and where you are going and have a rough idea of how to get there.

You can't plan without iteration: Early decisions have to be made under much uncertainty and possibly with too many alternatives. Better first try the most reasonable sounding approach and course-correct based on experience.

Many teams run into a problem with that latter item: They start with the most reasonable sounding approach and stick with it at all costs because scrapping a few weeks of work and starting over is seen as a loss rather than a gain in institutional knowledge.


👤 PaulHoule
Both.

My current side project leans heavily on "iterate". I started out with the goal of "print something everyday so my inkjet printer doesn't dry out".

Along the way I discovered new problems (I can't give away images unless I document them) and new goals (interactive cards that can change the environment and transform how I relate to people.) I built a grid of 8x8 squares that covered about 30% of a wall and discovered it had unacceptable effects on radiative transfer in the room and had to be modified. I showed the installation to somebody and fumbled when I tried to explain it, so I made this react app

https://gen5.info/$/XQ*42RXF-TLY:$B.8/

This "project" (really a "program") runs into a crisis every so often which requires rethinking how I do it. I want to get the title "Master of Images" but I am a Journeyman now who headed towards the west coast for a road trip and just turned his car into a hydrofoil to get across the ocean.

The actual printing and image processing is about as disciplined as can be. I work from checklists and change the checklists frequently when I learn things. For instance I started out printing the front side first but if the front had a lot of ink the paper might curl and I'd have trouble refeeding to print the back side. When I was trying to explain the process to other people it hit me that if I printed the back side first I wouldn't have this problem -- I updated the checklist and now I am not afraid of making prints with a lot of ink on humid days, which helps with those radiative transfer problems.


👤 JohnFen
I always do both -- it's hard to see how a project of any complexity can be effectively tackled otherwise.