Couple of questions for the author as well as the HN crowd:
1. Should I interpret each X number of users as stated in the article to be "simultaneous" or "generally active over the past Y amount of time" or even "total user count ever?" "Unique per day?"
2. What do you think in general about skipping the step "100 Users: Split Out the Clients" if one is reasonably certain to not want or need multiple clients? It would seem as though this could keep the deployments and testing simplified until later in the growth stage, as more code can be deployed/tested as a single bundle. But also, I want to be sure I'm not missing something by just trying to justify my own interests.
1. Should I interpret each X number of users as stated in the article to be "simultaneous" or "generally active over the past Y amount of time" or even "total user count ever?" "Unique per day?"
2. What do you think in general about skipping the step "100 Users: Split Out the Clients" if one is reasonably certain to not want or need multiple clients? It would seem as though this could keep the deployments and testing simplified until later in the growth stage, as more code can be deployed/tested as a single bundle. But also, I want to be sure I'm not missing something by just trying to justify my own interests.