Hacker News new | past | comments | ask | show | jobs | submit login
Deciding What Features To Implement (michaeldehaan.net)
12 points by shawnpage on March 24, 2011 | hide | past | favorite | 1 comment



Nice list. Such lists are quite personal and depend on one's values and philosophy. Here's my own triage list:

1. Is it reproducibly making the program crash or corrupting data? If yes, stop everything, fix that right now, today, and issue a hot fix before 24 hrs have passed, preferably before 2 hrs have passed.

2. Things that are obviously annoying the heck out of a lot (a lot as in most that I know of) of customers even though no one has said anything specifically about it, but it's become obvious to me because of the kinds of trouble they are having. This sort of fix will often be subtle workflow issues, or things that are nearly invisible but which take a huge amount of trouble to implement. Few of my competitors address this sort of thing at all.

3. Features I personally want, really need, and have been thinking about for a long time. This includes features that customers have asked for where I have been convinced it is necessary and won't screw up the program by breaking its underlying paradigm or making the interface too complicated.




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: