2 ms·
As a software developer the single most important thing is speed of iteration. If you can test an idea 10x faster than the next guy then you can fail nine times
by austin-cheney 2y ago
As a software developer the single most important thing is speed of iteration. If you can test an idea 10x faster than the next guy then you can fail nine times out of ten and yet still remain equally productive.
This sounds like common sense but I almost never see this practiced and it’s not forgiving of human behavior. Iterating faster means doing these things that I almost never see software developers do:
* measure things
* abandoning reliance upon popular conventions as determined by measurements
* reduce build/compile times
* abandon reliance of dependencies for slimmer home grown solutions
* use test automation to identify regression, and keep test automation fast (less than 30 seconds total run time)
* prefer WebSockets over http
* exchanging code vanity checks for more functional assessments
—-
The benefits of iterating faster include:
* dramatically faster defect resolution
* increased learning and more time for investigating alternatives
* lower risk of regression, as determined by sanity checks and automation like lint rules and test automation
* a given software team becomes more productive with fewer developers
* developers become more focused on their work due to loss of interruptions/distractions from frequent periodic delays