Comment by pavel_lishin
Comment by pavel_lishin 3 days ago
But he does back it up with actual facts (as far as we can trust the author to tell the truth) - the feature that the author gave feedback on shipped without any issues. (The article actually doesn't say whether A was fixed-and-bug-free before B shipped, but it certainly sounds like B was less stressful to ship.)
It’s also a biased view. The author admit that the feature he was involved in took longer to ship initially. Depending on the environment this can be an anti-pattern; don’t we say “release early, release often”.
In the same vein; the author says that the other feature took several releases to be stable. Were the other release purely bug fixes or did that give the engineer a chance to get early feedback and incorporate that into the feature ?
It’s clear that the author prefers a slow and careful approach, and he judges “success” and “failure” by that metric. It sometimes is the right approach. It sometimes isn’t.