Comment by ants_everywhere 10 months ago
The question isn't the "hey what's the first description that comes to mind?" The question is why people on HN are mocking the idea that saving HUGE_NUMBER*B is a bad investment just because there is some other A satisfying A > B.
The information we have is that an L7 and their manager believed the rewrite was valuable to the company
I was on that team. The rewrite was NOT valuable. Heck there is even a pinned Github issue at the top of the repo that tries to explain why this rewrite even happened and it doesn't have a good answer.
And high ranking managers never make silly judgments based on hype or cool-factor, is that right? Come on, the overlying context was specifically the claim that upper level management had lost touch with reality, and that this was just one example.
I get that people want to make fun of management. But the complaints were that Amazon was stagnating and frugal. The idea that they're spending extravagantly on programming language fads is amusing, but isn't really consistent with what others were saying or with what we know about Amazon generally.
Rewrites to save compute resources are done pretty frequently at huge companies. It's a relatively easy source of projects for high level engineers because they can see the fleet-wide metrics and have figured out how to choose projects that have clear dollar value in their performance write-ups.
Could an L7 have wasted a bunch of time on a fruitless project? Of course. Could an L7 have wasted a bunch of time on a fruitless project at a notoriously cutthroat and frugal company AND be super proud of it? That seems much less likely since they have to show some numeric impact at that level, but I obvious still possible. Would their manager then also be proud of it? That seems even less likely.
So when the only description of the project also accurately describes a large class of valuable projects (i.e. decrease fleet-wide cost of a ubiquitous tool) I was genuinely curious if there was more to the story.
Because B is quite possibly 0 in this case, given the limited information we have here.