That's a great point. I meant what's important to the company, which hopefully is the same thing as what's important to management. It's funny how thought pro

That's a great point. I meant what's important to the company, which hopefully is the same thing as what's important to management. It's funny how thought provoking your question is, though. Because a lot of developers, as you point out, care about far different things for their own reasons. One developer I knew, cared more about everything being 100% as "efficient" as possible before releasing it. She didn't care if things ever got released, really, that didn't matter as much as whether it fit her personal ideas of what's aesthetically ideal. To show you how extreme that was, she spent a week writing a caching layer (in python) using a hashtable + list for a least-recently-used algorithm. When I mentioned she could do this in 2 lines of code talking to memcached, she refused, pointing out how silly it is to talk to an external process with a context switch to look up some data. Who's right and who's wrong? Hard to say. I guess it depends on the context, and what the company is trying to do. I think the answers are different if you're trying to show good, sustainable results for customers without muddying up the code base, versus if you're writing software for lunar rovers.