I think I figured out the biggest thing that slows me down / makes me ineffective as a developer. It is: not knowing what Im working on. Sounds trivial, but most importantly whatever task Im doing *right this second* should have a deliverable. And heres an example of why that doesnt come automatically: Say my task is to fix an easy bug - say, fix number formatting so 123456789 displays as 123,456,789. I open the file in my favorite IDE and there are 20 warnings. Several relate to type errors and indicate actual brittle code or possible bugs, several are not bugs but I need to dig deep into configuration to turn off. I decide I want to work on the warnings because its really good practice to not have warnings. Do you see what happened? I went from fix the bug to work on the warnings. And the problem with work on the warnings isnt the warnings actually - thats a totally reasonable first agenda item! its, work on. Here I am drilling into documentation without having any idea of when Im done. How to solve this? I took a half hour and very, very manually put the warnings in a spreadsheet! I had to hand-write their text and hand-edit the links (you know why? because doing this automatically was its own rabbit hole and that one I did *not* think was worth learning right now). I then showed this spreadsheet to a teammate who said 80% of them were unsolvable and 20% worth throwing into a group discussion, and I finished working on them. The first deliverable I created was *make a spreadsheet.* And in fact that *solved the entire problem*. The problem of course being however much of the warnings I could fix until I was unblocked. The days of coding in the zone for 14 hours on end are nearly elusive for the typical developer - so much of the difficult coding is solved an effective developer is someone who knows how to work with six engineering teams and three business units to move data around safely and scalably and mostly writes glue code with a hundred lines of business logic. Turns out thats worth paying the big bucks. For myself, Ive discovered the best way I can navigate this is to create micro-deliverables on the order of 30 to 60 minutes. /cc Randy Zev Michael Vaibhav P.S. should I put things like this on medium or something? A little markdown (bold topic sentences) would be nice and I should be building a portfolio a bit better. Ive played with my own blog (echl.in) but never found the write groove of publishing cogent content in an natural voice, but perhaps I might revisit. Really the only things that have been truly worth writing for me are minor eureka moments that I just feel like writing out somewhere (c.f. pontificating thesis-first reasoning-second ideas), and FB is within arms reach.
Posted on: Wed, 03 Dec 2014 03:19:20 +0000
Trending Topics
Recently Viewed Topics
© 2015