Getting lost in the code library

   

A little context for the non-dork reader: in programming, rather than constantly re-inventing the wheel, we routinely rely on extensive code “libraries” and “frameworks” created by lots of other people. But for smaller projects, code libraries can easily be overwhelming overkill, like trying to find the one switch you need in a cockpit with a thousand of them. On the other hand, lots of things are just completely out of reach without the help of a code library.

That dilemma is what this is about. It was inspired by an abject failure to get something done using a code library.

The stages a soul being crushed by trying to do build a relatively “simple” website feature using a code library...

  1. Wow, what amazing features! I could use this to send a probe to Mars with 3 lines of code!
  2. Okay, so how do I use it for this minor thing that would be annoying to build from scratch? RTFM. RTFM some more…
  3. After slightly more work that it’s supposed to be, but not so much that you are alarmed yet: Oooh, cool! Almost have what I need! Just a couple more tweaks...
  4. Well that was a cryptic error! *chuckles nervously*
  5. I just want it in blue, dammit. That’s all. Just blue! Also, I need to stop that “Eye of the Tiger” autoplay. There has got to be a switch for that...
  6. Just when you think you’re on the brink of victory comes the inevitable 20 to 180 minutes of “suddenly I can’t get it do anything at all.”
  7. You learn from a StackOverflow post that “blue” is not supported. 🎶 “It’s the eye of the tiger, it’s the thrill of the fight…”

But the real tragedy is that the DIY approach actually has phases that are identical in spirit, and the only real difference is the main source of frustration. With code libraries, it’s the paradox of too-much-but-never-enough. With writing features from scratch, it’s the inevitable discovery that there has never been a “simple” feature in the history of programming.