I wound up stopping work on making TwistedSNMP work with PySNMP 4.x today. I had work on the VOIP project that needed doing, but that wasn't why I stopped.

Sometimes when you're faced with a task there are two (or more) mutually exclusive approaches, both of which will take a long time to implement. You have enough time to implement only one, and you won't know if you've chosen correctly until you have pretty much completed the task. This tends to produce a mental gridlock of sorts as you try to sort out which approach to take. That gridlock then eats up time as you try to find ways to estimate which way will work. If you can't come up with a reasonable way to make the estimate you just spin your wheels trying to guess.

At this point you basically take your best guess and move forward, knowing that even though you may have chosen incorrectly, spending more time trying to choose correctly would guarantee failure. However, when you have other projects sitting in the wings with their own critical paths that would otherwise eat into development time, you can switch to those projects and give your mind time to try to sort through the options.

So, VOIP work today, yawn. At least it didn't tax any brain cells. Wouldn't want to get them all worn out.