This is a delightful read!
My time-saving has always been split between automating the automatable and finding better and better tools to do so. Of course, I have fallen into the XKCD trap on occasion and spent more time automating a one-off task than it would have taken just to do the work. My excuse is always that I might find another use for the work in the future.
I can’t code, but it doesn’t mean that I never dabbled with Yahoo Pipes. I couldn’t understand dabbling with the APi’s, stringing them together and getting some output. Automating is fun, and I use IFTTT extensively to automate Twitter. I consider social media a cesspool. Why not automate it to keep a notional presence and reach out to specific accounts when it matters? As such, I limit my time on social media (and utilise more of it to blog or pursue other interests). For example, all grammatical checks are automated and happen continuously in the background.
Here’s what the author wrote and I’ll agree:
But IFTTT is limited by design. It is a single IF statement. And the biggest issue I’ve had over the years is the constant breakage of the connectors by the sites they connect to. This is no fault of IFTTT, just the ever-changing priorities and business models of lots of consumer websites and web apps. I’m starting to think we need an LTS for APIs or a commitment to not break them for at least five years from bigger organizations.
This is a useful addition to the discussion, and maintaining the API’s (while automating them, of course) is a rational choice in forward movement.