Can’t argue with this, I find it immensely frustrating. I’m on my fifth or sixth library for interacting with a database, my third component library, fourth or fifth config library, and the data science story especially has always been terrible because of the aversion to frameworks. This isn’t necessarily a great hardship, and for the most part old code just works, but so many libraries fall by the wayside or just never reach their potential cos they were the singular vision of one smart person, while all the other smart people were off working on other stuff.
It became a major headache for me when I saw multiple teams of developers trip over the same blind spots. Gluing together libraries works best when you're enough of a domain expert to understand every aspect.
When you're writing public-facing web services and forget about security while pulling together your stack, this ignorance can become very dangerous!
