Triage ideas from 2022 dCFPd

I surfaced a few ideas at [Code & Supply's Diversity CFP Day](https://www.meetup.com/Pittsburgh-Code-Supply/events/284600316/):

* Patterns for migrating Python apps from requirements.txt + setuptools to poetry
* Writing old fashioned formal letters with Pandoc
* Introduction to PromQL
* Introduction to InfluxQL
* Upgrading Rails major versions (I'm probably not the best person to do this, though)
* Shipping a data pipeline inside of a Conda container
* "The Unbearable Weight of Massive ~Talent~ Tech Debt" - Strategies for Fixing Legacy Code Undercover
* We need to teach Python differently to enable the sustainable development of proofs-of-concept into maintainable software
* More `jq` always makes the problem better, right?

Already WIP:

* [Observability for Data Pipelines](https://github.com/colindean/talks/tree/master/o11y_for_data_pipelines)

Others:

* @tealjulia wondered, "Why is Python the default choice for the data [science and engineering] world?" I can speculate about its use in _teaching_ data science at the graduate level in part because of the `numpy`, `pandas`, etc. ecosystems that make DS easy and fast, but it'd be great to investigate this further with some first-party research and surveying.