I hate writing and reading xml compared to json, I don’t really care if one is slightly leaner than the other. If your concern is the size or speed you should probably be rethinking how you serialize the data anyway (orotobuff/DB)

Anyone who says “a real artist wouldn’t …” Needs to spend some more time at the gallery.

This is a real thing, they are called operators and it is their job to oversee the cell, start and stop jobs, resolve bottlenecks, identify upstream problems and gracefully handle them, and emergency stop the system when needed.

Forever audits of free software are unsustainable in my opinion.

To truly audit every piece of software, you need an independent party to spend time (often more than the development) to look through the code, that person needs to be equally or more experienced than the developers of the software, and have specific knowledge for vulnerabilities and malicious techniques.

They then need to audit and monitor all of the channels of distribution for that software, including various websites and repositories. This needs to be done constantly.

You effectively need to double or more the total level of effort for all software.

Yes, high profile software (sometimes) gets audited regularly, but the assumption that anything you grab from your package manager has been truly audited leads to a false sense of security, additionally the assumption that an audit being performed means there are no issues with the code also leads to problems.

The reality is that most open source software doesn’t get audited because it is too much work.

My work gives me a $100 monthly stipend to cover internet and incidental costs for my home office. It’s not a whole lot but it’s a nice touch on top of my salary + cost savings of working from home.