A key line seems to be buried in the "Summary" at the end of the post: "Python native enums are great for what they were designed to do, but..." Barring tricks I'm unaware of, enums in other popular languages, such as Go and TypeScript, don't support the author's more complex usage either—probably by design.
Out of curiosity, are there languages for which the enum implementation supports and encourages encoding complex data like this?
The paper may not offer insightful prescriptions for experienced engineers, but can work like this still be useful for informing future studies in a meaningful way? The authors repeatedly note the widespread inadequacies of the current research landscape. (To anyone familiar with the literature, is their assessment accurate?) In my eyes, the message is that the paper represents an incremental step in the direction of a truly detailed understanding of the factors involved in developer productivity. Even if there's a broad intersection between the answers one would get from taxi drivers and from computer scientists, the distance between the two fields makes that an unexpected result, which should prompt us to change how we think about computer science* and/or how we think about studying it.
One of the linked articles [1] describes the arguments in favor of pro-ISP restrictions as concerns that municipal broadband would either be a waste of taxpayer money or an "unfair" threat to private-sector ISPs: "'The general rhetoric behind these laws, from the incumbents, is that cities are too incompetent to run their own networks, so it's a risk to taxpayers,' Craig Settles, a broadband consultant who works with cities to create municipal networks told me. 'But then, the other side of it is that cities are so competent that they represent unfair competition.'"
Is this a relatively accurate and complete characterization?
I've been using Notion (https://www.notion.so/) for Kanban-style boards for a few months. After using Trello, Asana, and Jira, it's the only one that does basically what I want it to without spending hours on setup (Jira) or sheer annoyance (Asana). It won't be perfect for everyone, but for me Notion's task management is good enough and actually a pleasure to use so far.
I'm curious about the "fine-tuning based detection" mentioned in the report ("Fine-tunes a language model to 'detect itself'... over a range of available settings"). Does anyone know good articles/papers (or have an off-the-top tl;dr) to get a high-level grasp of "self-detection" for generative models?
Out of curiosity, are there languages for which the enum implementation supports and encourages encoding complex data like this?