I was trying to point out that, for those folks, you can skip the middle step and just adopt a real workout plan created by fitness experts, and tested in real classes.

If your point is that there are a bunch of use cases for a fitness tracker/smartwatch that are not that, it's an interesting discussion to have, since my original comment made the assumption that the majority of customers focusing on those features were in fact using them to inform a workout plan/fitness plan (could be limited to diet, which Darebee also tries to address:

    * https://darebee.com/fitness/guide-to-meal-prep.html
    * https://darebee.com/fitness/portion-sizes-explained.html

I'm curious about other possibilities, like how we can incorporate AI to produce higher quality content, more quickly. Or more accurately. Bosses at work seem to assume that the way AI would benefit us to speed things up by some integer factor, and I kept explaining that tools like AI can provide benefits that don't appear in the "speed" dimension, per se.

One example is we can now vibe-out multiple potential ways for a feature to work, and actually play with them all over the course of an afternoon to help decide which one works best, and then actually code that particular approach for production.

I think the piece would be stronger if it explicitly explored ways to use AI in the authoring process that brought other benefits. As they note near the end of the article:

> It’s not going to be easy, because competitor publications are going to be able to produce immense quantities of writing in the time we are painstakingly correcting and adjusting a single article.

Right, it's fairly easy to generate a bunch of crap with AI. But what about having the AI critique your first drafts, or find flaws in the logic presented in an argument. I hope that writers work with AI in a sort of "rubber duck"/"sounding board" fashion, but it's hard to tell how they are actually working. The current front seems to be a hot debate about whether text was AI generated, but there's a lot to AI methodology beyond just that.

You don't have to know how much you weigh and you don't have to know your body fat percentage or even your heart rate to be effective at fitness. You don't need to buy equipment or have a special place to do it.

My point is the barrier to being fit is almost zero if you don't get distracted by all the shiny products they're selling.

This is intended to be an observation that applies to a general audience. I'm aware of elite athletes that require significantly more guidance to achieve the levels of performance they're looking for and I'm not trying to speak to any of that.

Darebee's a non-profit that's been around 15 years, and the founder of it intentionally changed its name to not be her identity because her mission is the fitness of real people.

My observation of people that are spending $600 on a special fitness watch and then a $20 month subscription to sync all of their health data and manage it with colorful graphs is they're not in better shape than the people that simply work out.

It's seems like the fitness equivalent of productivity porn.

For most folks, I feel like spending 20 minutes a day on Darebee would be more effective than their $20/month subscription to their $600 smartwatch.

https://darebee.com/

    * Downloaded the AppImage version for Linux (x64) from GH Releases
    * Downloaded the default Parakeet model
    * Tested with my BT headphones, worked perfectly, selected the correct device, no fiddling at all (Handy was much more difficult in this regard)
    * Actual dictation is pretty spectacular - live transcription as I speak Just Works, and it's returning 45-60 seconds of audio in <1s.  I think you've gotten the hardware detection right on Linux where Handy was struggling.

EDIT:

    * Text insertion works seamlessly, not sure what you did there. Other tools struggled with Emacs and on the command line -- Sayso just seems to work.

Copyright in compilation would be sufficient alone, requiring no new laws, to my reading. In particular:

https://en.wikipedia.org/wiki/Copyright_in_compilation

> In the copyright law in the United States, such copyright may exist when the materials in the compilation (or "collective work") are selected, coordinated, or arranged creatively such that a new work is produced. Copyright does not exist when content is compiled without creativity, such as in the production of a telephone directory. In the case of compilation copyright, the compiler does not receive copyright in the underlying material, but only in the selection, coordination, or arrangement of that material.

All my experience seeing orgs ship real production code with the assistance of AI over months/many-thousands-of-commits shows that the skill not only lies in what context you choose to provide in the first place (product and technical specifications, design guidelines, prompting process with the model(s)), but also in how to effectively review and test all of it, which is still done with human direction throughout. This is why people with no experience in coding can't do with AI what an industry veteran with 20 years shipping to production can.

In any case, I think existing reading of copyright law will be sufficient to say that large AI-driven software projects are a process of orchestration and also of curation in selecting what actually gets committed (or, if commit-by-default is enabled, then what to revert). This may not always be true in the future, though, as capabilities increase...one part of me thinks you could just do random stuff like "Make me an excel except every cell is a powerpoint presentation" and a good enough AI would reason through and produce something without further input. But I think in most cases, folks have a real set of requirements in mind, and get all of them into the AI through a process of iteration in various ways. I think that's real work that requires skill and, in many cases, creativity, at least in the sense that someone else walking up to the same tool and prompting on their own would produce a very different output than you'd see if you asked the same person to retry the same project.

Time will tell.

Given that photographs are copyrightable, I wouldn't hold my breath on projects like this being deemed ineligible for copyright.

My only addition to that point was that before I adopted a policy of not taking calls during certain times, I found the interruption annoying, especially since folks didn't have any context on my situation when they were calling. I solved this by first using "do not disturb" features, and then just stopped taking incoming calls entirely (wife and kids excepted). To my surprise, and why I thought it was worth writing about here, is that this worked better than I expected.

Turbo looks fantastic, though! It's precisely the niche I was looking to have filled, so I'll build a prototype or two in it to see what I think. Very much appreciate the pointer!

I already write Python services using Bottle that function both as a standalone server and as a CGI script, and I've been updating them to work with busybox's more constrained CGI environment (basically being able to handle the lack of .htaccess rewrite rules).

I definitely see a chance for simple deployment tools to make a resurgence as we create more one-off/disposable/personal software, and the low system requirements for something like busybox + lua stack are extremely minimal, which fits well with the current DDR4-is-good-enough zeitgeist. I'm running them on an old Raspberry Pi 2B and it's plenty performant for the ~10 people that access it regularly.

The point of my post was that I don't.