Failing videos, focus on the website, holidays, and the tech death spiral - Patroncast S03E05

I absolutely agree on the first idea. People on the fediverse definitely know what they want and what they like more, and their needs / likes are probably more technical than most! Regarding the guest creators, that's actually a really, really fantastic idea! I might not do it for this holiday, because it's fast approaching, but for the next one, delegating the news video / podcast to someone else who might benefit from exposure to my audience would be a great idea, and a pretty fun thing to do: they could keep the exact same format, put on a white shirt and pretend to be me, and plug their own channel / stuff after the intro! Thanks for the idea, I love it!

The Linux Experiment

I am late to this, but I wanted to comment on a couple things in this 'cast. On asking Mastodon users for what|how of video topics: It seems to me that the people that follow you on Mastodon (including me) are not the people that would pick stuff that works on a platform like Youtube. You can place fediverse users into the camp of not liking, not being served by, or simply wanting to escape "the algorithm." (Which is not to say that everyone uses Mastodon or Twitter, but I suspect I'm not in the minority of people that would identify as nonTwits^1). Folks that willingly populate corporate run social media networks, like Twitter, as you alluded to, are probably more generic given the larger population, and more likely to be positively served by "the algorithm." Without a shred of scientific evidence to back me up, it seems to follow that you'd get better direction from a group of people who respond positively to systems that use engagement algorithms to promote content. Last, about holidays: Just throwing this out, but more than one content creator I follow does "guest creators" when they go on holiday. Maybe the biggest example I know of being Tom Scott. Could Niccolò Venerandi (Nicco Loves Linux), Matt Weber (The Linux Cast), fasterthanlime (sorry, I don't know his name) do a guest spot while you take some needed R-n'-R? (As one of the Fediversers that would probably give you bad advice for guests, I'd dig seeing someone like DJ Ware (The Cyber Gizmo) make an appearance with a how-to) ^1 edit: which is not to say people who use Twitter are "twits." Sheesh, I need a better typing filter. >_<

Absolutely agree on the stock market. The argument that is often advanced to justify it is "how could smaller companies get any funding without it, it would kill innovation", but that's assuming we need innovation in the first place, and also assuming that investors wouldn't invest if the company wasn't on any stock exchange, which is flawed in my opinion. If investors could still reap dividends on their investments without being able to sell their shares in a company, it would still work, and companies would still be able to get funding. As per AI, it's true, there is an inherent bias based on the dataset, which is comprised of what we know, or what we understand currently. As long as an AI isn't capable of imagining other variables that we don't know about (which, I don't know how that would even be possible without a fully sentient program...), they'll be limited.

The Linux Experiment

Wow, very diverse patreon cast this time. For what they're worth, my 2 cents on the subjects: Views: Personally I did watch the Fediverse video, but not the Mastodon video. I was curious to learn more about the Fediverse and see if there were aspects I didn't know about. I really like the concept and it has diverse applications. For Mastodon I didn't need the video. I did watch the launch-keyboard review, but just out of curiosity what you think about it. I knew before watching that it isn't for me. Business and shareholders are always fickle. I think the invention of the stock exchange was a mistake and I wish it didn't exist. All things considered, it does more harm than good. The same can be said about some other inventions that probably wouldn't have existed without the stock exchange. Because of it, companies have gotten too big. You can't launch a new product or reduce capacity without hiring or firing thousands of people. Again a result from the stock exchange which just gives access to capital and provides the incentive to grow, and grow, and grow. Customers and supply chains can be just as fickle as shareholders, however. Sometimes they give too low estimates, and accordingly you allocate resources. But then suddenly they want much, much more that you can't deliver. Or alternatively, suddenly demand drops for one reason or another and employees stand around doing nothing. In all these cases the people trying to provide the estimates did their due diligence, but there were just other (new) variables that weren't known. Which also brings me to the AI. I don't think an AI is unbiased and objective. Same as any mathematical model, it rests on prior (incomplete) knowledge and assumptions. Take the degradation of plastics for example. There are over thirty different variables that affect how quickly a plastic degrades. These are the **known** variables. There are undoubtedly more. Yet no model I've ever seen even includes the known variables, because there would be too many scenarios to consider and for many variables the dataset is too limited to be useful. It is furthermore assumed that some variables are not so relevant, but this assumption may turn out false or discounts interactions between variables. So unless an AI is so advanced that it can discover additional variables in the world beyond what it is fed and subjected to, gather a complete set of data on all those variables, and anticipates interactions, its result will still be as limited (or good) as any other mathematical model. In other words, I also would not want an AI to determine the future of France or accept anything it produces as the complete truth. Unless the AI is programmed by an omniscient being, but then, you might as well ask the omniscient being for advice directly. Sorry for the wall of text >_<